Charles L. Edwards and Clarence W. Hahn 347 
tinuous with the hypoblast which does not pass under the chorda. As 
we shall point out at another place, the condition here referred to may, 
and perhaps should, have a slightly different interpretation than that 
which Mitsukuri has put forward. The spread of mesoblast laterally 
from the primitive plate begins later in Clemmys than in Phrynosoma 
if we may assume that the blastopore appears at the same time in each 
case. 
The early development of the Squamata differs but slightly in the two 
suborders. The Ophidia are like the Chelonia in having the cellular dif- 
ferentiation less marked than the Lacertilia. The shape of the embryonic 
area in the Ophidia is not clear from Hertwig’s two figures of the earliest 
stages (Hertwig, 03, Fig. 415, A and B). The blastopore is more or 
less posterior to the embryonic area. It is a broad, posteriorly arched de- 
pression with no sharp angles. In slightly iater stages (Hertwig, 03, 
Fig. 424, A and B) the embryonic area becomes narrow posteriorly, a 
condition not encountered in Phrynosoma at any time (Figs. 1 and 5). 
It is but shghtly elevated in the Ophidia and consequently not sharply 
marked off from the extra embryonic area by a bench (Phrynosoma, 
Fig. 1). The blastopore of both moves, relatively speaking, into the em- 
bryonic area as ontogeny progresses. ‘The mesoblast sac in the snake is 
clearly shown by Hertwig’s figures (03, Figs. 427 and 428). It differs 
from Phrynosoma in having the epiblast and chorda anlage less thick 
and less compact and in having no specially marked lateral pockets from 
the blastopore. Cell islands in the subgerminal region are characteristic 
of the snake. At the time the mesoblast sac comes into communication 
with the subgerminal space, isolated groups of cells remain where the 
lower wall of the mesoblast sac and the hypoblast broke away, thus estab-. 
ishing the neurenteric canal: (Hertwig, 03, Figs. 430, 431). While this 
condition probably exists in Phrynosoma, nothing of the kind has been 
observed in the stages thus far examined. Hertwig’s figures of cross 
sections (Hertwig, 03, Figs. 443-447) represent a solid outgrowth of 
mesoblast from the lateral angles of the blastopore and at no place do 
they reveal a lamination of the mesoblast suggesting a separate origin 
for somatic and splanchnic layers. 
The gastrulation of lizards as worked out by Wenckebach, g1, has 
nothing of striking contrast to the particular genus we are now studying. 
In Lacerta (Will, 95, Wenckebach, 91) and in Platydactylus (Will, 92, 
Taf. 6), there is not such a marked elevation of the embryonic area. 
Both have the mesoblast sac developed in a characteristic way, but con- 
ditions in the Gecko seem a little more closely allied to those of Phryno- 
soma. In a very early stage there is a much larger space between the 
