Albert C. Eyeleshymer and James Meredith Wilson 151 
and covered by a cellular layer which is considered as periblast. The 
author states that the figure is slightly schematized, but to what part of 
the figure he refers is uncertain. If to the periblast we have no criticism 
to offer. If the author, however, intended to represent the periblast as 
it is actually found in Amia we must emphasize the fact that our ma- 
terial shows nothing of the sort. There is no layer of cells between the 
periblast and the gastral cavity. The floor of the gastral cavity is, to 
our minds, the homologue of the periblast in teleosts. We find in-the 
ganoids a complete series of gradations from the teleostean to the am- 
phibian conditions. In Lepidosteus, as described and figured by the 
senior author, 03, we find the closest approach to the teleostean periblast. 
Amia comes next with its homologous layer in the floor of the gastral 
cavity; this floor is made up in part of detached cells and in part by the 
projecting ends of the large yolk masses. From this condition we can 
readily pass to the homologue of the periblast in Acipenser which is the 
layer of cells forming the floor of the gastral cavity. We are thus pre- 
pared to support the statement of the Zieglers, 92, that the floor of the 
gastral cavity in the amphibia is the homologue of the periblast in fishes. 
THE MESODERM. 
The origin of the mesoderm in Amia has been previously studied by 
both Dean and Sobotta. Their descriptions, however, are quite unlike. 
Dean, 96, in describing a stage in which the blastopore has nearly 
closed, says: “ The mesoblast is found to arise peristomal; on the dorsat 
side it arises from the undifferentiated tissue (of the tail mass), 
thence extending forward as a separate cell layer, and finally appears to 
be blended with the loosely associated cells of the entoblast; ventrally 
the mesoderm, although distinctly to be recognized, is not to be sepa- 
rated from the cellular elements of the entoderm. In its early growth it 
extends forward as a wide and flattened cell mass, thinning distally and 
becoming confluent with the inner germ layer. As in the teleosts, gas- 
tral mesoderm is absent, and the division of the middle layer into its 
somatic and splanchnic layers is not apparent until a comparatively late 
stage of development.” 
Sobotta, 96, describes the origin of the mesoderm in a much earlier 
stage in the following words: “ Nun tritt, noch ehe es zur Urdarmbild- 
ung, also zur eigentlichen Gastrulation kommt, eine Differenzirung der 
Furchungszellen zu Keimblittern auf, indem sich eine compacte mehr- 
schichtige Zellage an der Oberfiache des Kies durch einen feinen Spalt 
von den darunten gelegenen, mit grdsseren Dotterkérnern beladenen 
Zellen sondert (Fig. 3). Diese Erscheinung trennt bereits jetzt das Ek- 
