gidl^y: primitive mammalian foot 275 



phalanges, instead of distal as in the remaining metapodials ; 

 (c) the anomalous musculature of digit i on all mammals, the 

 object of which is clearly seen when the digit is opposable, but 

 is quite unexplained otherwise." In further substantiation of 

 the arboreal ancestry hypotheses, Matthew's first proposition 

 above quoted has, I believe, little weight as evidence. First, 

 because Notharctus can hardly be considered primitive, since 

 the known species, even in the Bridger epoch, had advanced in 

 foot, limb, and skull structure well toward the condition found 

 in the present-day South American apes; second, admitting NoiJi- 

 arctus to be primitive, the presence of true opposability in this 

 genus can aflfect the proposition under discussion only so far 

 as the Primates themselves are concerned, and cannot be taken 

 as evidence of arboreal ancestry for other orders of mammals, 

 especially as there is no indication of close relationship between 

 any of them and the Primates.' At most the evidence in the case 

 of Notharctus can not be interpreted to mean more than that the 

 group which it represents (the Primates) had adapted themselves 

 to an arboreal habitat at a comparatively early period. But 

 opposability probably followed or accompanied and did not 

 precede their adaptation to tree-living habits. If this be true 

 it explains why there are so many cases in which opposability 

 has not been developed even in strictly arboreal forms, and like- 

 wise why these cases seem to be confined to species of compara- 

 tively small size. For example, many species of rodents and 

 insectivores are living to-day almost exclusively in the trees, }'et, 

 so far as I am aware, there is not one example of true opposability 

 among the Insectivora, and but one among all the Rodentia, 

 living or extinct. This single exception is furnished by the 

 African genus, Lophiomys, in which the first digit of the hind 

 foot is thus developed, and this animal is not arboreal. Among 

 the Insectivora some of the Tupaiidae are mostly arboreal in 

 habit, more so than in any other members of the order, yet even 

 in this family there is not the slightest trace of opposability. 



^ It may be here remarked that, in my opinion, there is some evidence that this 

 important order, and probably some other orders of mammals as well, have been 

 derived independently from different though probably more or less closely related 

 premammalian ancestral groups. 



