TR1TUBERCULAR THEORY. 269 



triangular form and becomes more or less quadrangular. Here 

 is the fundamental type from which the molar teeth of most 

 herbivorous mammals may readily have been derived. The 

 homologue of the primitive reptilian cone, according to this 

 doctrine, is thus easily identifiable even in the most modified 

 molars, and it has received the distinctive name of protocone 

 in the upper, protoconid in the lower jaw. The other cones 

 are all regarded as constant and likewise named. Examples of 

 this nomenclature are given in the legends to figs. 183, 185, 

 187, 189, and 191. 



Some palaeontologists are inclined to believe that the advo- 

 cates of this Tritubercular Theory apply it in a purely fanciful 

 manner to the naming of the cusps in the more complicated 

 molars, and that there is no adequate proof of the homologies 

 they perceive. They also point to the primitive molars of the 

 Multituberculata as a serious obstacle to its acceptance. While 

 embryologists have adduced some striking evidence in its favour 

 (e.g. identifying the three supposed primitive cones in the germs 

 of such highly modified molars as those of Equus and Bos), 

 they also have done much to render its basis insecure, that is, 

 if ontogeny can be relied upon as generally parallel to phylogeny. 

 They have clearly shown that the order of calcification of the 

 various cusps in the molars of many recent genera is quite 

 different from the order of phylogenetic development of these 

 cusps assumed by the Tritubercular Theory. Hence this at- 

 first-sight brilliant generalization can only be accepted at pre- 

 sent as a convenient working hypothesis which remains on its 

 trial. 



As to the rival theories concerning tooth-succession in the 

 Eutheria, propounded and discussed by embryologists, Palaeon- 

 tology has not hitherto furnished any evidence of importance 

 for consideration. 



ORDER 1. CETACEA. 



As already remarked, the order of whales, dolphins, and 

 porpoises can be traced back to an early part of the Tertiary 

 period without obtaining any real clue to its origin. The 

 earliest forms exhibit a more normal rostrum, less unlike that 



