1924] Mann: Microsporogenesis of Ginlcgo biloba L. 245 



tube of Cycas which greatly resemble those seen in the pollen mother 

 cells of Ginkgo. 



The procedure described above raises a number of interesting 

 questions. Firstly, the position of the starch grains bears a definite 

 relation to the formation of the cell walls. They are generally dis- 

 tributed while the first wall is forming, and it ceases to thicken when 

 they withdraw. They lie near the forming cell plate, and near the 

 thick inner wall during its formation. Finally, they become generally 

 distributed during the formation of the pollen cell wall. They are also 

 smaller than they were during the formation of the thick inner wall 

 of the pollen case. It seems possible that they may provide the reserve 

 material which is utilized in wall formation. 



Secondly, the method of cell wall formation differs from that 

 common to the higher plants. This is of particular interest on account 

 of the phylogenetic position of Ginkgo. 



The changes of position of the starch grains are essentially the same 

 as those noted by Terni (1914) for the chondriosomes in the spermato- 

 genesis of Geotriton fuscus, and by Payne (1916) for certain scorpions. 

 The mitochondrial mass forms the tail sheath in such spermatozoa. It 

 would be interesting to know the fate of the plastids in spermatozoa 

 formation of Ginkgo. 



The similarity of behavior of the chondriosomes during cell division 

 to that observed for the starch-filled plastids of Ginkgo indicates that 

 the distributing mechanism is very similar in each case. It does not 

 seem necessary to postulate a separate mechanism for this purpose, the 

 forces already in action being of a type which would, it seems to me, 

 bring about essentially the observed distribution. Whatever the force 

 or forces may be by which the chromosomes are distributed during 

 reduction, the direction of movement of the chromosomes shows the 

 direction in which these forces act. One would expect that a large 

 number of movable cytoplasmic structures or inclusions would be 

 equally distributed between the cell wall and the nucleus, if in- and 

 out-going currents maintained an equilibrium during the early pro- 

 phase. After the spindle has formed, and the cell is a bipolar structure, 

 the forces (which we may think of as currents) move toward the poles, 

 and presumably back toward the equator. The changes of position 

 of the plastids at this stage indicate such lines of force. As the nuclei 

 grow, apparently by taking in fluid, their increase in size would also 

 tend to force the plastids out of the polar and into the equatorial 



