502 ^Y{scons^n Academy of Sciences, Arts, and Letters. 



group lying practically in the same plane. (Fig. 10.) During 

 these stages the spindle is frequently much elongated so that at 

 the beginning of the anaphases the central spindle is drawn out 

 into a fine thread in its middle part (Fig. 12.) During the 

 anaphases the chromosomes fonn at first somewhat rounded 

 dense masses that later form the daughter spirems. Xew 

 nuclear membranes are formed and the daughter nuclei are 

 completely organized. While the fonuation of the daughter 

 nuclei is taking place they undergo a remarkable change of po- 

 sition. With the disappearance of the central spindle the 

 nuclei approach each other so that by the end of the anaphases 

 they lie almost in contact. (Figs. 12-14.) The explanation of 

 this change in position is not at all evident. I have described 

 a similar phenomenon in connection with the formation of the 

 cell plate in the onion and larch and suggested that it might be 

 due to the mechanical pressure of the cytoplasm upon the daugh- 

 ter nuclei which pushes them into the space left vacant when the 

 central spindle disappears. The same explanation may pos- 

 sibly hold here although the small number of fibres constitut- 

 ing the central spindle would seem to render it less probable. 

 The juxtaposition of the daughter nuclei thus brought about 

 gives a striking appearance to the cells where all of the nuclei 

 are frequently in pairs. This position of the nuclei, as pre- 

 viously stated, may persist in the cleavage stages. Fig. 20 pre- 

 sents a curious condition in which it appears as if the chromo- 

 somes, instead of going back to the poles in compact bodies, as 

 above described, are here more or less strung out along the 

 spindle fibres either singly or in irregular clumps. The explana- 

 tion of these figures is not apparent. It is possibly due to some 

 peculiar effect of the fixing fluid although the other cell contents 

 show no abnormalities. 



Interesting variations in the size and form of the spindle are 

 shown in figures 17-19. In figures 18 and 19 is shown very 

 clearly that two spindles lying closely adjacent to one another 

 may vary as much in size as do the niiclei in some cases in which 

 division is not clearly in progress. That the apparent differ- 

 ence in size is not due to a possibly somewhat flattened spindle 

 viewed in different aspects is showm in polar views of the equa- 



