OSTEOLOGY OF CARIv^IVOKOVS DIXOSAXJRS. 115 



the other parts of the skeleton contribute to that effect. The long tail bemg raised 

 clear of the ground and acting as a counterpoise to balance the weight and compen- 

 sate for the swaying of the body and fore legs. 



The backbone stands largely in relief to the original sandstone matrix, but the 

 greater part of the background is covered with a mixture of sand and cement that 

 when chiseled gives so close an imitation of the original sandstone as to make it 

 difficult to distinguish one from the other. 



The entire length of the animal measured from the tip of the nose to the end 

 of the tail is 17 feet 6 inches; the tail measures 9 feet 4 inches; the height of the 

 head above the base is 6 feet 10 inches; the top of the hips above the base is 5 feet 

 5 inches. 



RESTORATIONS OF CERATOSAURUS. 



Plates 27-31. 



During the years that have elapsed since Professor Marsh described Cera- 

 iosurus nasicornis in 1884, there have appeared from time to time various pictorial 

 illustrations depicting the life appearance of this animal. These are of interest 

 as showing the diversity of opinions held as to its probable life appearance. 



Arranged in chronological order in plates 27 to 30 are the more striking of 

 these restorations. The history of each, so far as it is known, and comments on 

 the more important structural differences to be observed, are given below. 



In plate 27, upper figure, is shown the first restoration of the articulated skel- 

 eton, made under Professor Marsh's direction and first published by him in 1892.' 

 It was based on the tyjDe specimen (No. 4735, U.S.N.M.) described in the preced- 

 ing pages. From the study of allied forms it appears pretty certain that the dorsal 

 region was made too long by the insertion of at least six too many vertebrae, which 

 obviously throws the animal out of balance. A preliminary sketch restoration of 

 Ceratosaurus found among some drawings made for Marsh shows fewer presacrals 

 and thus makes a much better proportioned anunal than the final draft, which has 

 been so widely copied in scientific publications of this country and abroad. It 

 would also appear that at the time of making this restoration many of the bones, 

 such as certain presacrals, pelvic bones, etc., were more completely preserved than 

 they are to-day, as may be observed by comparing this figure with the line drawing 

 of the mounted skeleton shown in plate 30. 



The first life restoration to be made of Ceratosaurus, so far as I am aware, and 

 until now unpublished, is shown in plate 27, lower figure. It was drawn by Mr. 

 Frank Bond, in 1899, under the direction of the late Prof. W. C. Knight, of the Uni- 

 versity of Wyoming. The drawing was based on the earlier skeletal restoration by 

 Marsh, and in it are incorporated the faults of proportion of that restoration. 



In 1901 Lucas ^ published the restoration shown in plate 28, upper figure, a 

 drawing by J. M. Gleeson, done under the direction of Charles R. Knight. This 

 is the best and most artistic portrayal of the probable life appearance of Cerato- 

 saurus that has yet appeared. 



I Amer. Journ. Sci., vol. 44, 1S92, pi, 7. ' Lucas, F. A. Animal.s of the Past, 1911, fig. 23. 



