6-M PROF. H. G. SEELEY ON [Aug. I9OO, 



very small in Pareiasaurus. It appears to be homologous with the 

 Ichthyosaurian foramen. And when the vacuities in the back of 

 the skull are closed as in some Dicynodonts, the quadrate foramen 

 is comparable in position to this foramen in Omithorhynchus ; and 

 to the similarly-placed foramen in Ratteria, which opens into the 

 postorbital vacuity, and is defined externally by the quadratojugal 

 bone. It is not proved whether the divisions which appear to 

 separate this external film in Omithorhynchus into separate bones 

 may not be tension-fractures due to maceration. But since they 

 are in the place of the quadratojugal bone, there is a possibility 

 that the quadratojugal loses its individuality in the squamosal, 

 which may require examination. If that inference is suggested 

 from the persistence of the foramen, then it would seem worth 

 examination whether the articular area for the mandible repre- 

 sents the quadrate bone, which would also become lost as a portion 

 of the squamosal bone. Owen considered that the presence of a 

 distinct tympanic bone (with which he identified the quadrate bone 

 in the usual position in Ornitliorhynchus) ' nullifies the supposition 

 that the upper root of the zygoma can be the analogue of the 

 os quadratum in the Ovipara.' 1 (See fig. 8, p. 643.) 



The teeth of Aristodesmus are not only distinct from those of 

 Omithorhynchus, but appear to be different from those in Anomo- 

 donts, resembling a type like Rhopalodon in the individual form of 

 the molars, and a type like Procolophon in the absence of distinction 

 between teeth in different parts of the series. This resemblance 

 has led to a detailed examination of the evidences of structure in 

 ProcolojJion, which justify me in regarding that type as making (on 

 the whole) a nearer approximation to Aristodesmus than any other, 

 notwithstanding considerable differences in the form of the ilium 

 and the proximal end of the femur. It has also shown that the 

 original proposal to place the Procolophonia as a primary con- 

 stituent group of the Anomodontia, can be better sustained than the 

 subsequent proposal to make it a subordinate group of the Pareia- 

 sauria. The back of the skull is devoid of lateral perforations. 

 Aristodesmus resembles Procolophon in plan of skull, so far as the 

 parts can be compared ; and in plan of shoulder-girdle in having 

 the precoracoid extended in front of the scapula, by which they 

 differ from all other Anomodonts. 



The pubis and ischium appear to be of the same type in both ; 

 though in the limb-bones there are many differences in detail 

 in structures that can be compared. 



I propose to place the fossil in association with Procolophon as a 

 separate family in the tribe Procolophonia. 



The presence of Aristodesmus in the Bunter Sandstone is an 

 additional organic link between the Trias and the Permian strata in 

 which other remains of Anomodont reptiles have been found in 

 Scotland, Russia, and France. 



I am indebted to Prof. Charles Stewart, P. U.S., for enabling me to 



1 Since this was written Dr. J. F. van Bemmelen lias described these struc- 

 tures in the Monotremata ; see Koninklijke Akademie van Wetenschappen te 

 Amsterdam, 1899, Zoology, p. 81. 



