OSTEOLOGY OF CARNIVOROUS DINOSAURS. 127 



Since an ungual and three separate teeth apparently constitute all of the known 

 material of this species, such observations as "its proportions were extremely 

 slender; the hones arc verj^ light and hollow, the metapodials being much elongated 

 and their walls extremely thin," are difficult of understanding. That from sucli 

 scanty remains it is possible to positively recognize the genus to which they belong 

 seems to me exceedingly doubtful. And to use a species thus founded to swell a 

 list of like genera in order to prove the synchronous age of widely separated geologi- 

 cal formations is a mistaken and misleading practice. 



I have made a thorough search of the miscellaneous materials in the National 

 Museum collections from the Lance and Morrison formations for small unguals with 

 which to compare this type, but have failed in finding any that agree with it in 

 all particulars. Recently in going over a similar collection from the Belly River 

 formation of Canada, in the American Museum of Natural History, I found at 

 least four unguals, that, except for their larger size, were exact counterparts of 

 the type imder consideration. One of these (No. 5387, A. M. N. H.) selected 

 for reproduction is shown in figure 3, plate 36. None of these American 

 Museum imguals have been identified as yet, but it appears highly significant 

 that two bones from supposedly widely separated geological horizons should so 

 closely resemble one another in every particular (see pi. 36), especially since an 

 Ornithomimid dinosaur and other dinosaurs of Upper Cretaceous affinities ap- 

 pear to be present in the Arundel fauna. 



COELURUS FRAGILIS Marsh. 



Plate 34, figs. 4, 7, and 8. 



Coelurus fragilis Marsh, Amer. Jmirn. Sri., ser. .3, vol. 18, 1879, p. 504.— Cope, Amer. Nat., vol. 

 15, 1881, p. 413 ( Amphicoelias fragillismnus , a synonym). — Marsh, Amer, Jouru. Sci., ser. 3, 

 vol. 21, 1881, pp. 330-340, pi. 10; vol. 27, 1884, p. 340, pi. 13.— Cope, Amer. Naturalist, vol. 

 21, 1887, p, 360 ( = Coelophysis).—i^EELEY, H. G., Journ. Comp. Med. and Surg., vol. 9. 1888, 

 p. 5.— ZiTTEL, K, A,, Handhuch der Palaeontologie, vol, 3, 1800, p, 731, figs. 637-340.— Marsh, 

 Sixteenth Ann. Kept. U. S. Geol. Survey, 1806, pt. 1. pi. 7, p. 1.55.— Hay, O. P., Bull. No. 

 179, U. S. Geol. Surv., 1901, p, 493.— Nopcsa, F. B., Foldtani Kozlony, Budapest, vol. 31, 

 1901, p. 202,— GiLMORE, C. W., Proc. U. S. Nat, Mus,, vol. 37, 1909, pp. 39, 40, fig. 3; Bull. 

 No, 89, U, S, Nat. Mus,, 1914, p. 4.— Mook, C. C, Ann. New York Acad, of Sci., vol. 28, 1916, 

 p. 142. 



A number of vertebrae in the United States National Museum collection are 

 identified as pertaining to Coelurvs fragilis. These are sacral vertebra No. 5809 ; 

 cervical vertebra No. 5810; caudal vertebra No. 6624; caudal vertebra No. 6625; 

 caudal vertebra No. 6626; distal caudal vertebra No. 6627; caudal vertebra No. 

 6628. These vertebrae agree closely in dimensions and other characteristics with 

 the vertebrae figured by Marsh (on pi. 10) in his Dinosaurs of North America. 



All of the United States National Museum vertebrae were collected in Quarry 9 

 (from which Morrison mammals came), Como, Albany County, Wyoming. 



The type of Coelurus fragilis Marsh is in the Yale Museum. It consists, accord- 

 ing to a letter of June 28, 1915, from Prof. R. S. Lull, of a "Vertebra, dorsal, 

 figure 3 (pi. 7, figs. 3, 3a, 36); catalogue number 1991, and is marked 'type' in 

 Marsh's handwriting. This is the only bone of the so called type specimen figured, 

 the others being plesiotypes. " 



