400 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 48. 



and if regarded as ancestral furnishes very good evidence against the 

 earlier supposition that Myrmecobius is a direct and little-changed 

 descendant of some one of the "Jurassic mammal-like forms having 

 more than the normal number of teeth. It is indeed very doubtful 

 whether the latter gave rise to any of the higher mammals, and such 

 forms as Dryolestes, Dicrocynodon, Tinodon, etc., of the Jurassic, 

 with their many post-canine teeth, were probably not marsupials at 

 all. A recent restudy of these ancient forms leads me to believe that 

 there are good reasons for regarding them rather as monotremes, and 

 in this group may possibly be found the early representatives of the 

 living members of this strange order of mammals. 



In considering the derivation of the marsupials and placentals it 

 must not be overlooked that regardless of the origin of the fourth 

 molariform tooth in the marsupials the normal number of post-canine 

 teeth of primitive or generalized forms in both groups is invariably 

 seven, any deviation from this number being due to a loss or addition 

 through specialization. The obvious inference, then, is that the 

 common ancestral forms from which these great groups were origi- 

 nally derived had a like normal dental formula. Such a genus, there- 

 fore, as Triconodon, or some other form having four premolariform 

 and three molariform teeth behind the canine, would be a more 

 logical Jurassic ancestral type for the higher mammals than would 

 such forms as Dryolestes, Dicrocynodon, etc., which have many more 

 than the normal number of both premolariform and molariform teeth. 

 The ultimate origin of these great groups is, however, at best largely 

 speculative with our present knowledge. 



NOTES ON THE ORIGIN OF THE MARSUPIALS. 



From expressing disbelief in Owen's hypothesis regarding the 

 origin of Myrmecobius, Bensley seems to have gone to the other 

 extreme in suggesting the derivation of all the living marsupials from 

 an Oligocene form (Peratlierium) of the Didelphidae. Osborn 1 seems 

 to have adopted this view also, while Gregory 2 in his recent contri- 

 bution on The Orders of Mammals, seemingly accepts Bensley's views 

 in general but gives much greater antiquity to the ancestral stock of 

 the marsupials. I can not regard Bensley's view as wholly tenable, 

 even if the didelphid prototype were carried to a much more remote 

 time than the Oligocene. If the Fort Union mammal here described, 

 whether directly ancestral or not, is in any way related to the living 

 Myrmecobius, we have evidence that the Myrmecobidae had at least 

 reached a marked degree of specialization which separated this family 

 distinctly from the other marsupials at a much earlier date than is 

 assumed by Bensley for the differentiation of the whole order. Even 



1 Evolution of the mammalian molar teeth. Biol. Studies and Addresses, vol. 1, 1907, p. 109. 



2 Orders of Mammals. Bull. Amer. Mus. Nat. Hist., vol. 27, 1910, p. 229. 



