IO56 CLASS REPTILIA. 



and were replaced after the Crocodilian manner • their crowns being 

 somewhat compressed and grooved. The premaxillae appear to 

 have been small, as in the Amphibia. There are 29 vertebrae, 

 of which 18 are presacral, and two are anchylosed together to 

 form a sacrum ; while wedge-shaped intercentra are also present. 

 The neural spines are extremely short, and the centra of the dorsal 

 vertebrae are very small in proportion to their arches ; and only the 

 first sacral vertebra supports the ilia of the pelvis. Small ribs are 

 present in the caudal region ; and there was probably a dermal 

 armour. 



Professor Seeley concludes that this very remarkable and Amphibian- 

 like Reptile is a direct descendant from the Labyrinthodonts ; the chief 

 affinities to that group being displayed in the characters of the skull ; 

 in the notochordal canal, and the large arches of the vertebrae ; in the 

 support of the pelvis by a single vertebra ; as well as in the characters 

 of the pectoral and pelvic girdles. The latter features, together with 

 the general structure of the palate, being identical with those of typical 

 Anomodonts, there appears every reason for referring this family to a 

 suborder of that group. 



The genus Propappus is founded upon a humerus, from the 

 Karoo system of the Cape, of the above-mentioned type, and there 

 is no direct evidence of its distinctness from Pariasaurus. The 

 innominate bone referred to Dicynodon leoniceps probably belongs, 

 however, to Propappus, which may thus be entitled to stand. 

 The pelvis and sacrum described under the name of Dicynodon 

 tigriceps also seem to indicate a member of this suborder, since the 

 ilium is of the same type as in Pariasaurus, and is connected with 

 the sacrum by only a single rib, while there is no obturator 

 foramen. 



From the general resemblance of its skull to that of Pariasaurus, 

 we may refer to this family the genus Anthodon, of the South 

 African Karoo system, which, although originally regarded by Sir R. 

 Owen as a Dinosaur, must be included in the present order. It 

 agrees with Pariasaurus in the roofing over of the quadratic region, 

 and the continuous replacement of the teeth, which are in a uniform 

 series ; but differs in the form of the teeth, which resemble those of 

 the Dinosaurian A cant hop ho lis. 



Family Pariotichid^e. — This family, although agreeing with the 

 Pariasauridce in the sculptured cranial bones and the roofing over 

 of the quadratic region, differs in the dentition being of a carni- 

 vorous type. All the known genera are from the reputed Permian 

 of North America ; and the family is included by Professor Cope in 

 the next suborder. The three genera are Pariotichus, Ectocynodon, 

 and Pantylus. In Ectocynodon the first premaxillary tooth is tusk- 

 like, and there is also an enlarged tooth in the middle of the maxil- 



