286 FRANZ SCHRADER. 



that the period of fusion has now been passed. It is indeed pos- 

 sible that there is an occasional combination of a giant cell with 

 one of the yolk cells, the latter representing nothing but cleavage 

 cells that failed to migrate to the periphery to establish the 

 blastoderm. Fusion with what were formerly cleavage cells but 

 must now be termed blastoderm cells, can of course take place 

 only at the edge of the giant cell area, where the two types of 

 cells are in contact. But aside from the fact that the giant cells 

 in that location are not larger as a rule than those more centrally 

 placed, it must be considered that the blastoderm cells in their 

 expansion seem to exert actual pressure on the giant cells. The 

 latter are heaped up and finally actually leave the periphery 

 altogether. If such a pressure really exists, it would be constantly 

 cancelled by fusion of adjoining giant and blastoderm cells. 



Nor can there be a continued tendency of giant cells to fuse 

 with each other. Such occurrences are at least not general, for 

 the number of giant cells is at this time slowly but steadily in- 

 creasing, while their individual size is decreasing. Nevertheless 

 the size is evidently variable, so that an occasional division of the 

 chromosomes unaccompanied by cytoplasmic division remains as 

 the most plausible explanation. 



Once the giant cells have migrated to the symbionts and 

 entered into association with them, divisions become rarer. At 

 the same time it must be observed that in the adult Pseudococctis, 

 the former giant cells, then called mycetocytes, contain relatively 

 enormous numbers of chromosomes. The association with the 

 symbionts must therefore have a disturbing effect on the few 

 divisions that still occur, and most probably it is the failure of 

 cytoplasmic division following a normal division of the chromo- 

 somes that thus causes a multiplication of the chromosomes. 

 This idea has already been expressed by Buchner ('21). 



CYTOLOGICAL EVIDENCE. 



A complete cytological consideration of the problem should 

 begin with a study of the maturation phenomena in the egg. 

 Since these primary steps have already been considered at some 

 length in my previous paper ('23), it is sufficient to begin the 

 present account with the polar nucleus, i.e., the combined first 



