THECOSPONDTLTTS DAVIESI. 85 



narrowness of the sacrum in Thecospondylus Borneri, which mea- 

 sures 20 centimetres between the iliac bones, no less than the 

 elongation of the ilium over six or seven vertebrae, and the general 

 build of the animal, lead to the conclusion that its habits were 

 active, its limbs long, and its extremities specialized. 



In endeavouring to form an opinion on the systematic position of 

 Thecospondylus, there is little evidence available beyond these 

 British specimens. Por although Professor Marsh * states that he 

 possesses portions of ten or twelve individuals of Coelurus fragilis 

 from the middle Secondary rocks of Wyoming, which yielded the 

 Atlantosaurs, little is known of their structure beyond what his 

 figures demonstrate. He places the animal in a new subordinal 

 group, and observes (p. 340) : — " its remains preserved suggest re- 

 semblance to Dinosaurs, to Pterodactyles, and more remotely to 

 Birds, and it is apparently a generalized Sauropsid which, when 

 fully investigated, may serve to bridge over some of the present 

 breaks in the lines of descent. The sum of its known characters 

 indicates that it is a reptile and not a bird. Its structure, so far as 

 known, presents more similarity to Dinosaurs than to Pterodactyles ; 

 but for its nearer affinities we must await the discovery of further 

 remains." I concur in the general spirit of these conclusions, but 

 they admit of some elaboration. 



When the first notices of Ornitliojpsis were published, Dinosaurs 

 with a pneumatic skeleton were unknown ; and it was remarked of 

 the type specimens (Annals and Mag. IsTat. Hist., April 1870) : — - 

 " both vertebrse agree in being constructed after the lightest and 

 airiest plan, such as is only seen in Pterodactyles and Birds ; and 

 they agree in possessing pneumatic foramina, which are an Avian 

 and Ornithosaurian peculiarity. The foramina are of enormous 

 size, and approximate to those of the Pterodactyles rather than to 

 those of birds." On its afiinities it was observed "it does not con- 

 form closely in the shape of vertebrae to either Pterodactyles or 

 birds. And from the bones preserved, and many indications of 

 allied animals which I have seen from the Wealden and Potton Sands, 

 I anticipate that it will form the type of a new Order of animals, 

 which will bridge over something of the interval between birds and 

 Pterodactyles, and probably manifest some affinity with the Dino- 

 saurs." 



At that time, Hypsilophodon, Iguanodon, Sadrosaurus, and 

 Scelidosaurus gave the best conception of the Dinosauria. It has 

 been the fortune of others to elaborate the group thus indicated. 

 Mr. Hulke has admirably described Ornitlwpsis, and Professors Cope, 

 Huxley, and Marsh have investigated the structure of its allies, and 

 their classification. But it has seemed to me that the classifi- 

 cational value of the pneumatic skeleton has been underestimated 

 in the results which have been formulated. Professor Marsh has 

 been led to form an order of animals for the type to which Ornitliopsis 

 belongs. This order was named Sauropodaf, and has for its European 

 types Ceiiosaurus and Ornitliopsis. 



* Amer. Journ. Sci. rol. xxi, (April 1881). 



t Ibid. vol. xvi. p. 412 (Nov. 1878), vol. xxiii. p. 83 (Jan. 1882). 



