308 The American Naturalist. [ ApriIr 



Now in Tapirulus the two external concave lobes of the superior 

 molars have nearly the same structure as those of Dacrytherium, and 

 the internal primitive bunodont elements of the crown, have been trans- 

 formed into crests, this occurred before the Anoplotherium type of 

 molar had reached its present stage of evolution. It is then not difficult 

 to derive the form of superior molar found in Tapindus from the true 

 buno-selenodont type, and I believe this to have been the origin of the 

 peculiarly modified molar occurring in Tapirulus. 



The superior and inferior premolars in Tapirulus have the same 

 elongated form so characteristic of the Anoplotheriidce in general, and 

 like the latter group, the canines were not differentiated in form from 

 the anterior premolars. The lower true molars of Tapirulm depart 

 widely in structure from those found in any of the known genera of the 

 Anoplotheridv, although like the upper molars they are much special- 

 ized and can be derived from a less modified type, which was the 

 common stock of both Tapirulus and Anoplotherium. A peculiarity 

 111 the structure of the lower molars of Tapirulus is the presence on 

 each tooth of a well developed third lobe, hypoconulid, which projects 

 posteriorly a considerable distance. The portion of the molar crown 

 anterior to the hypoconulid consists of two high transverse crests the 

 antero-external termination of the front crest exhibiting a rudiment of 

 the anterior spur, which is so largely developed in all the other genera 

 of the AnophfheHdtt. The lower jaw in Tapirulus is long and ex- 

 tremely slender and its form closely resembles that of Dacrytherimn. 



In conclusion I believe the natural position of Tapirulus is in a sub- 

 family of the Anoplotheriidce, and that both Tapirulus and Aunplnther- 

 ium have been derived from a common stock, the ancestral form hav- 

 ing had the pure type of buno-selenodont molar. As already shown 

 above, the resemblance in structure of the molars of Tapirulus to that 

 of the Tapir is not exact, the Tapirs tooth having been derived direct 

 from the bunodont type (See Euprotogonia and Systemodon^ 

 III. 

 On the validity, „„, systematic position „( Mi.rfntherimn Filhol— The 

 genus MLrtotheriim* is another interesting form, not as aberrant in it* 

 characters as Tapirulus, but which unites in a certain degree the 

 feuillines and the Anoplotheroids. Prof. Zittel in his "Traite de 

 PalaBontologie" does not consider Mixtotherium as a good genus and 

 refers it to the milk dentition of Diplobune. I am quite confident that 

 1-roi. Zittel is mistaken in this determination, and I shall endeavor to 

 5 Mem. quelques Mam. fossils, Toulouse, 1882. 



