1 34 Gager . — The Development of the Pollinium and 
radiations at the poles of the nuclear spindle. These radia- 
tions have been described for Asclepias by Raciborski ( 1897 ), 
and by Strasburger 1 as centrosomes. Raciborski says that 
if the sections are very thin and the lens strong enough, the 
supposed spherical centrosomes appear only as centres of radi- 
ating plasma-threads. My own slides agree with Raciborski’s 
description. 
The Second Division of the Primary Pollinium-Cell. 
The daughter-nuclei resulting from the first division of the 
primary pollinium-cell enter upon the resting stage (Fig. 19), 
with the chromatin distributed upon a linin reticulum. 
Gradually they take up a position at the centre of their re- 
spective daughter-cells, and in the meantime the spindle- 
fibres, which at first took a violet stain from the gentian-violet, 
lose this colour, and take on the yellowish brown colour of 
the cytoplasm. They gradually shorten and finally disappear 
(Fig. 20). 
The nucleoli of the daughter-nuclei are very much smaller, 
relative to the size of the nucleus, than in the nucleus of the 
primary pollinium-cell. Stevens suggests from this fact that 
the substance of the nucleolus has gone to form the new 
structures in the cell. The writer found no direct evidence 
on this point. 
The second division (Figs. 21, 22) is quite similar to the 
first. Synapsis was not observed and probably does not 
occur here. The divisions result in four cells, the structural 
units of the pollinium, arranged normally in a radial row. 
No exception to this was observed. The cytoplasm of the 
young pollinium-cell is much more fibrous in appearance than 
that of the mother-cell. 
In no case were less than four pollinium-cells developed from 
each primary pollinium-cell, as Strasburger ( 1899 ) describes. 
In every instance each primary pollinium-cell divided and 
gave rise to four pollinium-cells arranged in a radial row 
( Fi g- 2 3 )- 
1900, p« 1 66, 
