496 REPORT OF COMMISSIONER OF FISH AND FISHERIES. [42] 



Strang, His) has been split off distinctly from the epidermal layer over- 

 lying it. The formation of the nenrula occnrs, as will be inferred from 

 what has already been said, in one of the radii of the blastodermic 

 disk. The sensory layer thickens perceptibly aloag this axis at a 

 very early stage. In the cod's egg this is perceptible on the eighth day 

 of development, as shown in Fig. 19, pi. IV, where the band of cells Pst, 

 which pass inwards towards the center of the disk of the blastoderm, 

 are the nidiment of neurula or medullary plate from which the nervous 

 system will be differentiated. This rudimentary nervous cord contin- 

 ues to become thicker in a vertical direction as development proceeds, 

 and at last begins to be apparent as a ridge on the under side of the 

 blastoderm. The ventral ridge or keel of the neurula becomes more 

 apparent ou the ninth day at the fore part of the embryonic axis, as 

 may be seen in a side view of an egg of that age with its blastoderm 

 in optic section represented in Fig. 20. Another view of an egg of the 

 same age is given in Fig. 21, but here the head end of the embryo is 

 directed toward the observer, and in the darkly shaded portion, which 

 is an optic section of the fore part of the body of the young embryo, 

 shows the neural keel prominently developed on the side next to the 

 yelk. In the embryo of other animals, with two or three exceptions, 

 the development of the neurula takes place in an entirely different 

 manner. In most forms in fact, when the neurula is forming, the epi- 

 blast becomes grooved on its external face, while, as the furrow so de- 

 veloped deepens, its sides fold over toward each other to join in the 

 middle line, leaving at the same time a canal throughout the whole 

 length of the nenrula and even continued at the hinder end into the 

 primitive intestine or archenteron. ifot so with the neurula 6f the 

 embryos of osseous fishes. Here the neurula is at first an absolutely 

 solid cord or strand of cells which by a slow thickening of the sensory 

 layer at last forms a deep laterally compressed mass of cells which juts 

 down into the yelk, pushing the hypoblast before it. It is even difficult 

 to prove that there is ever any such a thing as a medullary groove or 

 furrov7 developed at all in the sense in which we know it in the embryos of 

 Amphibians, for example. However, our Fig. 21, pi. IV, shows a de- 

 pression in the median dorsal line of the embryo at mg which we may 

 regard as representing the medullary groove in the blastoderm of the 

 cod of the ninth day. Such a feature is also shown by O^llacher in his 

 figures of sections of the blastoderm of the trout; it is also developed 

 in that of the shad. Oalberla in his studies of Syngnathus states that 

 the epidermal layer of epithelium is carried down by invagination into 

 the medullary plate along its mesial axis as the neurula is developed. 

 In such sections of similar stages of Clupeoids as I have had the op- 

 portunity to study I have seen no evidence of anything of the sort. 

 In this respect they are like the embryos of Lepidostens investigated 

 by Professor Balfour. In this regard (Ellacher's investigations upon 

 the trout seem to coincide with the results of Balfour and myself. As 



