276 C. O. WHITMAN. 



advance? Vipvoiul them, as seen in the posterior part of the figure 

 [e 2))- Kowalevsky {—) has noted the same thing in Euaxes. 

 Whether any of the epidermal cells enter into the nerve-chain in 

 the manner described by Kowalevsky is uncertain. Thus far I 

 have seen no evidence of this. 



d. Segment-cells. — In fig. 87 are seen two colossal cells fs), 

 with ])lain nuclei and nucleoli, lying just above the two outer 

 rows of nerve-cells. Somewhat later, owing to the concentration 

 of the nerve- cells and the growth of the mesoderm towards the 

 dorsal side, these segment-cells are no longer found above, but to 

 either side of the neural cell-group (figs. 88, 89). 



Fig. 88 would seem in this respect to contradict fig. 90 ; but 

 the seeming inconsistency in the relative positions of the nerve- 

 cells and the segment-cells is at once removed when we remember 

 that the germ-bands close earlier at the fore end than at the hind 

 end, and that consequently the difl'erentiations are further ad- 

 vanced in the former region than in the latter. A surface view 

 of the Neurula (fig. 91, 2^ days after exclusion) shows the paired 

 arrangement of these cells in each body-segment. Although 

 these cells are present all through the Neurula period, it is not 

 easy to obtain unbroken preparations much before the conjunction 

 of the germ-bands, on account of the want of coherence among 

 the embryonic elements. Such surface vit-ws reveal/bwr rows of 

 segment-cells, two on each band. A little later (twelve hours 

 after exclusion) they are found in pairs on each side of the gan- 

 glionic chain, so arranged that four cells lie in the same trans- 

 verse })lane — the plane of a septum. Two days after exclusion 

 the cells of the two median rows have both diminished in size 

 and changed their position. They ap})ear to be connected with 

 the groups of cells destined to become the segmental organs, and 

 to follow these in their growth towards the dorsal region. I am 

 unable to give any further account of them. The segment-cells 

 of the two outer rows can be followed for three or four days, but 

 are finally concealed by the tissues forming about them. They 

 can even be seen on the living embryo when viewed in profile 

 (figs. 49, 50). One of these cells is seen in fig. 92 in contact 

 with the cells of the segmental organ, which suggests j)erhaps 

 that it may become the ciliated mouth of the organ. 1 am more 

 inclined to think, however, that these cells are the mother-cells 

 of the future testes. Their position in the walls of tiie septa, 

 which they maintain so far as 1 have been able to follow their 

 history, favours this hy})othesis. The ciliated funnels of the 

 segmental organs are, on the contrary, always found in the 

 middle of each somatomere. 



{e) Segmental Organs. — The segmental organs appear first as 



