204 



GREGORY: NOTHARCTUS, AN AMERICAN EOCENE PRIMATE 



developed. Thus in the Chirogaleinae ps is very simple, almost peg-like; in Lemur and Lepilemur it is 

 compressed and pointed. 



in the Notharctinse is distinguished by the very delayed development of the posteroexternal cusp 

 or metacone, which in the history of the typical Adapinse appears very early. In Adapts sciureus, how- 

 ever, the very retarded condition of the hypoconid of p4 makes it extremely probable that in p"* the second 

 external cusp (metacone) if at all differentiated was connate with the paracone. In the Lemuridse this 

 tooth has usually only one external cusp (exclusive of the parastyle). In Myoxicehus (Hapalemur), 

 however, there is a large distinct metacone. The presence of a second external cusp in p^ of Myoxicehus 

 thus raises the question whether the single external cusp in the other Lemuridse is a primitive or a second- 

 ary character. The extremely simple form in Microcehus is accompanied with other specialized characters 

 and is no doubt secondary. In view of the evidence tending to show that p4 also is partly degenerate in 

 many Lemuridae it seems probable that Myoxicehus (Hapalemur) alone among the Lemuridse has retained 

 the primitive posteroexternal cusp of p*. 



The internal or "protocone" extension of p^ is also highly variable, being very wide and massive in 

 Myoxicehus, almost absent in Microcehus, well developed in Lemur, Lepilemur and Mixocehus. Here 

 again the conditions in Myoxicehus suggest those of Adapis magnus. 



P4 of the typical Lemuridse as compared with those of the Notharctinse and Adapinse gives the im- 

 pression of being variously degenerate; the talonid is widely unlike that of the molars, whereas in all 

 primitive Eocene mammals, including the Notharctinse and Adapinse, the talonid of p4 shows more or 

 less distinct beginnings of a posterior V bearing an incipient hypoconid. In Myoxicehus, which as above 

 noted has a less degenerate p^ than Lemurs, p4 has also, as might be expected, a well-developed talonid 

 with a distinct hypoconid. In Lepilemur and still more in Lemur, on the contrary, the talonid is much 

 reduced; Lepilemur, however, has the trigonid of p4 in an unreduced condition, with a very distinct 

 metaconid, the whole tooth being fundamentally identical with that of Adapis save for the reduction of 

 the talonid. In other words, the simplification of the first three premolars, which is very marked in 

 Adapis parisiensis, may have begun to affect the fourth premolars in the typical Lemuridse more than 

 in Myoxicehus. 



Molars 

 Platis LIV, LV 



In the primitive Notharctinse m| are notably wider than mi; m| are notably narrower than ml; 

 in the later forms there is a tendency for m| and mi to be of nearly equal width. In the Adapinse there 

 is considerable variation in proportions: there is a frequent tendency for the molars to become approxi- 

 mately equal in size, or m'^ may still be quite small (Adapis magnus var. leenhardti, Stehlin, 1912, p. 1279, 

 fig. 287). In the Lemuridse Myoxicehus suggests Adapis parisiensis var. schlosseri in the fact that 

 is nearly as large as m^. In Lemur on the other hand m^ is quite small. 



With regard to the detailed characters of the upper molars it has already been stated that in the 

 primitive Notharctinse these are tritubercular with only a faint beginning of the pseudohypocones and 

 mesostyles which become progressively developed in the later forms; these two features exclude all the 

 Notharctinse from direct ancestry either to the Adapinse or to the Lemuridse. In the Adapinse the 

 molars are primitively tritubercular {A. sciureus) but always have a cingulum-hypocone ; the protoloph 

 is always sharp; typically the metaconule is less distinct than in the primitive A. sciureus. In the 

 Lemuridse the molars are more or less degenerate in the finer details; thus the proto- and metaconules 



