Cook: EmBryoLocy oF RHIZOPHORA 273 
three ovules are in an advanced stage of degeneracy (FIGURE a, 
The cells of the endosperm appear to radiate rather indefinitely 
from a central point in the sac (FIGURE 6). The growth of the 
embryo may be divided into three periods: (1) first growth of 
the cotyledons, (2) growth of the hypocotyl, (3) second growth 
of the cotyledons. The youngest embryo secured was pear- 
Shaped with well-developed suspensor, but was not attached to 
the wall of the sac (FIGURES 6, 8). The suspensor persists until 
late in the development of the embryo (FicuREs 8, 9, 14). Ap- 
parently the embryo had originated as a series of cells in linear 
arrangement and developed in accordance with the capsella type. 
The only embryo of this species figured by Karsten is somewhat 
younger than this one but does not show a suspensor. The coty- 
ledonary end of the embryo grows quite rapidly and is sometimes 
* quite regular in outline (FIGURES 9, II, 12) but more frequently is 
very irregular (FIGURE 10) and apparently very much crowded by 
the Surrounding structures. At a point in the longitudinal axis 
of the embryo and not far from the suspensor, the cells are very 
small and very numerous (FIGURE 9, +). Ina short time the sep- 
aration of certain cells at this point gives rise to the plumule (FiG- 
URES 10, 12, 24); this is only discernible in section. The cells 
on the side next to the root-tip are smaller, contain more proto- 
plasm, and are apparently more active than those on the cotyle- 
donary side (FIGURE 24). The endosperm is very thin but the 
cells of the integument which come in contact with it are very 
Protoplasmic (FIGURE 25). This union between the endosperm 
cells and the integument is very close; frequently through shrink- 
age of the internal structures there would be a separation of cells, - 
but this was usually within the integument and not between the 
Cells of the endosperm and the wall of the sac (FIGURE 2 5). The 
growth of the cotyledonary part of the embryo results in the 
Pushing of the opposite end of the embryo and part of the endo- 
- Sperm out through the micropyle, and they now lie within the 
cavity of the ovary (FIGURE 12). The endosperm over the sur- 
face of the cotyledons is now very thin, thus bringing them into 
close contact with the walls of the sac (integument) (FIGURE 12). 
The cells of that part of the endosperm which has been forced 
from the sac show a rather indefinite arrangement in rows extend- 
