ON THE DEVELOPMENT OF NEBALTA. 395 



of five cells of the blastoderm, and oii each side of this in- 

 pushing there is a cell (fig. 6, mes.). 



There can be little doubt that the entrance to the groove 

 represents a blastopore, and that there is here an invagination 

 of blastoderm cells to form the endoderm. The cells on 

 either side of the invagination I take to be mesoderm, and to 

 have been budded off from the blastoderm in situ. 



In the section close to the anterior limit of the groove the 

 lumen is narrower. This makes one think the closure begins 

 at the front end of the groove. Just dorsal to the narrow 

 depression there are a few large rounded cells (fig. 5). These 

 I take to be endoderm cells which have been invaginated or 

 have arisen by proliferation from invaginated cells. 



The section behind the blastopore also shows two layers of 

 cells in the ventral region (fig. 7). I find it hard to say 

 whether the inner layer consists entirely of mesoderm or not. 

 The large almost central cell may be endodermal, the others 

 are certainly mesoderm. 



All the cells in this and in the next stage contain a very 

 granular protoplasm. 



Stage B' (fig. b'.). 



External View. — The embryo lying on what will be the 

 ventral surface of the yolk shows three distinct regions of 

 thickening, namely, the two optic thickenings at the anterior 

 end, and the caudal thickening at the posterior end {c.t.). 

 These three thickenings are connected by strands of cells in 

 which strands there are again thickenings foreshadowing the 

 antennse. 



One can also see in this external view an indication of the 

 caudal groove between the caudal thickening and the blasto- 

 derm in front of it. Many dividing nuclei can be seen in the 

 central region of the ventral surface of the blastoderm. 



Internal Structure. — Transverse sections through the 

 optic thickenings show them to consist of cylindrical cells 

 (fig. 8), and of only one layer of these, though here and there 



