NEAL: NERVOUS SYSTEM IN SQUALUS ACANTHIAS. 199 
from the ventral to form the muse. obliquus superior, can by virtue of its 
topographic relations to chorda and aorta be regarded as the somatic 
portion of this mesoderm segment (Plate 7, Fig. 56). Its ventral portion, 
which later becomes differentiated into the musc. adductor mandibule, 
is therefore splanchnic. While the indications of the differentiation of 
the 2d somite into myotome and sclerotome are less clearly expressed 
than in the case of the 3d and 4th, I have no reason to question the 
correctness of Killian’s (91) interpretation that such appear. The 
great enlargement of the cavity of the somite is the chief factor in 
modifying its form and the relations of its constituent parts. While 
Miss Platt (91) finds the musculature to arise first in the median wall 
of the somite, that is to say, the dorsal part of the so called “mandibular 
cavity,” Hoffmann (’96) states that the muse. obliquus superior arises in 
its upper and lateral walls. In my opinion their conclusions are not so 
divergent as they might at first sight seem to be, for I believe that the 
portion of the somite which Hoffmann calls dorsal is morphologically 
median ; in other words, that it is the portion which in early stages lies 
against the wall of the neural tube (Plate 7, Fig. 56). I agree with 
Hoffmann that the musc. obliquus superior arises in the dorsal and lateral 
walls of the second (van Wijhe’s) somite, but with the qualification that 
the dorsal wall is morphologically median." 
The first (premandibular) somite shows in its development even greater 
peculiarities than those of the mandibular; yet it appears to me to pos- 
sess somatic value as unquestionably as the latter does. The first and 
most important question to answer is whether this segment represents 
dorsal mesoderm or a diverticulum from the alimentary canal, and for 
this purpose the relations of the connecting stalk furnish us with the 
decisive evidence. In a median sagittal section of an embryo with 
14 or 15 somites, such as that shown in Figure C, the tissue which is later 
differentiated as the connecting stalk of the first somite appears as a 
mass of cells between the base of the brain, in that region which lies 
just posterior to the pit of the infundibulum, and the dorsal wall of the 
alimentary canal. Posteriorly this mass of cells is continued into the 
chorda and its relations are seen to be such that, if the chorda is dorsal, 
so must the mass of cells be also. The lumen of the alimentary canal 
may be traced to a point directly ventral to the pit of the infundibulum, 
where it ends as the so called “ Seessel'sche Tasche ” (Kupffer’s “ prä- 
1 Miss Platt’s (’91*) evidence of the continuity of the cavity of the alimentary 
canal and that of the mandibular cavity, as well as her evidence of two segments 
in the latter, appears to me illusory. 
