174 



becomes from two to three or more cells deep. This thickened portion, 

 of the epidermis has nothing whatever to do with the formation of any 

 part of the nervous system. The deeper cells of this thickening form 

 provisional gland-cells , which serve to attach the embryo, after its es- 

 cape from the egg-membrane, to the ventral side of the parent. In this 

 manner the young are carried about until the posterior sucker is deve- 

 loped sufficiently to serve as an organ of attachment. These gland- 

 cells are stained dark brown, and are thus very easily distinguished 

 from the lighter-colored nerve-cells lying beneath them. 



In Aulostomum and Nephelis^ according to Bergh, the primitive 

 epidermal layer is lost, and the second layer of the germ-bands gives 



A diagrammatic surface view of the second layer at the 

 posterior end of the nearly completed germ -bands. 



n =■ nerve-chain. 



ne'ph = nephridia. 



m = mesoblast (?) 



ah cd a'b'c'd' = the terminal proliferating blastomeres. 



origin to the nerve-chord and the definitive epidermis. Ten proliferat- 

 ing cells are found at the posterior ends of the germ-bands, as in Clep- 

 sine;' and in Nephelis five distinct lines of cells are traced to the 

 five proliferating cells of each band. These lines of cells are present 

 in Aulostomum^ but are not so regularly arranged as in Clepsine and 

 Nephelis. It is interesting to note that the two median rows in Aulo- 

 stomum are perfectly distinct from the other rows, a fact which makes 

 it all the more probable that they have the same destination as in Clep- 

 sine. It would seem from one of Bergh's figures (Semper's Arbeiten, 

 VII, PI. XII, Fig. 5) that the larval nephridia may arise from the same 



