1901.] AiTINITlES or UDENODON. 185 



the humervTS is short and broad. As pointed out by Owen in 

 PJatypodosaurus, the humerus bears a most remarkable resemblance 

 to that in Echidna. There is little doubt, however, that the 

 humeri in both the Mouotremes and the Anomodonts have been 

 derived by a parallel development from the more elongated 

 Theriodont type. 



The radius and ulna closely resemble these bones in both the 

 Theriodonts and in the Mammals. The ulna is remarkable by its 

 having a very well-developed olecranon, somewhat resembling that 

 which Seeley has shown to occur in Theriodesmus, but developed 

 to a much greater extent. 



The carpus is strikingly like that of both the Theriodont and 

 the Mammal. In Theriodesmus it is moderately certain that there 

 are four bones in the proximal row as shown by von Bardeleben 

 (14), and not three as according to Seeley's restoration (15). 

 JFrom the condition found in Udenodon it will be seen that 

 Seeley's objection to placing four bones in the proximal row is of 

 no weight. Bardeleben figures two centralia, but as these two 

 are figured by Seeley as a single bone, there must still remain 

 some doubt as to whether Tlieriodesonus has one or two centralia. 

 The carpus of Theriodesmus resembles that of Udeyiodon in the 

 small size of the 2nd carpale, but differs in the apparent absence 

 of the 5th carpale. Theriodesmus further differs from Udenodon 

 in having a small praepollex. In the structure of its carpus 

 Udenodon seems to be more primitive than the Theriodont, and 

 shows affinities with a number of the earlier reptilian types. In 

 Sjphenodon the carpus is very similar, but differs in having two 

 centralia. The carpus of Lacerta as figured by Wiedersheim is 

 almost similar to that of Udenodon, only differing slightly in the 

 relative size of one or two of the elements. In Proterosaurus 

 the carpus is again formed on the same type, though apparently 

 with the absence of the pisiform ; and in the Chelonians we have 

 another slight modification of type. 



In many mammals we find the carpus vei"y closely agreeing 

 with that of Udenodon, but with the 5th carpale invariably lost. 

 It is remarkable that in the lowest mammalian orders — the 

 Monotremata, the Marsupialia, and the Edentata — the os centrale 

 is absent. It is, however, retained in many Eodents, most 

 Insectivores, the Seals, in the Hyrax, and in the Primates. Even 

 in Man in early foetal life a distinct central element can be 

 seen. 



The manus agrees with that of the mammal, not only in the 

 number of digits, but in the number of phalanges in each. 



The pelvis belongs to that type with expanded ilium and small 

 obturator foramen, which can be seen as far back as Myops 

 among the Labyrinthodonts, and which is still found in the 

 mammals of to-day. As in the Pariasaurians, the ilium lies 

 mainly in front of the acetabulum, and the ischium and pubis are 

 well developed. The obturator foramen, however, is considerably 

 larger than in these primitive forms. In Gynognathus the ilium 



