140 Gager . — The Development of the Pollinium and 
is to be inferred from Stevens’s statement ? The answer to 
this question is found in the history of their development. In 
the case of the development of true pollen-grains, and of the 
spores of Bryophytes and Pteridophytes, numerical reduction 
of the chromosomes takes place in the mother-cell. In 
Asclepias this reduction occurs in the primary pollinium- 
cell, the two successive divisions giving rise to four cells, each 
of which eventually develops two sperm-cells and a larger cell 
which germinates into a tube through which the sperm-cells 
pass to the ovule. The course of development, so far as the 
number of nuclear divisions, the place of the numerical 
reduction of the chromosomes, and the mode of formation 
of the sperm-nuclei are concerned, agrees precisely with what 
takes place in other plants that form pollen-grains. We 
must conclude, therefore, that the primary pollinium-cells 
are pollen-mother-cells, and that the pollinium-cells are true 
pollen-grains. 
The chief peculiarity in the division of the primary pollinium- 
cell is in the arrangement of the spindles of the successive 
mitoses. Two different orientations have hitherto been 
described in Angiosperms. First, spindles at right angles 
to each other ; second, spindles parallel but not in the same 
line 1 . In Asclepias we find a third variation, where the axes 
of the spindles are in the same line, resulting in the disposition 
of the four daughter-cells in a radial row. The only other 
instances of a similar arrangement are exceptions in the plants 
where they occur 2 . This arrangement, however, would not 
affect the homology of the structures. 
The close union of the pollinium-cells, each surrounded by 
a firm cellulose wall, and the fact that neither they nor the 
tetrads are ever set free, would be misleading in attempting to 
homologize them. Brongniart’s sections of the pollen-mother- 
cells of Cucnrbita maxima suggest in appearance the cross 
sections of the pollinium of Asclepias , yet Rosanoff has shown 
for Mimoseae a similar condition of pollen-grains. 
1 Coulter (1898) for Ranunculus ; Fullmer (1899) for Hetnerocallis fulva. 
3 Wimmel (1850' for Fuchsia ; Wille (1886) for Orchis mascula. 
