farr: cytokinesis of pollen-mother-cells 301 



most common. This is, of course, what we have observed 

 above. But it should be remembered that the tetrahedral 

 arrangement has been shown in Nicotiana to be determined, 

 not during the division of the mother-cell, but previously near the 

 close of the nuclear divisions. It is quite probable then that the 

 restraining tension of the wall has very little to do with initiating 

 the tetrahedral arrangement, though it may play a large part in the 

 maintenance of such orientation. It is quite possible, as noted, 

 that the appearance of the tetrahedral arrangement is to be 

 attributed to a mutual repulsion between the daughter nuclei, and 

 an attraction between them and the plasma membranes. It may 

 be also that the movements of the plasma membrane in forming 

 the furrow during cell-division, as described above, are in part 

 at least an expression of this same force of attraction between it 

 and the nuclei. 



Each isthmus of the single mother-cell does not necessarily 

 constrict at the same rate as the others (figs. 34, 35, 36). The 

 division may, in fact, be entirely completed between two nuclei, 

 before the isthmus between others has fairly begun to narrow. 

 As the division proceeds the two sides of the same furrow come to 

 lie more nearly parallel (fig. 35), resulting in each of the daughter 

 cells approaching more nearly to the spherical form. It is quite 

 likely that this narrowing of the furrow is, in fact, to be attri- 

 buted to the surface tension of the daughter cells. The portion 

 of the surface of mother-cell over each nucleus, however, retains 

 the form of an arc with a longer radius, indicating that the 

 mother-wall is quite resistant to the rounding up process. 



As division is completed, there is quite a definite system of 

 radiating fibers in the cytoplasm (fig. 36), running from the 

 nuclear membrane to the plasma membrane and quite evenly 

 distributed over the latter. Finally, however, the cytoplasm 

 returns to the reticular appearance which it bore at the beginning 

 of synapsis (fig. 37), and the spores lie imbedded in the mother- 

 cell-wall as a sort of matrix while the spore-coats are being de- 

 veloped. 



VIII. Other dicotyledons 



In the five other dicotyledons studied, the process of cyto- 

 kinesis in the pollen-mother-cells was found to agree in all general 

 particulars, as far as studied, with that in Nicotiana. In view of 



