76 UNDERHILL : 



ical to the truly historical ; here we suddenly come upon a 

 mammalian fauna exhibiting a state of advancement far 

 beyond the prophecy of any fossil thus far recovered from 

 preceding geological strata ; reptiles, which had been domin- 

 ant in the MesQzoic, are decreasing in size and number and 

 taking a subordinate position to a more modern class, mam- 

 malian quadrupeds become the dominant land animals, and 

 the rulers of the sea are great sharks and whales and those of 

 the air birds and bats. Thus the modern faunal aspect com- 

 mences, or, more properly, this is the first record that we have 

 of it, for the predominance and comparative advancement of 

 the mammals of the L,ower Tertiary indicate that there must 

 have been a higher development of the class during the Cre- 

 taceous era than has yet been revealed, and that the lost link 

 between the scarce little pouched mammals of the time of the 

 dinosaurs and these numerous and relatively well developed 

 forms is probably in some locality of the preceding geological 

 age as yet unknown to us. It seems reasonable to suppose 

 that this fauna assembled in America about the end of the 

 Cretaceous by migration from a country where future explor- 

 ation may yet reveal the steps in its intermediate development. 

 There was, of course, no such specialization among these 

 Eocene mammals as we observe in the class in later times. 

 They were all very generalized types — built upon the same 

 general plan, and, although Cope found ninety-three species 

 and some of the main divisions of the class, such as Carni- 

 vora, Herbivora, Insectivora and Primates, they are not yet 

 distinctly separated by the long process of adaptive modifica- 

 tion as we now know them. If they converge to a common 

 order, the Insectivora, in the Cretaceous, at the beginning of 

 the Tertiary the primitive representatives of modern orders 

 still approached each other so closely that if they were con- 

 temporary with us we would probably group nearly all of 

 them in the same family. It is from an evolutionary point of 

 view that the Tertiary mammals are especiall}^ of interest to 

 us, and it is merely the aim of this article to review in an ele- 



