136 THE DEVELOPMENT OF 



except in its closing, it has nothing in common with a 

 blastopore. 



Nusbaum ('86) failed to see the first stages in Oniscus 

 but that he takes essentially the same view of the blas- 

 topore as that here advocated is readily seen. ^' Au mil- 

 ieudu disque de segmentation (blastopore) forme par une 

 seule couche de cellules, apparait une accumulation des 

 cellules (gastrulation) , dont une partie, comme Bobretzki 

 l'a bien remarque s'enfonce dans le vitellus, pour l'absorber 

 et pour former des cellules vitellines (Dotterzellen) une 

 autre partie reste diffuse au dessous de l'ectoblaste et donne 

 naissance, d'apres Fauteur cite, aux Clements du meso- 

 blaste ; selon mes recherches cette seconde partie donne 

 non seulement le mesoblaste, mais encore l'entoblaste." 

 Nusbaum does not regard the " Dotterzellen " as true en- 

 doderm but thinks that they play a part in the softening 

 of the yolk. His endoderm is derived from two lateral 

 masses at the anterior end of the meso-endo dermal thick- 

 ening and these give rise to the endoderm of the he- 

 patic coeca which Bobretzky thought arose from the yolk 

 cells. He refers to a recent Russian paper by Kowalevsky 

 on the development of the scorpion as agreeing with his 

 results. This I have not Seen. Kowalevsky and Schulgin 

 have a short paper on the embryology of this form ('86) 

 which came to band after the present paper was in the 

 hands of the printer, and which doubtless contains the same 

 facts as that referred to by Nusbaum. Their youngest eggs 

 had the blastoderm complete and occupying one pole of 

 the egg, while neither nuclei nor cells were to be seen in 

 the yolk. The rudiments of the middle and inner germ 

 layers first appeared as a thickening in the middle of the 

 under surface of the blastoderm ; and " nicht selten kann 

 man konstatieren, dass mehrere Zellen von den obern 

 Schicht nach innen getreten sind," certainly indicating a 



