Vlll INTRODUCTION. 



tive size as in other Ungulates, and are of equal functional import- 

 ance ; but in the Elepliantidce the tendency to the gradual increase 

 in relative size of the hinder cheek-teeth, and especially of the third 

 true molar, makes it evident that in the more specialized forms the 

 occurrence of successional cheek-teeth in the earlier part of the 

 series would be merely an encumbrance, and we accordingly find 

 such teeth totally absent. In some of the earlier Mastodons, how- 

 ever, the premolars were probably of some slight functional import- 

 ance ; but their relatively small dimensions indicates that immedi- 

 ately the elephantine type of dental succession had commenced 

 these teeth were practically doomed to disappear ; and it is inter- 

 esting to observe that while they had totally disappeared in the 

 simple-toothed Mastodon americanus of the Pleistocene, they still 

 lingered on in some of the generalized true Elephants of the Pliocene 

 (ElepTias clifti and E. planifrons), and thus exemplify the well- 

 known fact that a much longer period is frequently required to 

 eradicate functionally useless parts of the organism than to develop 

 other parts to a comparatively high degree of specialization. This 

 complete disappearance of the whole of the premolar and the reten- 

 tion of the milk-molar dentition is a feature quite peculiar to the 

 Elephantidce, although, if the writer's interpretation be correct, a 

 similar instance occurs in respect of the first cheek-tooth of the 

 Perissodactyla. 



The specialization of the cheek-teeth in the higher members of 

 the Proboscidea has followed to a considerable extent a line analo- 

 gous to that obtaining in the Perissodactjda and Artiodactyla, and 

 shows itself in the increased height and complexity of the crowns 

 of these teeth, and in the final attainment of a nearly horizontal 

 and continuous plane of wear. Prom the structure of the cheek- 

 teeth in the more generalized members of the suborder, it may be 

 inferred that the action of the molars of one jaw upon those of the 

 other must have been mainly a scissor-like or snapping one, while 

 in the more specialized forms this action has been converted into a 

 perfect grinding motion. 



The collection of Proboscidean remains recorded in the following 

 pages is probably by far the largest in the world, and since it con- 

 tains specimens belonging to nearly all the more important species, 

 it is admirably adapted to exhibit the almost complete transition in 

 respect of the structure of the cheek-teeth, which can be traced from 

 the most generalized to the most specialized member of the Elcphaa- 

 ticlce. So complete indeed is this transition (as will be indicated in 



