108 Bulletin American Museum of Natural History. [Vol. XXVII; 



(c) What a wide difference between the many-plated molars of Hijdro- 

 chwrus and the simple molars of Scitirus; and yet there can be no doubt 

 that both are correctly referred to the same order; again the feet are more 

 alike than the cheek teeth. 



(d) The Australian Dii)rotodont iVIarsupials show a great adaptive 

 radiation in the dentition, which includes such very diverse types as the 

 specialized sectorial dentition of Thylacoleo and the prismatic rodent-like 

 dentition of Phascolomys; and yet all these Marsupials are closely related in 

 numerous weighty characters; and here again the foot-structures, while 

 differing greatly in adai)tive features, retain clear evidence of derivation 

 from a single type. 



(e) Between the two living genera of Monotremes most striking differ- 

 ences obtain in the dentition: the transitory teeth of Ornithorhynchus are 

 of the many-cusped type and all the adai)tations of the adult are for a shell- 

 crushing diet; but in Echidna, in adaptation to ant-eating habits, the teeth 

 are lacking entirely. The feet of these animals, on the contrary, while dif- 

 fering in external adaptations to aquatic and dry land habits respectively, are 

 full of the most significant, detailed evidences of close relationship (p. 154). 



Examples of this kind might be multiplied almost indefinitely, showing 

 that great differences in the dentition are quite compatible with near (sub- 

 ordinal or ordinal) relationship of the forms in question. 



(2) (_)n the other hand very many striking resemblances between the 

 dentition of dift'erent forms have frequently been brought about either by 

 parallel or by convergent evolution, in cases where the ordinal separation 

 is clearly indicated in the feet. 



(a) The cheek teeth of Hyra.r, for example, superficially resemble those 

 of Rhinoceroses, but it is practically certain (pp. 360) that these resemblances 

 are secondary; and here again the detailed structure of the feet (apart from 

 resemblances in the mere number of digits) correctly indicates the wide 

 gap between the two groups. 



(b) The cheek teeth of Meniscotherium (p. 355), an Eocene Condylarth, 

 furnish in most respects an ideal prototype for the cheek teeth of Chali- 

 cotherium (as first pointed out by Osborn, 1893, pp. 127-130); but evidence 

 (p. 399) that the ancestors of Chalicotherium were only very indirectly related 

 to Meniscotherium is again revealed in the feet. 



(6) The molar teeth of Macrotheriiim, another member of the Perisso- 

 dactyl-Chalicotheriidffi are so much like those of /Uwplotherium that 

 de Blainville (1839-1864) and Deperet (1892), apparently regarded the two 

 forms as being ordinally related and standing near the border line between 

 the Perissodactyls and Artiodactyls. But Macrotherium and its allies are 

 shown by the structure of the feet and skull to be related to the Perissodactyls 



