CRANGON VULGARIS. 135 



and, rapidly increasing in numbers, split into two layers, 

 thus becoming differentiated into meso- and endoderm. 

 The cells of the latter layer sink into the yolk, "den gan- 

 zen Nahrungsdotter in sich einsaugen," and thus convert 

 the interior of the egg into a mass of yolk spheres which 

 Bobretzky says form the endoderm. In Scorpio, Mets- 

 chnikofl''s plates certainly leave much to be desired ; but 

 his figures 7 and 9 on plate XIV can be reconciled with the 

 condition occurring in Oniscus, if we interpret the layer 

 cells of fig. 7 and the " zweites Blatt " of fig. 9 as meso- 

 derm. HaeckeFs Interpretation is apparently wrong. In 

 Cymothoa the conditions are apparently the same as in 

 Oniscus. Van Beneden teils ns nothing of the formation 

 of either mesoderm or endoderm in Mysis, but the subse- 

 quent development would indicate a similarity to Oniscus. 

 The same remarks would apply to Nebalia. 



In Oniscus, Bobretzky says there is no invagination, 

 but he says that here the lower layer cells arise from 

 the blastoderm. Delamination here is of a difFerent na- 

 ture from that in the Coelenterates and is easily seen to be 

 but a slight modification of gastrulation. This gastrula- 

 tion takes place at or near the middle of the germinal 

 area, and there is to be sought the blastopore. We know 

 nothing about the presence or absence of a primitive 

 groove in any of these forms. Such being the case, the 

 gastrulation of these genera is of a very difFerent char- 

 acter from that of the teleosts and some other vertebrates 

 where the rim of the blastoderm is clearly the blastopore. 

 In these arthropods there is merely an acceleration of de- 

 velopment, wherel^y the gastrula is formed before the 

 blastoderm has had time to spread over the very large 

 yolk. The case öfters a close parallel to that of the chick. 

 The edge of the blastoderm has nothing to do with the 

 formation of either mesoderm or endoderm, and, hence, 



