OSTEOLOGY OF CARNrVOKOTJS DINOSAURS. 93 



distinguish detached teeth of Ceratosaurus from those of Antrodemus. So far as 

 the material at hand will admit of comparison, no essential differences were detected. 

 They have the same number of denticulations within a given space, the same irregu- 

 larity of size, same curvature of tooth, the same placing of the carina, and apparently 

 the same method of replacement of lost or worn out teeth by younger germ teeth. 



BEAIN. 



The superior and lateral views of the cast of the brain cavity of Ceratosmirus 

 nasicomis shown in figures 1 and 2, on plate 36, were made under the direction of Pro- 

 fessor Marsh, and in all probability at the time the skull was being prepared. Since the 

 investing bones of the brain case have been again securely cemented together it 

 was not deemed expedient to attempt their separation for fear of doing damage to 

 this rare specimen, so at this time there is no way of verifying their accuracy. 



The cast shows but few of the exits of the cranial nerves that lead from the brain 

 to the outer surface of the cranium, but it gives a good idea of the general shape 

 and extent of the dura mater envelope. Either the cranial openings for the exit 

 of the nerves were closed by the crushing to which the skull has been subjected or 

 else they were not properly developed at the time of clearing the matrix from the 

 brain cavity. The present condition of the skull is such that none of the nerve exits 

 can be positively recognized. 



As Marsh has pointed out, the brain in Ceratosaurus is "comparatively much 

 larger than in herbivorous dinosaurs. " When compared with the superior aspect 

 of the brain cast of Tyrannosaurus rex, there is a striking similarity between them, 

 though they differ somewhat in proportion. The protrusions observed in the 

 superior and lateral views of the Tyrannosaurus cast- are not present in the Cerato- 

 saurus brain, but again their absence may be due to lack of preparation. 



The transverse swelling of the cerebral portion of the brain is relatively greater, 

 the constriction of the olfactory region is more decided and the olfactory nerves are 

 more widely separated than in Tyrannosaurus, and the pituitary body appears more 

 robustly developed. 



VERTEBRAL COLUMN. 



Plates 19-22. 



Although parts of nearly all of the vertebrae of the backbone of Ceratosaurus 

 nasicoi'nis are preserved, there were at least three breaks in the articulated series, 

 which renders somewhat uncertain a precise determination of the vertebral formula. 



The mounted skeleton has the following vertebral formula. C, 9; D, 15; S, 5; 

 C 51 + . It was the intention at the time of mounting the skeleton to make the 

 presacral series correspond in number with a complete articulated series of Antrode- 

 mus (No. 666) in the American Museum of Natural History, New York, which has 

 twenty-three presacral vertebrae,' for there is every reason to expect the vertebral 

 formula of Antrodemus and Ceratosaurus to be the same. 



There are no true lumbars. The first restoration of the skeleton of Ceratosau- 

 rus^ shows 29 presacral vertebrae, the back being made too long by the unwarranted 



1 Regarding this specimen Dr. W. D. Matthew writes me as follows: "It is positively known (auct. Granger) in one of our 

 skeletons No. 66G from Bone Cabin Quarry, in whjch all the presacrals were in series with the skull, except the last, which was a 

 little displaced. There is some uncertainty as to the number of dorsals and cervicals, but none as to their combined number." 



» Marsh, Amer. Journ. Sci., ser. 3, vol. 44, pi. 7. 



