( 360 ) 
(fg. 175). Preparations showing early cleavages were not 
obtained, but figure rz5 appears to indicate that they may 
have been transverse. In this case at least the subsequent 
divisions are in various directions. The same embryo in its 
relation to the endosperm may be seen in figure 776. 
Citrullus 
Several longitudinal series of elevations occur on the pla- 
centae and these at an early stage appear merely as undula- 
tions on the folds as seen in transverse section. When these 
elevations have reached a height equal to their diameter at 
the base or slightly more, the first suggestion of an arche- 
sporium is apparent. Up to this time all the cells of the pro- 
tuberance are of about the same character, staining deeply. 
The archesporium is of very indefinite character. No en- 
larged cell adjacent to the epidermis (such as occurred in 
frevillea) was observed. ‘The sporogenous cell (fg. 177) ap- 
pears in an axial row and is recognizable at first only by its 
size and the open character of its nucleus. When first dis- 
tinguishable it is covered by two tabular tapetal cells, some- 
times very thin and again of an appearance quite similar to 
the sporogenous cell itself. There is always one axial row ap- 
parent, sometimes one or more parietal rows identical in ap- 
pearance with the axial one. While in its earliest stages the 
sporogenous cell may not be easily distinguished from its 
neighbors, the same does not hold true in its later stages. 
After a short period has elapsed the spore-mother-cell becomes 
quite distinct both in size and structural features and there is 
never more than one cell apparent in that condition in the 
same ovule. 
In the same ovary the early phases of the archesporium may 
be found about alike in all the ovules, but the rate of subse- 
quent development is dependent to some extent upon the posi- 
tion upon the placenta (see text-figure 5, A7), those arising on 
each placenta nearest its median line having the advantage. 
The differentiation of the parts of the ovule does not begin 
until after the sporogenous cell appears. By the time that 
