40 
Buccinator (Figures 8, 9, 6). A sheet-like muscle, which extends ver- 
tically from maxilla to dentary, with a possible attachment to the angle 
of the mouth. In most genera the insertion ridge on the dentary extends 
almost from coronoid to predentary. Lull (1) interpreted this to mean 
that the oral opening was restricted to the premaxillary region. In his 
recent revision (3) he has modified this interpretation somewhat. The 
gape is extended back almost to the articulation of the jaw, thus eliminat- 
ing the stiff integument from the cheek envelope, which is formed by the 
buccinator muscle. Nevertheless, the actual oral opening is still restricted 
to the premaxillary region. In the writer’s restoration it was assumed, per- 
haps incorrectly, that the integument of this region was sufficiently flexible 
to form the outer layer of the cheek. 
Parieto-mandibularis (depressor mandibuli of Lull) (Figure 8, pm). 
This corresponds, in function at least, to the digastricus of mammals. Lull 
postulated the origin from the posterior surface of the quadrate. In some 
ceratopsian skulls there is a ridge on the back of this bone, but usually it 
is quite smooth. It seems more probable that the origin was from a rather 
large, often vague area on squamosal and parietal, near the extremity of 
the par occipital process (Figure 7, pm?). Lull assigned this area to the 
latissimus dorsi. If it was the origin of the parieto-mandibularis the fibres 
would converge downward to the rather restricted insertion area at the 
posterior end of the mandible, mostly on the articular. As noted below, 
the latissimus dorsi almost certainly did not reach the skull. 
Pterygoidei. These are not clearly definable in Chas?nosaurus. The 
internal pterygoid probably was large, as it had a well-developed insertion 
along the ventral margin of the posterior half of the mandible. 
Massetericus (Figure 9, m.). This extended from the inner surface of 
the jugal to the outside of the coronoid process. It probably mingled fibres 
with the temporalis. 
Figure 1 . Protocera- 
iops andretesi Greg- 
ory and Granger, 
dorsal view of crest, 
drawn from a cast, 
with temporalis 
muscle (tm.) 
restored; XsV 
Figure 2. Chasmosaurus belli 
(Lambe), plesiotype, Geol. Surv., 
Canada, No. 2245, dorsal view of crest 
with temporalis muscle ( tm .) 
restored; X T V 
