356 
the internal surface of the orbital process toward the oral opening 
of the spiracular canal, but ends, as a distinet groove, approximately 
at interbranchial distance anterior to that opening. ‘The posterior 
margin of the groove is marked by a thickening of the lining tissues 
of the mouth cavity, and from the bottom of the groove a thick fold 
of tissue arises and projects postero-ventrally, parallel to the under- 
lying upper, but internal, surface of the palato-quadrate. This fold 
bears on its outer edge a mass of dermal papillae, these papillae 
abutting against the mandibular teeth immediately dorso-posterior 
to the line of contact of the maxillary teeth. Posterior to the man- 
dibular teeth a similar fold of tissue projects from a groove that lies, 
on either side, between the mandible and the tongue of the fish. 
The whole appearance of the groove thus here suggests a serial homo- 
logue of the branchial clefts, and if it be such a cleft the related parts 
of the palato-quadrate would seem necessarily to be parts of one or 
more premandibular visceral arches. 
In Carcharias and Torpedo conditions strictly similar to those 
in Mustelus are found, but in both these fishes the axis of the palatine 
process is inclined decidedly dorso-anteriorly. In Heptanchus the 
axis of the palatine process is inclined dorso-posteriorly, somewhat 
as it is in Chlamydoselachus, but it les near the hind end of the 
subethmoidal depression; and although there is a large pocket here in 
the roof of the mouth cavity the pocket extends but little if at all 
dorso-anterior to the dorso-posterior edge of the palatine process. 
In Cestracion there is no pocket, but there is a groove somewhat 
similar to that in Mustelus. In Chimaera neither groove nor pocket 
is found, probably because of the large dental plates in this fish. In 
Scaphirhynchus, a specimen of which I have also examined, there is 
a large pocket leading forward, above the palato-quadrate apparatus, 
from about the middle of the mouth cavity and apparently lying 
along the dorso-lateral (external) surface of the cartilage; and PARKER’S 
(1882) figures of the sturgeon would seem to indicate that a similar 
pocket is there also found. In Polyodon I can find no trace of it. 
The palatine process of Chlamydoselachus does not definitely 
articulate with the lateral edge of the solum nasi, simply resting 
against its ventral surface in certain positions of the upper jaw, 
and probably sliding slightly upon it in a latero-mesial direction. 
If this slight sliding contact were to become a definite articulation, 
such as is found in teleosts and the bony ganoids, it is evident that 
