16 THE DEVELOPMENT OF 



the formation of separate epithelial caps which later grow 

 together as in Crangon and produce the lobed condition of 

 the adult. In Curaa (Dohrn, I.e., p. 6) : The liver "besteht 

 aus einer massig gewolbten kuppelformigen Erhohung, 

 deren Basis ringformig eine Oeffnung umschliesst, durch 

 welche die communication desLebersackes mit dem Dotter, 

 spiiter mit dem Darme stattfindet." All of which agrees 

 well with what is given above. 



I regret that the material at hand does not decide more 

 definitely the question of how much of the alimentary ca- 

 nal is of entodermal origin, but I think the inspection of 

 the figures referred to in connection with this section of 

 the present article will show that scarcely more than the 

 "liver" and its ducts can be derived from the entodermal 

 cells. Views of living specimens at a later stage than that 

 represented in fig. 23 show that the hind-gut extends itself 

 still farther forward, its inner termination retaining its fun- 

 nel-like expansion. Up to the stage represented in fig. 

 54 the approximation of proctodcum and stornodeum has 

 been effected not by additions from the entodermal cells 

 but by cell division in these regions of true ectodermal 

 origin. As in later stages, after hatching, the whole middle 

 region of the alimentary tract retains a uniform histologi- 

 cal structure it seems but fair to infer that the whole of its 

 extent has a common origin. In other words, I believe that 

 the whole of the straight canal is produced by the stomo- 

 deum and proctodeum and that the entoderm is limited to 

 the liver or mid-gut gland of Frenzel, and the immediate 

 vicinity of its ducts. 



It will readily be recalled that this view is not new. A 

 close parallel will be found in Oniscus as described by 

 Bobretzky ('74). There the straight part of the tract is 

 described as being formed exclusively of stomodeum and 

 proctodeum while those cells which, in the former part of 



