422 BOTANICAL GAZETTE [JUNE 
one another, plates of hypodermal archesporial cells ( figs. 30, 
31,32). All of these divide once to form axial rows ( fig. 32). 
These rows are radially placed, forming a mass of cells shaped 
like a truncated cone with its apex inward. Of the cells form- 
ing the inner layer, only a limited number, and, as a general 
thing, but one becomes the ancestor of the spore mother cells 
(fig. 33). The outer layer divides at once into two layers, 
thrusting the primary potential sporogenous cells down where 
there is even less room to divide (fig. 33). Among these one 
cell soon may be distinguished by superior size and different 
stain as the forerunner of the spore mother cells referred to 
above. The hypodermal layer of cells divides by periclinal 
walls into two layers (fig. 34). The tapetum is developed from 
the wall, and is frequently differentiated some time before the 
mother-cell stage (fig. 34). The development of the spore 
mother cells (fig. 36) and a clearly defined synapsis stage (fig. 
37), with the usual division into tetrads, follow (jigs. 35-40), 
In one preparation the mother celis had divided into fives and 
sixes instead of tetrads (figs. 2-46.) Nothing was observed to 
suggest how they were formed. It may be noted from the 
figures, however, that unlike the cases of Hemerocallis described 
by Juel,3 the spores are of the same size. 
The ordinary sequence of events leads to the development 
of the pollen grain, which is a spherical body with three deep 
grooves running from pole to pole. A large vegetative and a 
smaller lenticular generative nucleus were seen in the pollen 
grain and in the tube, but no trace of a division of the generative 
cell preceding tube formation (figs. 48-51). The pollen tube, 
with one generative cell and the vegetative nucleus, were fre- 
quently seen in cross sections of the peculiar nucellar neck, 
which led to the inference that either the generative cell may 
not divide at all, or if it does, one of the male cells disappears 
before the pollen tube enters the micropyle. The actual fusion 
of the gametes was not observed. 
3Cytologische Studien aus dem Bonner Botanischen Institut. 51-72. 1897. 
