] 899.] PLEXODONT MOLARS OF MAMMALS, 571, 



deciduous molars in function, and before the first persistent molar 

 has appeared, it can be seen that the last deciduous tooth, having 

 more than the necessary space for its development, is strongly 

 inclined posteriorly, so that it diminishes the space which will 

 have to be occupied by the persistent molars, and increases in the 

 same proportion the space assigned to the replacing molars. 



This inequality in the development of the molars also explains 

 why the last upper replacing molar of Eumioants and Artio- 

 dactyla generally is notably smaller and simpler, not only than the 

 one on its posterior end, but also than the one anterior to it. 

 This fourth replacing molar is the last to cut the gum, and must 

 adapt itself to the space left free by the penultimate replacing and 

 the first persistent tooth. 



Lastly, I have to observe that the sexcuspidate form of 

 tooth, which is represented as the last term of evolution of molars, 

 is very frequent in the oldest Tertiary Mammalia of Europe, and 

 especially in those of the Cernaysian Fauna. To judge from the 

 figures of the recent publication by Mr. Matthew on the Mamma- 

 lian Fauna of the Puerco, a great number of Mammals of this 

 epoch also have sexcuspidate inferior molars. Going a step 

 backwards, we can perceive, with the help of Osborn's and Marsh's 

 publications, that almost all the Mammals of the Upper Cretaceous 

 of North America are provided with sexcuspidate or even more 

 complicated (multituberculntej molars. Going another step 

 backwards, the figures published by Marsh enable us to recognize 

 the same type amongst several Jurassic genera, Peralestes, Pera- 

 spalax, Paurodon, Laodon, Dri/olestes, &c., which show their 

 posterior molars resembling those of Didelphyidiv and of Proteo- 

 didelpliys. Going still farther backwards, we find the oldest 

 known fragments belonging undoubtedly to a mammal, Microlestes 

 antiquus, with plexodont molars not far removed from those of 

 Proteodidelphys, and with a crown more closely resembling the 

 crown of unworn deciduous molars of certain primitive Ungulates 

 (ProsotJierium, Prolmjetotherium, &c.) than the molars of the 

 Plagiaulacidse {Plagiaidax, Neoplagiaidax, &c.). 



I do not maintain that the first complicated molars were sex- 

 cuspidate, rather than quadri- or quinque-cnspidate. On this 

 point I have sufiiciently explained my opinion in my memoir 

 " Sur I'Evolution des Dents des Mammiferes." The clear result of 

 all these facts is, that the famous theory of the gradual compli- 

 cation, of triconodonty and trituberculy, is an untenable hypothesis. 

 Nowhere do we meet with the stages leading from baplodonty to 

 plexodonty ; all those which have been mentioned are, on the 

 contrary, as I believe I have demonstrated, but the result of 

 simplificatioD of molars which were formerly more complicated. 

 Plexodonty therefore presents itself as a primitive character, 

 having made its appearance suddenly ; and it is only the theory of 

 fusion which can explain it in a satisfactory manner. 



