298 BOTANICAL GAZETTE [APRIL 
observed that some short-styled plants growing by themselves at a 
considerable distance from any long-styled plants produced mostly 
sterile capsules. From this he concludes that the short-styled 
form is very sterile with it own pollen. 
The pollen mother cells are small and a large number are con- 
tained in one loculus, often as many as 40 appearing in a single 
longitudinal section. Considerable variation is generally shown by 
the cells of a loculus, but there does not seem to be any very regular 
succession of stages. Frequently, to be sure, an anther shows a 
progressive series with the most advanced stages at the top; but 
this is by no means a uniform condition, for occasionally a loculus 
shows the mother cells near the middle in a more advanced condi- 
tion than the cells at either end. The relative position of the cells 
in an anther does not, then, in Houstonia, furnish reliable evidence 
of the succession of the stages. This makes the exact significance 
of some stages rather uncertain, and some of them are open to more 
than one interpretation. 
THE REDUCTION DIVISION OF THE POLLEN MOTHER CELLS 
PropHasE.—The pollen mother cells first become distinguish- 
able by their increased size and the possession of a very large nucleus. 
The nuclear reticulum appears as a network of very fine irregular 
threads, and contains numerous granules, none of which take the 
chromatin stain. There is usually only a single large nucleolus, 
and this is surrounded by the clear zone already described in the 
buckwheat. 
The nuclei of the pollen mother cells present at this stage an 
appearance which has been variously interpreted. Figs. 29-33 
show what appears like a progressive ‘‘ budding off” of chromatin 
staining material from the nucleolus. A similar condition has been 
observed by DARLING (5, p. 184) in Acer Negundo, and described 
as a budding off of actual chromatin which goes to make up the 
spirem thread. Miss NicHoLs (30, p. 35) has observed this con- 
dition in Sarracenia, and considers that it represents a movement 
of chromatin, which has been elaborated in the nucleolus, to the 
nuclear reticulum. GaTEs (15, p. 6) interprets a similar appear- 
ance in Oenothera in an entirely different manner. He regards the 
