474 PKOP. H. G. SEELEY ON DELPHINOGNATHTJS CONOCEPHALTIS. 



owing to the concavity of the inferior border ; by this contraction 

 it is diminished to about 6 centim. behind the last tooth, and not 

 more than 5 centim. at the anterior extremity of the jaw, which 

 is manifestly imperfect. The external layer of bone is lost in the 

 alveolar region, but about half-a-dozen closely- set cylindrical teeth 

 with conical crowns are preserved. The interspaces between the 

 teeth are narrow. A slight cinguloid constriction defines the crown 

 of the tooth, which is convex, and inflated as far as it is exposed, 

 but the summit of the crown is not seen. The cylindrical roots are 

 about 5 or 6 millim. in diameter. Successional teeth rise within the 

 sockets on the inner side. On the left side the traces of the teeth 

 are less satisfactory. 



The lower jaw is composite and appears to show on the external 

 surface the dentary, coronoid, angular, surangular, and articular 

 elements, though their limits are indefinite. The jaw is flat 

 externally, like that of a porpoise, and the rami similarly converge 

 forward, but no trace of a symphysis is to be seen. There is a 

 vascular foramen above the condyle, about midway between its upper 

 margin and the summit of the coronoid process. 



In the general form of the skull there is some resemblance to an 

 Artiodactylate mammal in the large size and backward position of 

 the vertical orbit, with its posterior boundary, and in the elongated 

 form of the pre-orbital region, though the arch which supports the 

 lower jaw is unlike that in any mammal. The lower jaw can only 

 be compared to that of the Dolphin tribe in form and in dentition, 

 as preserved. These superficial resemblances are the more inte- 

 resting because the characters of the occiput, in so far as they are 

 preserved, so closely conform to the Anomodont type of the Dicy- 

 nodontia as to leave no ground for reasonable doubt that the genus 

 must be included in the Anomodont order. The characters of the 

 post-orbital region, however, are all different from those of every 

 type hitherto comprised in that order. The differences appear to be 

 more than those of a family ; they are especially the forward direc- 

 tion of the quadrate bone, the excavation above the condyle which 

 it forms, the short post-orbital region, and the depth of the zygo- 

 matic bar, relatively to its short length. The teeth are quite 

 consistent with reference to the Theriodontia, though there is 

 no evidence preserved of incisors or canines ; for I conceive that 

 Theriodonts may develop dental modifications as varied as those of 

 Mammals. j^urosaurus has the orbit large, sub-circular, placed 

 far backward, with a short post-orbital region, and the quadrate 

 bone directed downward beneath the orbit so as to form an articu- 

 lation which faces forward.^ The lower jaw in jElurosaurus simi- 

 larly increases in depth as it extends backward. Consideration of 

 these facts, which are so dissimilar to the corresponding conditions 

 in the type of the Theriodontia, Galesaurus planiceps, led me to 

 regard JElurosaurus felinus, Lycosaurus curvimola, and their allies 

 as referable to a sub-order Gennetotheria, which is nearly related 

 apparently to the Pelycosauria, and lies midway between the typical 

 ^ Owen, Quart. Journ. Geol. See. vol. xxxvii. (1881) pi. ix. fig. 1. 



