ARTIFICIAL PARTHENOGENESIS IN CUMINGIA 31 



four nuclei. If the first polar body is formed in the normal 

 manner, the second polar spindle may form two nuclei which 

 fuse. Lefevre finds, .however, that there is no difference in the 

 course of development, whether the eggs are fertilized by the 

 first or by the second polar nucleus or obliged to develop with 

 the haploid number of chromosomes. 



The results obtained by Allyn ('12) in her study of artificial 

 parthenogenesis in Chaetopterus seem to be of the same nature; 

 though one cannot be sure, because of the small amount of ma- 

 terial she had for cytological study. There is evidence that the 

 first or the second polar body may be retained in the egg and 

 that the cleavage spindles have the diploid number of chromo- 

 somes; but it is not shown conclusively whether this is a neces- 

 sary preliminary to development or not. 



Kostanecki's study ('04, '11) of parthenogenesis in Mactra is 

 interesting as the only other cytological study of parthenogene- 

 sis in moUuscs. Here, as in the cases already mentioned, one 

 or both polar bodies may be formed, or the egg may omit matura- 

 tion entirely. No self-fertilization was observed here. The eggs 

 developed to swimming larvae, often by a process of repeated 

 nuclear divisions followed by a simultaneous cleavage into a 

 number of small cells. 



In all of these cases, then, the formation or suppression of 

 polar bodies seems to be a matter of indifference. They are 

 hardly to be compared with the development of naturally par- 

 thenogenetic eggs, for the cleavages are irregular, and the whole 

 condition evidently pathological. In the sea urchin, indeed, 

 where both maturation divisions occur, development is fairly 

 normal. Artificial parthenogenesis in this form might be com- 

 pared to natural parthenogenesis in the hymenoptera in which 

 both polar bodies are formed. The phenomena in Cumingia, 

 on the other hand are comparable to the cases of natural parthen- 

 ogenesis in which one polar body is suppressed. To make the 

 comparison exact, one must suppose reduction to occur normally'' 

 with the formation of the first polar body in Cumingia. This 

 seems a warrantable assumption when we consider the relation 

 of the two maturation divisions to parthenogenetic development. 



