260 S. W. WILLISTON^NEW GENERA OF PERMIAN VERTEBRATES 



spinous expansions overlap like shingles the ones preceding. The roof 

 also differs in the greater angularit}^ of the ridge, the sides sloping at a 

 greater angle, while in Cacops there is a median depression or shallow 

 groove, and the lateral plates are more convex. The spines supporting 

 these plates are much more expanded above than is the case in Cacops. 

 From Broili's figure of the type of Aspidosaurvs it is not certain that the 

 otic notch is inclosed by bone, as in the Dissorophidje, though the emargi- 

 nation shown would suggest that such may be the case. 



A specimen recently described by Case"^ and referred doubtfully to 

 Zatrachys apicalis is in all probability a species of Aspidosaurus, a genus 

 apparently overlooked by the author. The author described the carapace 

 as having overlapping plates, with great angularity, apparently quite as 

 in the specimen herewith figured, and as in the type specimen figured by 

 Broili. If the specimen l)e conspecific, or at least congeneric with Aspido- 

 saurus, as I doubt not, then an additional generic character is furnished 

 by the ribs. These have an elongate hooklike process directed upward 

 and backward over the succeeding ribs, quite unlike the structure in 

 Cacops. 



Altogether the differences presented by Aspidosaurus are so fundamen- 

 tal that I believe Broili was justified in separating the genus from Dis- 

 soroplius on the character of the carapace alone. If so, then certainly 

 Cacops can not be united with Aspidosaurus, even though the size and 

 extent of the carapacial development is much more like that of this genus 

 than of Dissorophus. Possibly the Aspidosaurus type of roof has been 

 derived from that characteristic of the Dissorophidae by the anchylosis of 

 the dermal scutes with the superior expansions of the spines, but even 

 such a difference as this is of generic importance. 



Other genera of temnospondyles are known to possess similar carapa- 

 cial developments. Euchirosaurus, from the Lower Permian of France, 

 has a broad dilatation of the spines above, though not platelike. Very 

 similarly expanded spines I have seen in specimens from the Texas Per- 

 mian, but have been inclined to refer them to the basal caudal region of 

 Eryops. 



The structure of the dorsal expansion of Zatrachys is very imperfectly 

 known, unless, indeed, Case was right in referring his species apicalis to 

 that genus. The vertebrae are unknown in the type species. In another 

 species, referred to the genus by Cope later, the real Z. apicalis, he de- 

 scribed the spines as follows : "The summits of the neural spines are ex- 

 panded and the superior faces of the expansion are tubercular and have 



* Bulletin of the American Museum of Natural History, vol. xxiii, 1907, p. 665. 



