198 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 39.^ 



same size (fig. 5).^ Again these are surrounded by considerable sur- 

 plus material and now there are several accessory nuclear centers. 

 Again also we can follow the migrating material for some distance into 

 the yolk. We may call this a two-cell stage, but we must remember 

 that it is not such in the sense that the two cells are the halves of the 

 original one. They have rather been built up independently of each 

 other, and are each formed of separate material that has migrated 

 out of the yolk. 



This same accumulation of material around nuclear centers goes on 

 until there are four or five of these large cells, of about the same size 

 as the original one. Then, while the same process continues, a seg- 

 mentation of these large cells also begins, and we find eggs with from 

 12 to 16 medium-sized cehs, only half their former diameter (fig. 6). 



There is still much surplus material around them and many small 

 nuclear centers, but the migrating material can not be followed as 

 readily into the yolk. 



The amount of migrating material and the rapidity with which it 

 comes out from the egg would seem to exert a controlling influence 

 upon the kind of segmentation which results. 



If the cytoplasm and nuclear material emerge all at once, the 

 former gathers around the latter in a single mass, which by subse- 

 quent division forms the blastoderm. This is true discoidal seg- 

 mentation. But if the substances migrate more slowly, those which 

 reach the surface first have time to accumulate while the others are 

 still emerging. The result is not a single cell but several such cells of 

 nearly the same size, as in the present instance. Moreover the 

 removal of the last of this material from between the yolk globules is 

 accomplished more slowly than at first, either because, coming from 

 the center of the egg, it had a longer distance to travel, or because the 

 actual rate of emergence has diminished, or possibly both of these 

 combined. The blastoderm cells first formed thus have plenty of 

 time to segment before the material is ah out of the egg. And since 

 there is not as much of the migrating material now as there was at 

 first, the size of the accessory cells which it forms and adds to the 

 blastoderm is about proportional to the size of the cells in the latter. 

 Since these accessory cehs become a part of the blastoderm and enter 

 into the formation of the embryo, this can not be regarded as discoidal 

 segmentation. 



BLASTULA. 



The two processes of aggregation and cell division go on simul- 

 taneously until the blastoderm has spread superficially over the entire 

 yolk. In the intermediate stage represented in fig. 7 it can be seen 

 that the blastoderm cells have appreciably diminished in size, and it 



a In the figure the right cell is cut at a different level and so appears smaller than 

 the left one; in the second section following it is fully the size of the latter. 



