632 PKOF. H. G. SEELE5T ON [Aug. I9OO, 



The distal end of the bone has a sort of hammer-headed 

 appearance. The superior surface is smooth, inflated proximally 

 on the ulnar side by a broad convex ridge, which prolongs the 

 convexity of the shaft obliquely across the expansion of the distal 

 end of the bone towards the ulnar side, exactly as in Dicynodon. 

 The ridge makes the middle part of the distal end a shallow con- 

 cavity, with the middle of the distal margin notched ; but the 

 concavity is less marked than in Monotremes. There is a small 

 flattened triangular articular surface for the ulna ; and there appears 

 to be a second long and narrow articular surface for the radius. 



The essential differences from Echidna are in the absence of 

 the strong lateral curvature, which causes the twisted proximal 

 end of the bone in Monotremes to curve laterally towards the distal 

 end. This divergence from the Monotreme type is emphasized by 

 three other differences. Firstly, absence of a vascular perforation 

 through the distal end of the bone in the fossil ; secondly, limita- 

 tion of the two facets for the distal articulation to the inferior 

 surface of the bone ; and thirdly, the lateral border is a simple open 

 curve from the proximal end to the distal expansion, unbroken by 

 processes. 



Comparisons with the humerus in burrowing animals, such as the 

 mole, suggest that the transverse expansions of the extremities of 

 the bone, the curve of the proximal articulation downward, and the 

 lateral position of the distal articulation, may all be adaptive modifi- 

 cations ; and that the difference between the Insectivore and the 

 Monotreme in structure of the humerus is not greater than, even 

 if so important as, the difference between the Monotreme and this 

 fossil. 



The value of the resemblances to mammalia as marks of affinity 

 can only be determined by the extent to which the characters are 

 shared by animals which there is no ground for regarding as 

 mammalian. Thus the transverse expansion of the proximal and 

 distal ends of the bones, the constriction of the middle waist of the 

 humerus in the fossil, the downward direction of the radial crest, 

 the unbroken contour of the ulnar border, the absence of a foramen 

 opening on to the superior surface on the ulnar side, as well as the 

 long narrow articular surface at the proximal end, are characters 

 found in the humerus of Protorosaurus, though they are not developed 

 in the same degree ; and therefore in all those points in which differ- 

 ence from the Monotreme is most pronounced, there is a reptilian 

 approximation. 



Most of the characters seen in the Permian reptile Protorosaurus 

 are also found in the different animal types which have been referred 

 to the Anomodontia. Sir Kichard Owen figured four in the 

 humerus of Dicynodon — the transverse expansion of the proximal 

 and distal ends, the constriction of the waist of the bone, the smooth 

 contour of the ulnar border, and the downward reflexion of the 

 radial crest at the proximal end. In details there are striking 

 differences, for all these points are intensified in Aristodesmus. On 

 the other hand, in Dicynodon the articulation of the bone makes a 



