282 BOTANICAL GAZETTE [APRIL 
Stamens 
The stamens bear four narrow sporangia, whose hypodermal cell 
walls show the usual thickening, but not prominently so, and whose 
cavities before dehiscence unite in pairs. The pollen mother cells of 
each sporangium are in a single row, surrounded completely by @ 
tapetum, and externally by two or sometimes three cell layers (jig. 6). 
They are somewhat flattened lengthwise of the anther, and in con- 
formity with this peculiarity of shape and arrangement they exhibit 
an interesting polarity, the spindles of both divisions standing at 
right angles to the long axis of the sporangium. Many of the stamens 
fail to mature their microspores. The ripe microspores contain 4 
large vegetative nucleus and two male cells. In good sections the 
latter can be seen to have an elongated pointed outline, and to be 
distinctly delimited from the remaining cytoplasm by a plasma mem- 
brane (fig. 7). The exine possesses strong thickening bands studded 
with minute spines (jig. 8). 
Ovule and embryo sac 
The position of the ovules and their relation to the carpels are 
illustrated in fig. 5. It will be seen that they are distinctly axial. 
They point outward at first toward the carpels, but gradually bend 
downward so that finally the micropyle is directed toward the base 
of the flower. The archesporial cell remains undivided for @ long 
time, lengthening with the enlarging ovule (figs. 9, 10). Finally, 
without cutting off any parietal cell it segments into four megaspor 
the innermost of which becomes the embryo sac (fig. I o). ina 
large number of cases several or all of the megaspores germinate, 
both by increase of size and by division of their nuclei. While thus 
increasing in size, some of them accommodate themselves to the space 
in which they lie by slipping past one another so as to lie side by 
side. Many cases somewhat similar to jig. 12 were observed; also 
in sections of the same age as fig. 13 a group of five or six nuclel 
above the embryo sac, and no doubt consisting of the unabsorbed 
nuclei of the sterile megaspores, can frequently be seen. But W . 
it is perhaps the rule for more than one megaspore to germinate, I 
have never found two fully formed embryo sacs in one ovule, sei 
other than the lowermost megaspore the successful competitor” 
