CKANGON VULGARIS. 131 



come entanglecl among the iiivaginatiiig cells and thiis 

 be carriecl back iiito the yolk where they would form a 

 pai't of the endoderm without ever having taken part in 

 the formation of the blastoderm. Such an elFect might 

 result from an increase of deutoplasm in proportion to 

 the protophism. Still further increase it and more cells 

 woukl be delayed and finally enough would remain in the 

 yolk to form the whole of the endoderm. Such a process 

 is in perfect harmony with the theory of acceleration and 

 retardation of Professors Cope and Hyatt ; and it would 

 be accompanied by a considerable saving of vital force to 

 the egg, 



The endoderm cells in eggs with a large yolk need to 

 take a position in dose connection with the deutoplasm, 

 for from the moment of their formation they are actively 

 engagedin assimilathig it (cf. ßeichenbach, '86, pp. 101- 

 102, pl. VIll, fig. 67) ; hence any process whicli leaves 

 them scattered through the yolk is an evident advantage 

 to the embryo. The mesoderm, on the other band, is 

 first needed in the neighborhood of the developing appen- 

 dages where muscles, etc., will be earliest required, and 

 hence it is no economy to the individual to change the 

 mode of its formation. From this reason, as well as 

 from heredity, the egg would retain the appearance and 

 go through the motions of gastrulation, even though it 

 formed no endoderm by the Operation y and the result 

 would be such as has been described by the majority of 

 observers. 



The conflictins: accounts of recent date are those of 

 Bruce, Patten and Korotneff. In the case of Bruce ('85) 

 I think a reconciliation is to be effected on the supposition 

 that he has misinterpreted bis observations. A compari- 

 son of his account with that of the Hertwigs seems con- 

 clusively to shovv that the yolk cells are to be regarded as 



ESSEX INST. BULLETIN, VOL. XVIII. 17 



