OSTEOLOGY OF CARNIVOROUS DINOSAURS. 85 



Antrodemus sends forward and inward a thin vertical sheet-like process to meet the pterygoid, and 

 there is every reason to believe that the same condition prevails in the present specimen. The squamous 

 union of these two bones, although somewhat obscure can be traced in the specimen by an irregular 

 vertical line that passes in front of numeral 28 in figure 54. So that this numeral, as the relation of these 

 bones are now interpretated, does not represent the pterygoid but is the inner and forwardly directed 

 process from the quadrate.] Whether or not the pterygoid articulated with the quadrate the Whole 

 distance between the two points mentioned is somewhat uncertain. On the left side of the skull a 

 considerable part of the pterygoid from the quadrate forward is broken away. [The missing bone is 

 largely from the quadrate.] Marsh neither mentions any vacuity here nor does he represent it in bis 

 figure. On the right side there is a gap between the two bones, extending about 55 mm. along the 

 border of the quadrate and forward from it less than 15 mm. It seems probable that in life this gap was 

 filled in with bone. [It is in Antrodemus.] This region is represented in the figure as it appears on the 

 right side. In Sphenodon the quadrate sends forward a broad process between the upper and lower 

 processes of the pterygoid. It is quite certain that this was not the case in Ceratosaurus. [In Antro- 

 demus the union of the pterygoid with the quadrate is almost identical with that in Sphenodon and 

 there is no reason to expect it to be greatly different in Ceratosaurus.] 



Marsh informs us that there is a very short, thin epipterygoid closely articulated by suture with the 

 pterygoid. The position of the suture is obscure, but it was doubtless present. Marsh was probably 

 mistaken when he affirmed that the upper end of the epipterygoid articulated with the postfrontal. Its 

 connection must have been with the border of the parietal. The epipterygoid of the left side has its 

 upper end missing [fig. 54, numeral 8]. On the right side this bone is uninjured. Its upper end is 

 moved well outward toward the postfrontal, but it is probably joined to the parietal. The suture be- 

 tween the latter bone and the postfrontal is obscure. Just behind the base of the epipterygoid the 

 width of the pterygoid is 64 mm. 



The inner and anterior end of the transverse bone [fig. 54, numeral 36] overlaps the outer border of 

 the pterygoid, as if it had been pressed somewhat out of its natural relations; but in both Sphenodon and 

 Cyclura the transverse bone overlaps the pterygoid. From the outer border of the pterygoid and of the 

 transverse bones a strong process, 29, descends to a point a little below the upper border of the lower jaw. 

 Marsh indicates that this process belonged wholly to the transverse, but the suture between the two 

 bones is distinct. The transverse measures 160 mm. from its inner end to the extremity of the process 

 mentioned. Just in front of the inner end of the transverse the pterygoid is 45 mm. wide. 



The sutures separating the pterygoid, the palatine, and the vomer are not as distinct as could be 

 desired. Nevertheless their courses may be made out more or less satisfactorily. The bones of the right 

 and left sides did not articulate on the midline. Doubtless there was, at least posteriorly, a considerable 

 interval between them. In the skull, as preserved, the rounded inner, or upper, border of the bones 

 of the right side rise several millimeters above those of the left side. 



Mostly hidden by the lachrymal and the jugal, in a side view of the skull, is a considerable vacuity 

 in the bones of the roof of the mouth. Its fore and aft extent is 90 mm. This is called by Lambe the 

 suborbital vacuity. It seems to correspond to the postpalatine foramen of lizards, turtles, and spheno- 

 don. It is bounded behind by the transverse and mesially, in part, by the pterygoid. In figure 54 the 

 vacuity is mostly hidden by the lachrymal. A part of it, with light shading, appears between the two 

 numerals 20; the hinder part is shaded dark and is in front of numeral 36. 



Another fissure in the roof of the mouth is seen through the antorbital vacuity and is situated just 

 above the border of the maxilla. This is the internal nasal passage [fig. 54, numeral 16]. It extends 

 forward to the anterior end of the antorbital vacuity and backward to a bar of bone that bounds the 

 postpalatine vacuity in front. The whole extent of the internal nasal passage, as seen, is 85 mm. Pos- 

 sibly it extended forward some distance beyond the front of the antorbital vacuity. 



The bone that lies between the postpalatine vacuity and the nasal passage is undoubtedly the 

 palatine [fig. 54, numeral 20]. A branch of this bone rims backward against the inner face of the max- 

 illa, its hinder extremity being hidden by the maxilla, the front end of the jugal, and the base of the 

 lachrymal. The hinder border of the bone sweeps around and bounds in part the postpalatine vacuity 

 mesially. The suture between the palatine and the pterygoid appears to be where drawn in figure 

 54, B, running first directly forward from the inner side of the postpalatine vacuity to a distinct ridge, 

 then following upward and inward this ridge about 25 mm. Behind this ridge the fibers of the bone are 

 144035°— 20 7 



