328 Bulletin American Museum of Natural Hi.storij. [Vol. XXVII, 



typical Placental orders. But Dr. Wortman's argument seems to be weak- 

 ened to some extent by the following considerations: 



(1) While the premolars in certain groups {c. g., Zalambdodonts) ap- 

 pear to furnish reliable indications of the early history of the molar cusps, 

 in other groups {e. g., Ungulates) they appear not to do so. 



(2) Poramys has already departed from the typical Placentals rather 

 widely in the skull structure; and in its lower teeth the primitive trigonid may 

 have been modified in the manner described below. 



(3) Dr. Wortman takes the view that the true protoconid in the lower 

 molar of Paramys is on the internal side. A frequently reliable criterion 

 of the homology of cusps is not position but function with reference to the 

 parts of the opposing teeth. The main internal cusp of the upper molars 

 of Paramys is certainly analogous with the protocone of most mammals 

 because in life (as shown by comparison with Ardomys and Diclelphis) it 

 fitted into the basin of the talonid of the lower molars. In the lower molars 

 the functional protoconid is the reduced antero-external cusp of the lower 

 molars; because, like normal protoconids, it fitted into the space behind the 

 metacone of an upper molar and in front of the paracone of the next upper 

 molar. This space in the more advanced forms was gradually filled by the 

 rising parastyle-cingidum ridge, while the tip of the protoconid was lowered. 

 The high antero-internal cusp of the lower molars, which Dr. Wortman 

 regards as the original tip of the crown, is probably not the true protoconid 

 but either the paraconid, the metaconid or the paraconid + metaconid; 

 because, like these cusps in normal trituberculate mammals, it fitted into 

 the internal space between two upper molars. The raising of the parastyle, 

 lowering of the protoconid and decided elevation of the antero-internal lower 

 cusp in the modern Sciurid are seen to be mechanically correlated phe- 

 nomena (see below). 



(4) The mechanical reason for the peculiar obliquity and atypical de- 

 velopment of the Sciurid lower molar is the oblicjuely lateral vibration of the 

 mandible in mastication. True propalinal (antero-posterior) motion is 

 only possible in gnawing with the incisors, when the molar crowns are not 

 in contact. Anteroposterior motion during mastication would be precluded 

 by the irregular wearing surface of the crown. Even in Rodents (except 

 Caviidae) in which the wearing surface of the complexly folded crown is 

 perfectly plane, the movement of the mandible in mastication (so far as can 

 be judged from the skull and jaw alone), seems to be obliquely lateral 

 rather than strictly antero-posterior. 



The molar pattern of Paramys, which all authorities regard as the most 

 primitive pattern among Rodents, accordingly seems to offer no insuperable 

 obstacle to the view that the Rodents have been derived from mammals 



