262 BOTANICAL GAZETTE [march 



Discussion 



The two main points wherein this description disagrees with 

 that of Woodburn (12) are as follows. First, according to that 

 author there are no indications of centrosomes in the spermato- 

 genous mitoses, the blepharoplast first appearing as a cytoplasmic 

 differentiation in the androcyte. On the contrary, the present 

 writer finds that centrosomes are present at all stages of the last 

 mitosis, and that these persist as the blepharoplasts of the andro- 

 cytes. Second, Woodburn states that the blepharoplast in the 

 androcyte undergoes a simple elongation to form the cilia-bearing 

 thread, whereas the present writer sees it fragmenting to several 

 pieces which coalesce to form the thread somewhat after the man- 

 ner of the blepharoplasts of Equisetum and Marsilia (Sharp 8, 9). 



It is not improbable that this disagreement is due in part to 

 actual differences in the two lots of material studied. Although 

 the single species of the genus, Blasia pusilla, was used in both 

 instances, a comparison will show that the cells described in the 

 present account are little more than half the size of those figured 

 by Woodburn. Although it is possible, therefore, that the two 

 lots of material represent two varieties, too much weight should 

 not be placed upon a size difference, for it is known in certain 

 cases {Equisetum, Sharp 8) that androcytes and spermatozoids 

 often vary considerably in size in the same lot of material. 



Lack of agreement as to the presence of centrosomes during 

 mitosis is perhaps not surprising. Because of their extreme 

 minuteness the centrosomes might easily be overlooked in the 

 stages previous to that at which Woodburn first finds them, and 

 at which they enlarge and become really conspicuous for the first 

 time. With regard to the fragmentation of the blepharoplast, on 

 the other hand, it is more difficult to understand why material 

 actually the same should be interpreted so differently. In the 

 writer's material the process of fragmentation is shown with great 

 clearness; only occasionally is anything found in good prepa- 

 rations which might be interpreted as a uniformly elongating ble- 

 pharoplast. Moreover, in no case has a condition approaching 

 that shown in Woodburn's fig. 11 been observed. The nucleus 

 becomes closely applied to the blepharoplast when the latter is in 



