Mor\TF.n srr.ciMi'.xs 65 



an animal as Centrostmrus. They were made by a digitigrade form and exhibit the correct number 

 of digits. The rear pad-Jike part of the foot, however, did not make a full impression, which is 

 understandable, as the toes would dig into the sand somewhat more deeply at the end of the step 

 when the heel of the font wis lilted from the ground. At the same time the fore foot, which may 

 have toed in at first, would be brought into line with the hind foot. The animal is therefore posed 

 with an easy stride, with the elbows flexed and bowed outward, and the humerus nearly horizontal. 

 The hind limb, on the other hand, was nearly straight, with a slight flexing of the knee at the 

 beginning of the stride, before the weight was fully borne. All four feet are on the ground although 

 the several feet probably struck the ground at slightly different intervals. The normal stride must 

 have been slow, the body swaying in front to bring the weight toward the hand that bore it when the 

 other one was lifted. One imagines the foot movement to have been somewhat shuffling, as with a 

 modern elephant, the feet never rising far above the ground. In charging, the movement must have 

 been more rapid, but only for a short distance. 



The head was probably carried low, with the muzzle near the ground. This was determined in 

 part by the articulation of the cervicals, which do not indicate a great deal of intervertebral play, and 

 in part by the direction of the pedunculate occipital condyle, the axis of which must have been 

 carried normally in line with that of the three coalesced anterior cervicals. This is also in harmony 

 with the findings of Tait and Brown based upon the position of the horizontal semicircular canal 

 in Anchtceratops. 41 A further discussion of this problem will be found under the section on the 

 Brain and Special Sense Organs, p. 73. The jaws in our specimen have been set somewhat agape. 



On the right side the flesh (PI. II, B) of the animal has been restored. The original model- 

 ing was done synthetically, the muscles, especially those of the limbs, being built on from within 

 outward although, as the completed restoration is in the form of a shell, only such as determine 

 the outer contour of the muscular system are preserved. In accordance with my former practice, 

 not only is the entire right side of the head, body, and tail covered by flesh but that of both right and 

 left limbs as well, in so far as the latter may be seen from the right aspect. Thus, when viewed 

 squarely from the right, one gets a complete silhouette of flesh, no bone being visible, while from 

 the left the entire skeleton may be seen, including the inside of the right limbs and the outside of 

 the left. 4 "' 



In restoring the musculature, reference was made to the very careful studies on "The Pelvic 

 Musculature of Ornithischian Dinosaurs" by Alfred S. Romer, 4 " as well as to papers on the myology 

 of the chameleon and the iguana by St. George Mivart, 44 together with some dissection of a young 

 alligator. 



The processes at the rear of the crest in this genus were a problem. Brown, in his published 

 drawing, 4 '' encloses them in the general integument of the crest, which would require a very con- 

 siderable thickness of subcutaneous tissue, muscular or otherwise, and which is out of harmony with 

 the bloodvessel impressions on the surface of the crest, since the latter imply a close-fitting, resistant 

 investiture. We have therefore deemed it wise to show these as processes sheathed like the horns, 

 without trying to solve the question of their possible utility. Of course, the horny covering must 

 have overlain the parietal fenestrae, the presence of which is indicated by a slightly depressed area. 



In restoring the mouth, we compared the beak and bones of a loggerhead turtle with the rostral 

 and predentary of Centrosaurus. With the remainder of the mouth we have varied somewhat from 

 my previous ceratopsian restorations, in which the gape was limited to the prehensile part only, the 

 masticating portion being covered by integument to retain the food in the mouth. But the integu- 

 ment of the face was horny and resistant, and would be either creased and folded over opposite the 

 tooth row when the mouth was closed or, as in the Crocodilia, the gape would apparently extend 

 to the articulation of the jaws. The crocodiles show, when the mouth is open, certain muscular soft 



41 Tait, J., and Brown, B., 1928. 



4J Note. The photograph of the right or flesh side of the animal has been reversed in the plate, for ease of comparison. 



43 Romer, A. S., 1927. 



44 Mivart, St. G., 1867, 1870. 



4 'Brown, B., 1917, PI. XIV. 



