OSTEOLOGY OF CARNIVOROUS DINOSAURS. 



117 



The vertebra (fig. 3), [66] evidently a posterior presacral, is of peculiar proportions: nearly as deep as 

 long and strongly compressed laterally, the lower surface forming a distinct keel, which is nearly straight 

 in profile. The neural canal is broad, though somewhat constricted in the middle of tlie centrum. 



The vertebra differs materially from those of AHo.tnunisin lacking the decided constriction in the 

 middle and in the presence of the keel. It resembles most nearly the vertebra belonging to the type 

 of Creosaurus alrox (No. 1800. Yale Museum), not however, the one figured by Marsh as Creosaurus.' 



The vertebra in question is much larger than those of the type of C. atrox, but the latter show the 



same compressed form, though with a less straight ventral outline. 



The present vertebra is slightly oiusthocoelous and its dimensions are as follows: 



mm. 



Length 140. 



Depth of centrum 128. 



Width anterior face 98. 5 



Width posterior face 97. 



Least diameter of centrum 78. 



This vertebra represents by far the largest carnivore known from the Arundel formation. 



The vertebral centrum described above as "evi- 

 dently a i)osterior presacral" is most certainly a ^ -^^^^^^^^ '^^§ 

 proximal caudal, as is clearly shown by a compari- { gs^ " 

 son with the anterior caudals belonging to Antro- |^"" 

 detiius valens Leidy, No. 8367, U.S.N.M. (see pi. 

 32), and also with the caudal centra of Tyrannosaurus 

 as figured by Osborn. V^ 



The absence of a decided transverse constriction tj 

 of the middle of the centrum, the beveled chevron tt^K^^^B 

 border on the postero-ventral end, and the cancel- 

 lous internal structure of the bone, are all features 

 indicating its caudal affinities. The posterior dor.sal 



^ _ .,..,, Fig. tjb.— AntKKMR c VUI>-\L VERTEHRV OF 



centra of Antrodemus, as shown by two mdividuals dryptosaukus?potens(Lull). type. 

 now before me, have a decided transverse constric- '^°- ^''*^- u.s.n.m. j natur.il size. 



(After Lull.) 



tion at their middle, with flai'ing flange-like ends, 



and with the internal structure made up of large pneumatic cells. 



The nearly straight ventral border, as Lull has pointed out, forms "a distinct 

 keel, which is nearly straight in profile," and is one of the characters cited for dis- 

 tinguishing it from Antrodemus {Allosaurus). When compared with a proximal 

 caudal of Antrodemus (see pi. 32), it will be seen that this difference exists between 

 them, — a difference that seriously reflects on its assignm_ent to the genus Creosaurus, 

 here considered a synonym of Antrodemus, as will be discussed below. 



In Marsh's original description,- as well as in later discussions ^ of the genus, 

 but few characters were given for Creosaurus, and apparently Marsh was not able 

 to satisfactorily distinguish it from Antrodemus. Williston * in 1901 and Hay ^ 

 in 1908 both question the validity of the genus Creosaurus, though both were misled 



1 Marsh, 16th Ann. Rept. U. S. Geol. Survey, pt. 1, l.S9ii, pi. 1 ', figs. 5, fi. 

 = Marsh. O. C, Amer. Joum. Sci., vol. 15, 1S78, p. 2«. 



> Marsh, O. C, Idem, vol, 17, 1879, pp. 90, 91, pi. 10, figs. 1 and 2; vol 27, 1S.S4, pp, 3:j4, 337, pi, 9, figs. 3, 3a, and 36 and nl. 14; 

 Dinosaurs of North America, 1896, pp, 163-239, pi. 13. 



< Williston, S. W. Amer. Joum. Sci., vol. 11, 1901, pp. 111-114. 

 i Hay, O. P. Proc. U.S.Nat. Mus, vol. 35, 1908, pp. 353-356. 



144035°— 20 9 



