CRANGON VULGAKIS. 113 



it is the protoplasm of the egg, the whole being enveloped 

 with the deutoplasm, a condition just the reverse of that 

 implied by the terms endolecithal or centrolecithal. The 

 first segmentation is confined to this central protoplasm, 

 and it is not until the second segmentation is nearly com- 

 pleted (vide fig. 2) that any of the protoplasm reaches the 

 surface ; and for a long time afterward that which remains 

 behind continues to undergo cell-division as well as that 

 which has earlier reached the surface and has there begun 

 to form the blastoderm. Hence at first, the segmentation 

 is clearly not superficial. 



The same state of affairs is recognizable throughout the 

 whole of the series of so-called centrolecithal eggs, as can 

 readily be seen by an examination of the results of all who 

 have studied arthropodan segmentation by means of sec- 

 tions. It is even to be recognized in the results of many 

 of the earlier workers. It is not necessary to give an ex- 

 haustive resume of the work of previous students but a 

 few may be instanced in support of this position. 



In the Crustacea but few have carefully studied the phe- 

 nomena of segmentation, and in some instances (Moina, 

 Cetochilus, Lucifer) they throw but little light upon the 

 present point. Haeckel's observations on the segmenta- 

 tion of Peneus ('75) seem at first sight to conflict with 

 this view, for he represents the egg at the end of the second 

 segmentation as divided into four segmentation spheres, in 

 each of which is a nucleus, while the spheres are united 

 at their inner surfaces in an undivided mass of yolk. The 

 later stages present the same appearance. When we con- 

 sider that Haeckel depended entirely on optical sections, 

 an explanation readily suggests itself. He does not give 

 the first segmentation, and if we regard his nuclei as really 

 nuclei enveloped with protoplasm like those of Crangon, 

 which are migrating toward the surface, the correspond- 

 ence between the two is at once evident. 



