﻿1903] EMBRYOGENY OF ZAMIA 187 



our preparations. Webber (11) observed the second division 

 and states that the first two divisions occur near the center of 

 the egg, and that the nuclei then become scattered, and that 

 "finally in the first stage of the organization of the embryo form 

 a layer of cells about the periphery of the ^gg cell." We were 

 able to count the nuclei in the i6-nucleate and 32-nucleate 

 stages, being of course the product of the fourth and fifth simul- 

 taneous divisions. The 64-nucleate and 128-nucleate stages 

 were counted approximately. In two cases the mitotic figures 

 of the eighth division, which gives rise to the 256-nucleate stage, 

 ) were counted with considerable certainty, assuming, as seems 



safe, that these early divisions are all simultaneous. In both of 

 these series, which consisted of more than fifty sections, from 

 several of which the proembryo had been lost in making the 

 preparations, over one hundred figures were counted. Since the 

 seventh division would show only 64 figures and the ninth divi- 

 sion 256, there could hardly be a possibility of mistaking this 

 I stage, A section from one of these series is represented '\vijig.4. 



. The entire mitotic figure is formed within the nuclear mem- 

 brane, which does not disappear until the anaphases are reached. 

 The kinoplasmic portions of the figure are highly developed, the 

 polar radiations being particularly conspicuous. The whole 

 figure differs decidedly from the one concerned in the formation 

 of the ventral canal nucleus [fig, i), for in this case the nuclear 

 membrane disappears before the metaphase is reached, and there 

 are no polar radiations whatever; and also from the later divi- 

 sions in the embryo, in which the nuclear membrane disappears 

 early. That simultaneous divisions cease with the eighth divi- 

 sion may be regarded as certain, for numerous preparations of 

 the proembryo just before the formation of the cell wall {figs-5^ 

 ^) fail to show even approximately the large number of nuclei 

 (512) which should be found if a ninth division had taken place. 

 The appearance of the proembryo at the close of free nuclear 

 division, but before the formation of any cell walls, is shown in 

 J^g^' 5 and 6. In both these figures the nuclei are scattered 

 throughout the entire ^gg, there being no tendency to form a 

 large central vacuole with consequent parietal placing of the 



