On the larval development of Amin calva. 659 



section of a larva of about the fourteenth day, Fig. L, the liver may 

 be seeu to occupy this positiou: inferioiiy it is conflueiit with the 

 intestine, i, in the region of the defiuitive gall duct ; dorsally its finely 

 raraose substance appears to be attached to the Peritoneum, aiul be- 

 tweeu the gullet aud the iutestine it exteuds to the opposite visceral 

 wall: it here becomes apposed to the ventral wall of the yolk sac, 

 although not fusing with it. 



The relation of the yolk sac to the alimeutary canal may next 

 be noted. The origin of the dorsal wall of the gut has been out- 

 lined in the writer's paper "On the early development of Amia." The 

 lumen of the mid-gut appears to be formed by the gradual uplifting 

 of the head region of the embryo, and the gut comes to be constricted 

 otf from before backward: thus at the stage of hatchiug the hinder 

 gullet and the anterior region of the stomach, Figs. K, J, H, is en- 

 tirely separated from the yolk; the hinder portion of the stomach, 

 however, Fig. H, is confluent with the yolk. Hindward of this position 

 the dorsal wall of the gut becomes constricted and more and more 

 flattened, the lumen becoming reduced to a mere fissure. The out- 

 growth of the tail region causes an enlargement of the cavity of the 

 gut, Fig. G, as well as giving rise to the evagination which is to 

 form the hinder intestine, I and FAG; hindward of this evagination 

 the wall of the gut flattens to the underlying yolk and shortly merges 

 with it. It will thus be seen that the relations of the yolk to the 

 alimentary canal difier but little from those of the typical mer^blastic 

 Vertebrates, the Teleosts in particular. From the latter, however, 

 Amia differs broadly in the mode of its yolk absorption. In Serrantis 

 (H. V. Wilson) the liver becomes attached to the yolk-mass and aids 

 directly in its absorption — a process, "probably akin to intj&cellular 

 digestion", in which the liver "cells establishing a connection with 

 the yolk, form a feeding or absorbing surface, which, as it incorporates 

 new material on its yolk side, as constantly splits otf new cells on 

 its liver side". In Amia, on the contrary, the liver has no direct 

 connection with the yolk, its growth being supplied largely by the 

 vitelline (and intestinal) veins. That this condition may well prove 

 ancestral in the evolution of bony fishes {Serranus) appears not un- 

 natural when the following relations of the later yolk absorption in 

 Amia are considered. In the larval Amia of seven days the trans- 

 verse section of the liver region, Fig. L, already noted has shown 

 that the yolk material has become largely absorbed: within three or 

 four days, indeed, it will have entirely disappeared. In this section 



43* 



