184 Bulletin American Museum of Natural History. [Vol. XXVII, 



Tertiary families is at first small, like the internal shelf-like "deuterocone" 

 of the upper premolars, and because the two develop together jpari passu so 

 that the deuterocone finally develops into a cusp resembling the protocone 

 of the molars, Gidley seems to conclude that the same is true of the molars 

 of Jurassic mammals. 



According to this a very low and small talonid ought to imply a small 

 protocone; but as shown below in the Jurassic Trituberculates a high pointed 

 protocone of the type seen in Dryolestes was probably received into a low 

 small talonid; and instead of the protocone being a "heel" evoked as in 

 the premolars by the preexistence of the talonid, the reverse has probably 

 been the case. The peculiar recurved protoconid and small concave talonid 

 of Peramus and Leptocladus (Osborn, 1907, p. 28, fig. 18), on this hypothesis 

 imply the existence of a high protocone. The high large protocone of 

 Peralestes was (it has been argued above, p. 174) received into a low talonid, 

 while the small talonid of Ampltiiherium (p. 180) seems to require a large 

 protocone. In short, the theory that the protocone is of more recent origin 

 than the paracone and has been called into existence as a "heel," pari passu 

 with the development of the talonid below is not favored by the present 

 interpretation of the evidence furnished by the Jurassic Trituberculata. 



The Theory of Trituberculy correct in its basal postulate, the originally re- 

 versed relations of trigon and trigonid. 



The well known plasticity of the dentition in mammals and the frequency 

 of convergent evolution raises an a priori presumption that the tritubercular 

 type has arisen more than once in mammalian history; and it is indeed 

 possible that the so called protocone is not always strictly homologous 

 throughout the mammalia; while even among Jurassic mammals the more 

 or less triangular upper molars of Peralestes and Dryolestes may have attained 

 their observed features by two different roads of development. But this 

 does not imply that the celebrated theory of trituberculy has been entirely 

 disproven by its numerous critics. Stripped of all non-essentials and inter- 

 preted in a new manner the fundamental thesis of the Theory of Trituberculy 

 may still be correct: namely, that the trigon above and the trigonid below are 

 morphologically reversed structures, and that all the accessory parts are the 

 offspring of a main "protocone," the tip of which lies on the inner side of the 

 upper, and on the outer side of the lower, molars. The writer therefore 

 ventures to put forth into the already crowded field still another hypothesis, 

 which may at least help to make clear some mechanical principles that will 

 be of some assistance later on in this work. 



