THE TRIAS 'OF GRAAFP EEINET, S. AFRICA. 423 



longitudinal extent and well -defined ridge-like form of the great 

 trochanter, in which, as in the proportions of breadth and fore- 

 and-aft compression of the proximal (probably) half of the bone, I 

 find the closest agreement in the femur (PL XVII. fig. 9) of the 

 Spiny Monotreme (Echidna liystrix). 



From the number of correspondences presented by parts of the 

 skeleton of Platy][jodosaurus with homologous ones in the skele- 

 tons of the two existing genera of Monotrematous Mammals, 

 one is led to relieve the dry work of comparison by speculations on. 

 the vast number and variety in gradually advancing structure of 

 air-breathing vertebrates, offsprings of the strange reptilian forms 

 exemplified by their remains in Cape localities — but which, of old, 

 may have spread over lands extending thence northward and east- 

 ward, now in great part submerged, and of whose inhabitants a 

 remuant still survives and lurks in the burrows and waters of 

 Australia. One may also conjecture, on the derivative hypothesis, 

 that the higher class of Vertebrates, as represented by the low 

 ovoviviparous group now limited to Australasia, may have branched 

 off from a family of Triassic Eeptilia represented, and at present 

 known only, by the fragmentary evidences of such extinct kinds as 

 that which forms the subject of the present communication. Eut 

 this is far from being the only instance of correspondences between 

 organisms, both animal and vegetable, of the Cape of Good Hope 

 and those of New Guinea, Australia, and Tasmania. 



The European Triassic reptile Placodus, nearest akin to the Cape 

 EndotJiiodon, has the molar dentition of each mandibular ramus 

 represented in one species by two low, broad, subquadrate plates 

 which overhang the border of the bone (PL XVII. fig. 11). In the 

 young OrnithorliyncJius the molar dentition of each mandibular 

 ramus is represented by two similar plates, which in the adult be- 

 come confluent, also overhanging the supporting bone, and still indica- 

 ting their primitive separation (ib. fig. 10). True it is they are not 

 calcified in the duck-mole ; but, in other more significant characters, 

 this mammal alone repeats the Placodont ones. I may add that the 

 rami of the mandible in both divaricate anterior to the symphysis, 

 as in figs. 10 k 11, PL XVII. 



The fossils of Platypodosaurus which have afforded the foregoing 

 descriptions were obtained by E. J. Dunn, Esq., of Oaklands, Clare- 

 mont, Cape of Good Hope, from a quartzose Karoo formation 350O 

 feet above the sea-level, near Graaff' Eeinet, and were transmitted as 

 a donation to the British Museum. 



Postscript (British Museum, 22nd June, 1880). — I have received 

 this day a copy of a " Second Contribution to the History of the 

 Vertebrata of the Permian Formations of Texas," by E. D, Cope 

 (read before the American Philosophical Society, May 7th, 1880). 

 In this paper the Author states: — "The order Theromorplia 

 approximates the Mammalia more closely than any other division of 

 Reptilia. This approximation is seen in the scapular arch and 

 humerus, which nearly resemble those of the Monotremata^ espe- 

 <5ially Echidna, and in the pelvic arch, which Owen has shown, in 



