OSTEOLOGY OF AMERICAN PERMIAN VERTEBRATES 367 



further excavation of the original deposit could be made. Unfortu- 

 nately, explorations the past year have brought to light only a few 

 more bones, including, however, the skull which has served as the 

 key to the solution of its structure. This delay is the more to be 

 regretted as it has caused Dr. Broom to misinterpret the genus and 

 redescribe it under another name.^ 



SKULL 



The material studied consists of seven skulls, none complete. 

 They are all more or less distorted, with the different elements 

 more or less separated, and in some misplaced and broken. They 

 also differ in size; two preserved in blocks of matrix in association 

 with cervical vertebrae are of adults ; the loose ones are all more or 

 less immature, though none is of a quite young animal. 



The first specimen studied, at the time I gave the preliminary 

 description of the genus, had the right temporal region exposed, 

 showing conspicuously what seemed to be the smooth border of an 

 entirely open temporal region. Within the depression, however, 

 there is a large flat bone, and the fragment of another, which seemed 

 to be misplaced elements that had been crowded into the cavity. 

 On the opposite side the same free border was visible with the 

 inclosed space filled more or less by matrix. It was because of 

 this apparent structure that I stated in my preliminary description 

 of the genus that "almost certainly there is a large temporal 

 vacuity" and that "nothing definite can be said about Araeoscelis 

 till the skulls have been cleaned and studied, and possibly not even 

 then, save the presence of a temporal vacuity." 



Unfortunately for my first belief, when the matrix was removed 

 from the left temporal region of the above-mentioned skull it too 

 was found to have a broad expanse of bone at its bottom, filling out 

 nearly the whole space, definitely proving that the bone of the 

 opposite side belonged where it was found. Nevertheless, distinct 

 evidence of a free border above, and the whole structure of the 

 skeleton so definitely uncotylosaurian, made it difficult for me to 



' Araeoscelis gracilis Williston, Jour. GeoL, XVIII (1910), 518; American Per- 

 mian Vertebrates (1911), p. 6; Jour. Geol. (1913), p. 743; Science, November 9, 1913; 

 Huene, Morph. Jahrb., 1912 (not Araeoscelis Schultz, 1911); Ophiodeirus casei Broom, 

 Bull. Amer. Mus. Nat. Hist. (1913). 



