6 
J. T. CUNNINGHAM. 
the diagram fig. 8, and is referred to in a subsequent part of 
this paper. 
In the stage represented in figs. 5, 6, and 7, the canal of the 
neurochord has appeared. In the posterior part of the embryo 
it is deep within the cord ; in the anterior part it opens out above 
into a flat cavity covered only by a thin layer of epiblast. I 
was not able to satisfy myself whether this layer was only one 
cell deep, but it seemed to be so both above the neurochord 
and over the otocvsts : it seems as if the lower layer went to 
form the neurochord and the wall of the auditory vesicle. 
The explanation of Kupflfer’s vesicle in which my reflection 
on the subject has resulted is a very simple one. 
In the Elasmobranch the invagination of the hypoblast 
which takes place at the posterior end of the blastoderm forms 
a tubular cavity open to the exterior by the blastopore. The 
dorsal wall of this cavity is formed by columnar hypoblast ; 
its floor is occupied by the superficial layer of the yolk con- 
taining yolk-nuclei, that is by the periblast. This cavity n 
the Elasmobranch becomes shut off from the exterior by the 
closing of the medullary groove over the blastopore. In this 
way the cavity comes to be continuous posteriorly with the 
canal of the neurochord, a neurenteric canal being constituted 
which subsequently disappears. The history of this stage n 
the Amphibian is precisely similar, with the exception that in 
the Amphibian the segmentation is complete, and the yolk 
is wholly composed of large nucleated cells instead of con- 
taining nuclei only near its surface. The cavity of invagina- 
tion of which I am speaking never obliterates, it becomes later 
the lumen of the intestine. The intestine is formed simply by 
the differentiation of the walls of the cavity. The dorsal wall 
of hypoblast is hollowed out longitudinally and its lateral parts 
approach one another; the floor of the alimentary canal 
is formed by cells differentiated round the nuclei of the 
periblast. 
Kupflfer’s vesicle in the Teleostean is the homologue of the 
cavity of invagination in the Elasmobranch and Amphibians : 
it is the rudiment of the primitive gastrula-cavity, of that part 
