352 BOTANICAL GAZETTE [NOVEMBER 
The microspores of Potamogeton become separate immedi- 
ately after the second division of the mother-cell.. The anthers 
are at this time still quite small, the subsequent growth being in 
reality for the purpose of accommodating the increase in size of 
the pollen grains. The microspores at first have a thin, although 
distinct and homogeneous cell wall surrounding the cytoplasm, 
and a very large nucleus. The latter fills at least one fourth of 
the whole cavity of the cell (fig. go). The limited amount of 
cytoplasm present at this time is decidedly much more homo- 
geneous than in the mature pollen, and stains with the gentian- 
violet a uniform pale violet similar to that of the cell wall. For 
a short time after the wall of the mother-cell disintegrates the 
pollen grains are still held together by the remains of these 
walls. In fact they are as if imbedded in a ground mass of 
some viscid matter. 
The young grains very soon begin to increase in size, but 
the cytoplasm does not keep pace. As a result, the latter at 
length is confined to the parietal layer, but with a considerable 
increase in thickness on the side where the nucleus is located. 
These stages occur when the embryo-sac is yet one-celled, and 
of course while the spike of flowers is still enclosed within the 
bud. 
Just before the nucleus begins to prepare for div 
the following conditions: The cytoplasm is decidedly “cay 
granular, and stains more deeply with the orange. The large 
vesicular nucleus possesses a very distinct membrane. ce ie 
close against this is the linin thread which is rather extenst" 
for Potamogeton. The thread however is nearly destitute ° 
chromatin. The nucleolar-like body is smaller than usual, af 
in some cases more than one may occur, tare 
The first division of the microsporial nucleus takes P : 
much earlier than in Convallaria, and while the whole je 
yet enclosed in the bud. The spores reach their full size 
the division, and it is at this time that the exine first ae 
show signs of the thickening which produces the very gl 
roughened surface of the mature spore. Owing to me 
ision we find 
