290 MEMOIRS OF THE NEW YORK BOTANICAL GARDEN 



Many stages may be found showing different degrees of this con- 

 striction (figs. 9, 10, 12). The daughter-cells are not at first 

 spherical. This may be related to the resistance of the thick 

 mother-wall to transformation of form, or to the unequal rates at 

 which the isthmuses narrow (fig. ii). Upon the completion of 

 the division each of the four cells is separated from each of the 

 other three by a lateral wall which is thicker in its periphery and 

 thinner at its center. Its form persists even after the protoplasts 

 undergo plasmolysis (fig. 13). Since in living cells the protoplast 

 completely fills the space enclosed by the cell-wall, it will be seen that 

 the daughter-cell is not perfectly spherical, but rather has in section 

 the form of an ellipsoid. Such a condition is shown to the right in 

 FIGURE 14. To the left, however, are cells which have proceeded 

 a step farther and have approached the spherical form. Later 

 (fig. 15) the cells become exact spheres. This last change is 

 accompanied by their enlargement and a decrease in thickness of 

 the mother-cell-wall. The latter may involve a dissolution of 

 the inner portion of the cell-wall. The entire cell does not retain 

 its initial outline at this time, but assumes more nearly that of a 

 tetrahedron; the division of the single spherical protoplast into 

 four spherical protoplasts and the growth of the later has involved 

 a considerable increase in the total volume, which has doubtless 

 led to a stretching of the outer portion of the mother-wall. As 

 the exine of the spores appears this mother-wall becomes thinner 

 and thinner and finally disappears entirely. 



FVom a study of the living material it has been shown that 

 normally the mother-cell of Nicotiana thickens its walls during 

 synapsis and the heterotypic karyokincsis, that no division of the 

 cytoplasm occurs after the latter, but that the homoeotypic karyo- 

 kincsis takes place almost immediately and is followed by a divi- 

 sion of the tetranucleate cell into four uninucleate spores. This 

 division involves a furrowing of the protoplast in the planes mid- 

 way between the nuclei. It is not possible in the living material 

 to determine the relation of cell to cell within the tissue, nor to 

 study critically the details of the process of division. 



V. Thickening of the mother-walls 

 In the fixed material it is found, as noted above, that the 

 mother-wall begins to thicken during synapsis, commencing first 



