1910.J The ''Pre-" ami "Pod"- Axis; PelvLs and Seapula. 435 



ward rotation of the scapula, (2) the drawing in of the elbows, (3) the move- 

 ment of the glenoid fossa away from the sternum. This caused the humerus 

 to assume a more erect position and to face forward rather than inward. 



Flower's view (op. cit., p. 260-270) that the permanent pronation of the 

 manus in quadrupedal mammals is altogether a "very modified and adaptive 

 position" appears very improbable. The manus was very likely turned 

 forivard as well as outward to a considerable degree even in Pareiasaiiru.s' 

 (Broom, 1903.4) and certainly is so held in Orniihorhynchn.'i. In fact, the 

 pose of the manus may have remained relatively fixed as compared with 

 the rapidly changing pose of the elbow, humerus and scapula. In the 

 standing pose of the limb in Monotremes the ulna is largely exiernal to the 

 radius with respect to the body, but po.sierior to the radius with respect to 

 what was later the front of the humerus. When the elbow was drawn 

 inward the olecranon would become posterior to the radius with respect to 

 the long axis of the body and would also retain its posterior relation to the 

 radius with respect to the "front" face of the humerus. But the lower 

 end of the ulna, being attached to the external part of the carpus in the 

 relatively fixed pose of the manus, would always remain external to the 

 radius. This, and not the pronation of the manus, may be the j^rincipal 

 cause of the slight crossing of the long axes of the radius and ulna which is 

 observed even in very ])rimitive mammals. The extreme crossing of these 

 bones in Man and the Elephant is a secondary condition. 



The Pelvis and Scapula, 



Flower also assumed the more or less complete homodynamy and serial 

 homology of the parts of the scapula and pelvis and of the fore and hind 

 limbs; his diagrams (op. cit., p. 364) are in fact suggestive of Owen's "arche- 

 typal" homologies. But even in the Paleozoic Crossopterygian fishes we 

 do not find any such correspondence. In the Permian Stegocephalia, Cotylo- 

 sauria, I'herapsida and Pelycosauria the scapula and pelvis were radically 

 different and lend no support to the proposition that in every mammal 

 "both scapida and ilium may be resolved into rods or bars of three sided 

 prismatic form" (Flower, p. 309). The primitive form of the sca])ula has 

 already been discussed (p. 432). If the mammals are derived from un- 

 known Cynodonts, then the trihedral ilium of Didelphis must be derived 

 from the flattened triangular ilium of the Diademodon type by the atrophy 

 of the posterior angle and hypertrophy of the anterior border (Broom, 

 1905.4). Remains of this posterior border are preserved in Oryc.teropus 

 (Broom). The ilium in very small primitive insectivorous forms like 

 Marmosa, Microgale, and Tiipaia is not trihedral but varies from a small 



