502 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [48] 



The next event in the history of the development of the neurula is its 

 separation from the epithelial layer of the epiblast. This occurs cotem- 

 poraneously with the development of its internal lumen, and proceeds 

 from before backwards. The epithelial layer, in fact probably the sub- 

 jacent layer of the corium itself, has the same history, being developed 

 somewhat sooner in the cephalic than in the caudal region. With this 

 the hitherto flat upper portion of the neurula becomes rounded off except 

 at the tail end. It is now separated from the skin. During this time 

 the skin has commenced to develop pigment in its deeper layer, as 

 shown in Fig. 31. These pigment cells are stellate and exhibit a slow 

 amoeboid or migratory movement as development proceeds, becoming 

 aggregated at a later period by this means into patches upon definite 

 regions of the body. 



With the further progress of development the tail commences to bud 

 out from the caudal end of the embryonic axis, which at this point con- 

 tinues to become gradually thicker and more prominent and finally 

 swells out into a hemispherical prominence just above the point of 

 closure of the blastoderm. This is the rudiment of the tail of the embryo. 

 In an embryo sixteen days old represented in Fig. 32 the tail has grown 

 out for a considerable distance, and it has been bent over to one side on 

 account of the confined space in the egg-membrane, so that its dorso- 

 ventral axis is turned at nearly right angles to that of the body. A 

 slight fold extends over its end and dorsally and ventrally which is 

 entirely composed of the skin layer. This fold, w/, may be regarded as 

 the beginning of the embryonic natatory fold and develops in height 

 from behind forwards as embryonic evolution advances. 



With the outgrowth of the tail the development of the neurula is 

 continued backwards and its extreme hinder extremity remains solid 

 as in Fig. 31 , and continuous with the same mass of cells from which 

 the chorda dorsalis or notochord ch takes its rise. This continuity of 

 the extreme posterior extremity of the notochord with the neurula is 

 maintained until the tail of the embryo is fully developed. Indeed, 

 after hatching even, in the young cod and most other forms studied by 

 the writer, the chorda is lost in a caudal cellular mass and its end is 

 not included within the urostyle until a considerable time after the 

 young fish is free. I have never met with young cod old enough to see 

 the development of the bones of the tail. The muscular layer is also 

 continued backwards into the tail and clasps, on either side, the neurula 

 and chorda. Lastly the skin or dermal layer is also developed in extent 

 in order to keep pace with the tail as the latter lengthens. The whole 

 process of the growth of the tail is a very remarkable one, and in the 

 cod's embryo, for example, it is hard to understand how the material for 

 so much new structure is transported to its new location without the 

 help of a vascular system, no trace of which has yet ai)peared. The 

 new matter seems to be added by apposition and intussusception. How 

 much of this process may be due to the amoeboid properties of the 



