STUDIES OF THE DEVELOPMENT OF PEPEROMIA HISPIDULA 375 
be seen in the oospore, as they were in P. pellucida. The two nucleoli 
of the oospore remain distinct and separate until the first division of 
the endosperm nucleus has been completed, at about which time they 
apparently fuse to one. The nucleus and nucleolus of the oospore 
both remain single from this time onward, usually until 10 or 12 endos- 
perm cells have been formed (figs. 84, 104). 
The oospore continues to grow and, at some time after the number 
of endosperm cells has reached 8 or 10, it begins to divide by successive 
walls whose sequence has not been made out. There is thus formed a 
small embryo which in the ripe seed shows but 8 or 10 cells in a longi- 
tudinal section (fig. 108). In the seed referred to above, which had 
passed two weeks in damp moss after being gathered, the form of the 
embryo is some what more elongated and the cells are more numerous. 
This probably represents the more usual form of the embryo, for this 
figure was drawn from sections of formalin-fixed material, which gave 
no evidence of shrinkage such as was seen in nearly all older embryos 
that had been through alcohol (fig. no). The shaded cell at the right 
of the embryo in the latter figure is probably the persistent synergid. 
The condition of embryo and synergid in the ripe seed is thus seen to 
be similar to that found in the seed of Peperomia pellucida (See John- 
son, 1900, figs. 12, 13, 14). 
F. Germination* OF the Seed 
After searching repeatedly for germinated seeds in the field, and 
attempting to germinate collected and partially dried seeds in the 
laboratory, success was at last attained with seeds that were placed in 
damp moss at once on gathering, and mailed from Jamaica to Balti- 
more. These reached Baltimore on February 11, 1914, and were at 
once placed between layers of damp filter paper in a closely covered 
glass jar. On March 25 the radicle of the embryo was seen projecting 
from several of the fruits (cf. Johnson, 1902, fig. 36). In the three 
weeks following some of the seeds had developed a hypocotyl twice as 
long as the seed and a primary root three times as long (fig. 112). 
Examples of all stages were fixed and preserved for surface study and 
sectioning. 
The first external indication that germination is going on is the 
pushing out of the wall of the fruit at the upper end. Soon the white, 
pearly layer of the endosperm becomes visible between the 4 or 5 
parted flaps of the carpellary tissue (fig. 112). At first the endosperm 
