4 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 139 



note, however, that the upper teeth seen in Clark Fork Apheliscus 

 nitidus seem intermediate in most, if not all, of these respects. A com- 

 parison of Matthew's figure (1918, fig. 24) for the Clark Fork speci- 

 men, which Simpson (1937b) made the type of A. nitidus, with P 4 

 and M 1 in Phenacodaptes sabulosus, here shown in plate 2, figure 2, 

 reveals little to distinguish them. The Sand Coulee lower teeth of 

 Apheliscus, figured by Matthew (1918, fig. 24) also seem to show 

 a little less compression of the trigonid than more typical Gray Bull 

 specimens. 



The foregoing comparisons strongly suggest that Phenacodaptes, 

 or at least a very closely related form, gave rise to Apheliscus. The 

 succession may well have been Phenacodaptes sabulosus-Apheliscus 

 nitidus-Apheliscus insidiosus. In the course of this postulated de- 

 velopment it would seem that the principal tendency was toward the 

 transverse narrowing of the teeth, both upper and lower series ; the 

 loss or weakening of the cingulum in the upper series ; the increas- 

 ingly Pentaco don-like development of P 4 ; the relative increase in 

 length of talonid of the lower teeth, P 4 as well as the molars ; together 

 with the shortening of the lower molar trigonids. 



RELATIONSHIPS OF APHELISCUS AND PHENACODAPTES 



The most nearly comparable development to that illustrated in the 

 Phenacodaptes-Apheliscus line would seem to be among the panto- 

 lestids. The suggested comparison is perhaps not so close to the 

 Bessoecetor-Propalaeosinopa-Palaeosinopa-Pantolestes succession as 

 it is to the middle Paleocene Pentacodontinae. The premolar develop- 

 ment would seem rather like that in both Aphronorus (see pi. 2, figs. 3 

 and 4) and Pentacodon (see pi. 2, figs. 5 and 6), except that there 

 tends to be no metaconid on P 4 or tritocone (uncertain for Pentaco- 

 don) on P 4 in the apheliscids. Aphronorus, moreover, differs in having 

 somewhat higher crowned, more definitely insectivore teeth. The 

 upper molars of Aphronorus show better developed and more laterally 

 directed anteroexternal and posteroexternal styles and the lower molar 

 trigonids are a little higher and show better development of the para- 

 conid. 



Much larger Pentacodon has a more enlarged fourth premolar, but 

 the upper molars (not previously illustrated) do not show the distinc- 

 tive outer styles seen in Aphronorus. Also the trigonids of the lower 

 molars do not appear to be so elevated, but, like Aphronorus and unlike 

 the apheliscids, show a prominent and forward-placed paraconid. The 

 talonid construction, nevertheless, is much alike in the two subfamilies, 

 except for relative length. 



