Sperm-Cells in Asclepias Cornuti , Decaisne . 127 
be sure to just what cells he refers as pollen-mother-cells. 
From his comparison of the large size of the cells relative to 
their respective nuclei, it seems as if he must refer to the cells 
which give rise to the individual cells of the pollinium, for this 
comparison does not apply very well to any of the other cells 
of the pollinium. Juel (1900) says that the manner of tetrad 
division in Asclepiadaceae is still a question. 
Neither Juel nor Strasburger (1900) mention Stevens’s 
(1898) paper, nor is any reference made in the latter to 
Wille’s work or to Strasburger’s statement. Stevens figures 
and describes in part the division of the cells from which 
the pollinium units originate, and calls them pollen-mother- 
cells. 
The general appearance and structure of the pollinium 
suggest that the cells which compose it are pollen-mother- 
cells rather than pollen-grains, and this, doubtless, lent ready 
acceptance to Wille’s interpretation of them. The develop- 
ment of these cells, four in a row, from the large primary 
cells, accompanied by a reduction of the number of chromo- 
somes, as described by Stevens 1 , suggests that the cells of the 
pollinium are pollen-grains. If this is the correct interpreta- 
tion of them, the arrangement of the cells in a row, rather 
than at the corners of a tetrahedron, differs, so far as our 
present knowledge goes, from the mode of formation of pollen 
in all other plants. 
If these cells which develop the fertilization (pollen) tube 
are pollen-mother-cells, we would expect to find the subsequent 
nuclear phenomena and the number of cell-divisions to be 
different from those in the development of the nuclei and 
sperm-cells in pollen-grains, that is, that there would be 
a shortening of the process from the mature archesporial cell 
to the sperm-cells if the microspore were cut out, which 
would be the case if Wille’s interpretation proved to be 
correct. This has already been suggested by Atkinson (1901). 
It was for the purpose of determining the course of events in 
the development of these cells and of the pollen-tube, and the 
1 Loc. cit., p. 81. 
