40 BOTANICAL GAZETTE [JULY 
persistence in older sacs, this mass is surely the egg and the two 
other masses the synergids. Fig. 17) is another view of the 
same sac and shows the synergids more distinctly. Fig. 16 
undoubtedly shows the two synergids, but no division of the sur- 
rounding protoplasm. In fig, 18 the egg appears with the spheri- 
cal bodies regularly arranged; the small nucleus is a synergid. 
Fig. 19 is the micropylar end of a sac in which figures of dividing 
endosperm nuclei were seen. The large cell is the egg, near which 
are the spindle-shaped figures mentioned before. From their 
position it is possible that they are the remains of the disintegrat- 
ing synergids. The large cell of fig. 24 is the egg at a later stage 
than the other figures show, because it is almost completely envel- 
oped by the endosperm; numerous illustrations of this stage were 
found. No more distinct segmentation of the egg was seen than 
appears in fig. 25, in which the protoplasm is divided, but only 
one nucleus could be found, which makes it a doubtful case. Even 
after the egg is completely surrounded by endosperm, the deeply 
staining globules remain, but they and the whole egg seem to lose 
the prominence shown in fig. 24. However, this is partly relative 
owing to the increased size of the whole ovule. Figs. 20a and 206 
show the curious chromosome-like bodies in the micropylar end 
of the sac; they are rodlike and twisted, resembling chromosomes, 
but not those of Diospyros, which are very small. The investiga- 
tion of these phases of the life history of the persimmon will be 
continued in the hope that the complete sequence of events in the 
embryo formation will be found. 
The youngest unmistakable embryo that has been seen consists 
of three cells (fig. 26). This embryo was in the extreme micro- 
pylar end of the sac, imbedded in endosperm. Its position agreed 
very nearly with that of the egg when surrounded by endosperm, 
but no proof could be found that it originated from that cell. 
Fig. 27 represents the embryo at a much later stage, but does not 
yet show differentiation into stem tip and cotyledons. Fig. 28 
is a variation from the common type, and fig. 29 is the appear- 
ance of the embryo about the time it can be distinguished without 
a lens. One case of polyembryony was found (fig. 30), and one 
lot of material contained freak embryos, one of which is shown in 
