126 THE DEVELOPMENT OF 



of the mid-gut and consequently are to be regarded as 

 endoderm cells." The large yolk cells are not regarded 

 as taking any part in the formation of the endoderm, but 

 nothing is said of their fate. Apparently (if we may 

 judge from the two figures given) the process in Thyri- 

 dopteryx is much like that in Noctua, as described by the 

 Hertwigs, and the layer which Bruce describes as endo- 

 derm is regarded as mesoderm by them. In Bruce's fig. 

 2, this layer bends around almost exactly as it does iu 

 Noctua to form the splanchnopleure. Bruce's "clear mi- 

 gratory cells" are not represented in Noctua. Bruce says 

 nothing of their fate nor does he indicate how the meso- 

 derm arises. While he quotes Balfour, Kowalevsk}^ 

 Tichomiroff and Dohrn, he fails to refer to the Hertwigs 

 in connection with the origin of the endoderm. 



In the bee, according to Grassi ('85) the blastoderm 

 is formed by a migration of the amoeboid cells to the sur- 

 face where, at one end of the egg, they at first form a layer 

 of cells which gradually increases until the whole is 

 covered, just as was described by Kowalevsky. Numer- 

 ous nuclei are left in the yolk. In the median ventral line 

 the formation of the mesoderm takes place, a broad plate 

 of the blastoderm sinking and being overgrown by the re- 

 mainder of the blastoderm. At first this mesoderm is a 

 single cell in thickness, but it soon becomes two or more 

 cells deep. This closing in takes place first in the anterior 

 third of the embryo and is concluded at the posterior 

 end of the germ. After this mesodermic plate is formed 

 and enclosed by the ectoderm it grows forwards and back- 

 wards beyond the limits of its origin, curving at either end 

 to surround the yolk. It is these terminations that Grassi 

 regards as forming the endoderm. In other words he de- 

 rives the endoderm from the anterior and posterior ends of 

 the mesodermal plate. He refers to numerous sections 



