6 BULLETIN 110, UNITED STATES NATIONAL MUSEUM. 



Nat. Mus., vol. 35, 1908, pp. 353-356.— Hutchinson, H. N., Extinct Monsters and Creatures of 

 other days, 1911, new ed., p. 140, pi. 15, fig. 44.— Jaekal, Otto, Die Wirbeltiere, 1911, p. 160, fig. 

 176, (as Allosaurus agilis). — Zittel, K. A., Grundzuge der Palaeontologie, vol. 2, 1911, pp. 282, fig. 

 427, (as Allosaurus agilis).— Osborn, H. F., Memoirs Amer. Mus. Nat. History, vol. 1, pt. 1, 1912, 

 text figs. 9, 10, 11, 26, and 27 (as Allosaurus agilis).— Gregory, W. K., Journ., Morph. vol. 24, 1913, 

 p. 11, fig. 8 (as Allosaurus agilis).— Gilmore, C. W., Bull. 89, U. S. Nat. Mus., 1914, p. 4; Proc. U. 

 S. Nat. Mus. vol. 49, 1915, pp. 501-513, text figs. 1-7.— Mook, C. C, Annals New York Acad. Sci., 

 vol. 27, 1916, p. 141.— Lambe, L. M., Memoir 100, Geol. Surv. Canada, 1917, p. 56, figs. 32, 37. 

 Allosaurus King, C, U. S. Geol. Explor. 40th Par., 1878, p. 346.— Cope, E. D., Amer. Nat., 1885, Jan. 

 p. 67.— Osborn, H. F., Bull. Amer. Mus. Nat. Hist., vol. 22, 1906, pp. 283-294, figs. 2 and 6 A.— 

 Matthew, W. D., Journ. Amer. Mus. Nat. Hist., vol. 8, 1908, pp. 3-5, pi. 1; Dinosaurs, Handbook 

 Series, No. 5, Amer. Mus. Nat. Hist., 1915, pp. 36-46, figs. 8, 9, 10, 11, 12. 



Genotype. — Cat. No. 218, U.S.N.M. Consists of the posterior half of a caudal 

 centrum (probably the sixth) from the anterior fourth of the tail. Collected by 

 Dr. F. V. Hayden, 1869. 



Type-locality. — Middle Park, Grand County, Colorado. 



Horizon. — Morrison, Jurassic. 



The original description is as follows : 



A specimen, consisting of less than the half of a vertebral body, was submitted to my examination 

 by Professor Hayden, who obtained it last summer during his geological survey. It is from Middle Park, 

 Colorado, and Professor Hayden thinks was derived from a cretaceous formation. Similar specimens 

 were reported to be not unfrequent, and were known under the appellative of "petrified horse hoofs." 

 The fossil indicates an elongated form of caudal vertebra of some large saurian. Much constricted 

 toward the middle, such specimens would be most liable to break in this position, and the halves from 

 their form might readily be taken, by the inexperienced in such matters, for what they are called. 



The vertebral body in its entire condition would resemble in form those of Megalosaurus, but in 

 form and other characters bears a near resemblance to those of Poitilopleuron Buchlandi. This is an 

 extinct reptile from the oolitic formation of Caen, in Normandy, described by Deslonchamps; and 

 remains apparently of the same animal from the Wealden of Tilgate, England, have also been described 

 by Professor Owen: 



Poicilopeluron has generally been viewed as a crocodilian reptile with biconcave vertebrae, but 

 probably pertains to the dinosaurs. The P. Buchlandi is estimated by Deslonchamps to have been 

 about 25 feet long. The Colorado fossil indicates a much larger animal, having been more than one- 

 third greater. 



One of the most remarkable characters of the Poicilopleuron is the presence of a large medullary 

 cavity within the bodies of the vertebrae, parallelled among living animals, so far as I know, only in 

 the caudal vertebrae of the ox. The same character is presented by the Colorado fossil. In the former 

 animal the cavity appears simple or unobstructed by osseous trabeculae. In the Colorado fossil, as 

 seen in the broken surface of the specimen, the medullary cavity occupies the lower two-thirds of the 

 interior of the body and is crossed by a few trabeculae. The sides of the cavity converging below are 

 constituted by a layer two lines thick and as compact as the walls of the medullary cavity in the limb 

 bones of most ordinary mammals. The upper third of the interior of the body is occupied by the ordi- 

 nary spongy substance which becomes more compacted, ascending into the interior of the neural arch. 

 The cavernous structure of the Colorado fossil is occupied with crystalline calcite. 



The estimated length of the vertebral body is 6 inches or more [estimate too great by at least an 

 inch]. The sides are much narrowed toward the middle, and they are concavely depressed just below 

 the sutural conjunction of the neural arch. A narrow groove occupies the lower border of the body, as 

 is indicated to be the case in the Tilgate specimens. The posterior articular surface of the body is mod- 

 erately depressed, but its lower fourth curves forward, producing a thick, convex ledge for the accom- 

 modation of a chevron. The breadth of the articular surface is scarcely 4 inches and its depth is about 

 the same measurement. 



The species represented by the fossil may be named Poicilopleuron valens. Should the division 

 of the medullary cavity of the vertebral body into smaller recesses by trabeculae be significant of other 



