CRANGON VULGARIS. 17 



this paper ('86ö) pp. 134-136, I have shown to be true 

 entoderm are utilized almost wliolly in the formation of the 

 voluminous liver. Between Oniscus and Crangon, as is 

 to be expected, there are minor differences. In Crangon, 

 as we have seen, there are six (three pairs of) liver lobes 

 outlined at first. In Oniscus, on the other band, the structure 

 is more simple. In this connection it may be noted that 

 Balfour, deriving bis facts from Bobretzky, says ('80, p. 

 439) that in Oniscus the alimentary tract "is mainly if not 

 whoUy formed from the proctodeum and stomodeum, " 

 while on the next page be thinks that the entoderm cells, 

 besides furnishing the hepatic epithelium, "probably also 

 supply the growth material for the later growth of the ap- 

 parent proctodeum" and onthis account this portion of the 

 digestive canal "does, in reality correspond to the procto- 

 deum and mesenteron together." 



Looking at Oniscus alone, as described and figured by 

 Bobretzky, it seems to me that the whole alimentary tract, 

 from the openings of the hepatic ducts to the anus is strictly 

 proctodeal in origin, while the light thrown upon the sub- 

 ject by Crangon seems to confirm this vievv. The yolk or 

 entoderm cells in Crangon are larger, and differ in their 

 histological characters from those of the bind gut, and I 

 have never seen a trace of their joining themselves to that 

 pai't of the canal. On the other band, they seem to remain 

 in a passive condition until a coraparatively late stage when 

 they unite, not to piece out either ingrowing portion but 

 to form the hepatic epithelium. 



In Astacus the resemblances would naturally be closer 

 and so I regard them as shown in Reichenbach's ('86) fig- 

 ures, especially in bis PI. xiv, lig. 217. His letter "1>" 

 ("übergangstelle des Mitteldarms in den Hinterdarm") 

 seems to have an arbitrary position while his "ilfZ>/ ("dor- 



ESSEX INST. BULLETIN, VOL. XXI. 2 



