GREGORY : 



NOTHAHCTl'S, AN AMERICAN EOCENE PRIMATE 



229 



Notharctus or its allies, a suggestion first made by Leidy in 1873 and independently put forth by Wort- 

 man in 1903. Indeed there is no feature hitherto observed in the entire skeleton of the New World 

 primates which may not very probably be regarded as a derivative of a more primitive condition in Noth- 

 arctus; in other words, in spite of the wide structural and palseontological hiatus between the Middle 

 Eocfene Notharctus on the one hand and the Miocene and recent New World primates on the other, we 

 can feel reasonably confident that although Notharctus may not be a direct ancestor of the latter group, 

 it is in general a structural ancestor. 



INFERRED ('HARA( TERS OF THE ( O.AIMON LEMI RIFORIM STOCK ANCESTRAL TO ADAPIN^E 



AND NOTHARCTIN.E 



While the earliest known members of the Notharctinse and of the Adapinte have already entered upon 

 the paths of specialization leading to their respective terminal stages, they are so much nearer to the 

 common stem form than are the later types that with the aid of the "Law of Trituberculy " it is not 

 difficult to reconstruct that form with some approach to accuracy. As to the general doctrine that the 

 Eocene representatives of many modern orders were tending back toward generalized insectivor(3us 

 dentitions, with the formula l|, C't, P|, M|, with tritubercular upper molars and tuberculosectorial lower 

 molars, the evidence is so abundant and so familiar to palaeontologists that it need not be itemized at 

 this point. Even Dr. Stehlin, whom none could justly accuse of being hasty in such matters, admits 

 that all the diverse patterns of the molars of Eocene Primates have been derived from a " tritubercular- 

 sectorial" structural plan. He says (1916, p. 1533) 



Es bedarf keiner unstandlichen Nachweise, dass der Molarstructur der Primaten der "tritubercular-sectoriale" 

 Bauplan zu Grunde liegt. Bei den meisten Stammen des Eocaens sind die Grundziige desselben noch sehr deutlich erhalten, 

 nur bei den neogenen Cercopitheciden finden wir sie vollig verwischt. Immerhin machen sich zahlreiche kleinere und 

 grossere Modificationen des Urplanes schon im Eocaen geltend. Sie erfolgen wie bei den Artiodactylen nach stark di- 

 vergierenden Richtungen und liefern iins fiir die Ergriindung der Stammesgeschichte schatzenswerthe Anhaltspunkte, 

 welche schon darum ein sorgfiiltiges Studium verdienen, well ja nur gar zu viele Formen vorderhand ausschliesslich oder 

 fast ausschliesslich durch ihre Backenl)ezahnung belegt sind. 



The probable characters which may be assigned provisionally to the dentition of the ancestral sub- 

 family of the Adapidse appear to be as follows. 



Paleocene of Asia (?). Size small, diet insectivorous or incipiently frugivorous. Upper molars 

 tritubercular, with neither hypocone nor pseudohypocone ; upper molar cingula well marked, para- and 

 metacone conical, not V-shaped. No mesostyles; m- wide transversely. Lower incisors of small size, 

 not chisel-shaped, not strongly procumbent; canines single-rooted, of moderate size; premolars simple, 

 p^ with a single external cusp and but little internal spur (protocone) ; p4 with very small low talonid, . 

 no distinct hypoconid; lower molars with high primitive trigonid, retaining the paraconid, and low, not 

 much expanded talonid, with very low entoconid; ms with basin-like hypoconulid; all molars in use 

 along with the deciduous molars. 



Some of the foregoing characters are retained in the more primitive members of the known subfami- 

 lies. 



To this hypothetical primitive subfamily may also be assigned such primitive dental characters as 

 are retained by both the known subfamilies, especially dentition if, Ct, P|, M|; DI|, DCt, DPf; proto- 

 conule of upper molars forming with the protocone a low oblique crest; metaconule tending to connect, 

 metacone and protocone. 



