,2 Bulletin American Museum of Natural History. [Vol. XIX, 



in Fig. 1 8, probably, but not certainly, belong to the 

 palatine. 



The hinder end of the parasphenoid is missing; but it is 

 not likely to have been much wider than the part present. 

 The species, therefore, appears to be characterized by a nar- 

 row parasphenoid. In one important respect this parasphe- 

 noid differs from that of .1. aratiis, figured in this paper. In 

 the latter species the parasphenoid extends forward nearly as 

 far as the palatines do. In A. javirostris the parasphenoidal 

 dental plate, at least, falls far short of the anterior end of the 

 palatines. I am not able to see that this is due to displace- 

 ment. 



The vertebrge (Fig. 19) are little constricted, devoid of 

 conspicuous lateral grooves, and provided with fine longi- 



tudinal ridges. 

 There appear to be 

 no lateral processes 

 such as are found 

 in some related 

 species. 



The paratype of 

 the species, No. 

 2109, consists of 

 the anterior half 

 of the skull, and 

 is of most interest 

 seen from below 

 (Fig. 20, natural 



pmx 



mx. 



Fig. 19. A KO^nius /Javi- 

 rostris Cope. No. 2 1 II. Xi. 

 Type. Three caudal ver- 

 tebrae. 



Fig. 20. Anogmius Javirostris Cope. No. 2109. Xi. 

 Paratype. den.y dentary ; itix.^ maxilla ; par., parasphe- 

 noid; pmx.y premaxillae. 



size). The specimen has evidently had pressure applied to 

 the snout, so as to drive it backward against the other bones. 

 The anterior ends of the dentaries overlap. Just behind and 



