﻿1 88 BOTANICAL GAZETTE [march 



nuclei, as inCjxas. In the stage represented \x\ fig, 5 the nuclei 

 are more numerous in the lower part of the proembryo, doubtless 

 on account of the settling of some of them. This settling is 

 sometimes very evident, the nuclei being lined up along strands 

 of cytoplasm, which appear almost as definite as bundles of 

 spindle fibers, although they are only such trails as might be 

 caused by nuclei passing rapidly through the cytoplasm. Very 

 similar appearances were noticed in ^%^^ which had been ruptured 

 by the pressure of the razor in cases where the portion containing 

 the archegonia had been cut out previous to placing in the fixing 

 fluid. In these cases the trails were caused by the rapid move- 

 ment of the cytoplasm toward the point of exit- 

 It is worthy of note that in most of our preparations of the 

 early stages, including the eighth simultaneous nuclear division 

 ^fiS' 4)y ^^ nuclei are nearly all in the lower half of the ^g%\ 

 while in later stages {figs, 5-10) the upper portion is never free 

 from nuclei. In this connection it should be noted also that 

 just before the appearance of cell walls {figs. 5-6) the number 

 of nuclei is often more than 256, although never so many as $12^ 

 the number which would result from a ninth simultaneous divi- 

 sion. It seems probable that many of the nuclei in the upper 

 part of the proembryo, which is not to form any part of the 

 embryo proper, are due to irregular or amitotic division. It is 

 certain that the nuclei of this region are irregular in size and 

 often amoeboid in outline. Some of these nuclei may have been 

 due to the entrance of two male cells, one remaining near the 

 apex and dividing. The two large irregular nuclei shown infig> 

 4 suggest this explanation. While the ventral canal nucleus 

 might divide and give rise to nuclei in this region, we have seen 

 nothing which would make such an explanation probable. 



Ikeno's (8) account indicates that even in the earlier stages 

 of development the embryo of Cycas differs considerably from 

 that of Z.amia, In Cycas, as in Zamia, fertilization is followed 

 by a period of free nuclear division, but in Cycas circinahs 



(Treub, 5) and also in Q^revohita (Ikeno, 8) a large central 



vacuole is soon formed. According to Ikeno, who gives a more 

 detailed account, the nuclear division near the neck of the 



