Sperm-Cells in Asclepias Cornuti , Decaisne . 133 
probably an early stage in the segmentation of the spireme, 
but the details of this process could not be accurately followed. 
Before the formation of the nuclear spindle, the nucleus 
elongates in the direction of the long axis of the mother-cell 
as described by Strasburger 1 for Senecio vulgaris. 
The mode of disappearance of the nucleolus and of the 
nuclear membrane, and the origin of the nuclear spindle, were 
not observed, though careful search was made for them. The 
nuclear plate stage of the division is shown in Fig. 17. The 
long axis of the spindle is always parallel to the long axis 
of the cell, and does not vary as described by Juel ( 1900 ) for 
Carex acuta. 
Fig. 17 a is a detail of this division. Here the U-shaped 
chromosomes at the equator of the spindle are shown to be 
twelve in number. Fig. 8 represents a tissue-cell in which 
twenty-four chromosomes are shown in the nucleus. After 
careful and extended counting, this has been determined to 
be the full complement for Asclepias 3 as Stevens also states. 
Fig. 3 a is a detail of the division cutting off the tapetal cell 
from the pollen-mother-cell. Here there are fifteen chromo- 
somes shown, indicating that reduction in number does not 
take place in this division. It thus becomes evident that 
numerical reduction takes place in the division figured in 
Fig. 17. 
Cell-division follows directly upon the division of the 
nucleus. The two daughter-nuclei form at the poles, and the 
cell-plate appears well defined. The daughter-nuclei become 
separated from the spindle-fibres very early, and the develop- 
ment of the cell-plate into a cell-wall takes place without any 
connexion with the daughter-nuclei (Fig. 18). This has been 
described by Stevens ( 1898 ), who showed its significance in the 
light of Haberlandt’s 2 ( 1887 ) declaration that the influence of 
the daughter-nuclei propagated along the connecting fibres 
is necessary for the development of the cell-plate into the 
cell -wall. 
Inspection of Figs. 18 and 20 shows delicate kinoplasmic 
1 1 ^79? P* 9* 3 Quoted by Stevens, 
