20906 BOTANICAL GAZETTE [APRIL 
could be demonstrated. No difference is apparent either in the rest- 
ing nuclei of the pollen mother cells or in the pollen mother cells 
themselves (figs. 1 and 2). The difference becomes evident, however, 
after the rounding off of the pollen mother cells during synapsis, 
and apparently reaches its maximum some time before diakinesis, 
(compare figs. 7 and 8; and figs. 11 and 12). It is of course 
much easier to measure accurately spherical cells than angular cells 
massed together, as it is extremely difficult to determine in the 
latter case whether the cells are cut in the same plane or not, and 
there may be a difference in the size of the somatic cells which could 
not be determined. It seems entirely probable, however, that the 
difference in the size of the pollen of the two forms is due to the 
fact that the pollen mother cells of the short-styled form grow 
more rapidly during the period from the beginning of synapsis to 
diakinesis, than do the microspore mother cells of the long-styled 
form. It is interesting to note in this connection that it is during 
this same period that the ovocyte in animals undergoes its great- 
est enlargement (GREGOIRE, 17, p. 243). 
SEPARATION OF THE POLLEN GRAINS 
The separation of the pollen grains occurs in much the same 
way as was described by Miss Fercuson for Pinus (14, p. 35): 
During the late telophase of the second mitosis in the microspore 
mother cell the four nuclei of the tetrad become connected with 
one another in all directions by ‘‘kinoplasmic’’ fibers (fig. 19). 
These fibers, however, are never very numerous, and they are visible 
for only a comparatively short time after the reconstruction of the 
daughter nuclei is complete. During this period a marked thick- 
ening of the wall of the pollen mother cell occurs (fig. 20); and, 
apparently continuous with this wall and extending out from it, 
walls appear separating the daughter cells (fig. 21). 
These walls attain a remarkable thickness, and are apparently 
homogeneous and extremely resistant to stains. They stain very 
lightly with haematoxylin or safranin, and with orange G only if 
the staining be considerably prolonged. The strength and definite- 
ness of the walls thus formed is shown by the fact that after the 
spores are mature and the wall of the mother cell is ruptured, the 
