i897. THE LEMURS AS ANCESTORS OF APES. 311 



incisors. In some of its cranial characters, Adapis more closely 

 resembles the apes than the lemurs. I refer to the nearer approach 

 of the lachrymal foramen to the orbit than in the typical lemurs ; and 

 then again the jaw symphysis is strongly anchylosed and the angle 

 rounded as in the apes. These characters, with the large canines of 

 Adapis magnus, give the skull of this species quite an ape-like 

 appearance. In general, the genus Adapis might be considered as 

 ancestral to some of the recent lemurs, as the complex character of 

 the last premolar of Adapis is found in the recent genera, Hapalemuv 

 and Galago. 



Turning to the characters of the skeleton of Tavsius, we find 

 them nearly all essentially those of the lemurs. As in the latter, the 

 fourth digit of the pes is longer than the others, and the index 

 and middle digits of the hind feet are furnished with claws. Again, 

 the calcaneal and navicular bones of the tarsus are elongated as in 

 Galago and Cheirogaleus. It is interesting to note, and this goes hand 

 in hand with the other primitive characters of Tavsius, that there is one 

 more claw on the hind foot of Tavsius than in other lemurs ; and I 

 believe that this point is an important one in its bearing on the 

 phylogeny of the Lemuroidea. 



The view has been advanced that the lemurs were descended 

 from condylarthrous ancestors. Now, by Professor Cope's latest 

 definition (3) of the Condylarthra, they were all hoofed quadrupeds, and 

 to my mind it would be impossible to derive the claws and nails of 

 the lemurs from any hoof-like type. I am convinced from a study of 

 the terminal phalanges of the lemurs, that these have been developed 

 from claws and not hoofs. Nails arise by the distal expansion of 

 claws, and this is proven in the case of the lemiirs by tracing this 

 development in the individual digits of living lemurs. 



There are a number of forms included in the Condylarthra, such 

 as Penptychus, which are very doubtful, and excluding these uncertain 

 condylarths, we are confined for our comparison to the type genus 

 Phenacodus. If we compare the structure of Phenacodus with that of 

 Adapis, the best known of the fossil lemurs, we find little in common 

 between them. Phenacodus is a typical ungulate closely related to the 

 perissodactyle division of that order. On the other hand Adapis 

 in its skeletal structure is in general different from Phenacodus, and 

 this applies as well to the details of its cranial anatomy as to the 

 structure of the limbs. If we take into consideration the difference in 

 structure between the lemurs and condylarths, I am sure that the 

 idea of deriving the former from the latter is untenable. The serial 

 arrangement of the carpus and tarsus in the condylarths and lemurs, 

 and the same number of teeth in the primitive extinct lemurs as in 

 the condylarths, can be better explained, it appears to me, as characters 

 common to the ancestors of the Mammalia in general, than evidence 

 that the lemurs and condylarths are closely related phylogenetically. 



Having attempted to show that the Condylarthra cannot be 



