222 GEOLOGY. 



in the revolution. In explanation of profound transformations of epi- 

 continental marine life, appeal has been made repeatedly to the with- 

 drawal of the sea, to the extension of the land, and to climatic changes 

 incident to deformative movements, and this appeal may now be made 

 so far as the change in the sea life is concerned; but the contrasted 

 phenomena on the land raises a new and unique question. The with- 

 drawal of the sea from its wide extension in Cretaceous times seems 

 in this case peculiarly well fitted to explain the transition in the epicon- 

 tinental sea life, because of the great differences in the areas of shallow 

 water in the two periods. It is worthy of note in passing, that the dis- 

 tribution of the harbors of refuge and other transition tracts of this 

 transformation had many points of analogy with those of previous 

 transformations, the Mediterranean region being again conspicuous in 

 this function. Such repeated service is a most significant illustration, 

 not simply of the persistency of continents, but of special continental 

 configurations. 



The increase of the land area and the establishment of new land 

 connections attendant on the post-Cretaceous withdrawal of the sea 

 might well have caused the vegetation to spread and flourish, if the 

 climate remained congenial; but why did not the animal life respond 

 in like manner? The record shows that plant life suffered little, although 

 plants are on the whole more responsive to climatic and topographic 

 influences than animals; why, then, did the saurians suffer so much? 



Closely correlated with this problem is the question, whence came 

 the placentals? Had their apparition anything to do with the extinc- 

 tion of the saurians and the repression of the rest of the reptile horde? 

 The origin of the placentals is one of the great outstanding problems 

 of paleontology. It is yet an open question whether the placental 

 mammals of North America and Eurasia arose from non-placental mam- 

 mals that had been natives of these provinces in the Jurassic and Creta- 

 ceous periods, or whether they were immigrants from some other region. 

 No satisfactory evidence of a transition from non-placental to placental 

 mammals in Eurasia or North America has yet been produced, but the 

 imperfection of the record may be appealed to. The relative sudden- 

 ness and overwhelming power of the placental irruption suggest inva- 

 sion from some other quarter in which the earlier evolution of the pla- 

 centals had been in progress for a long time previously ; whether from 

 marsupial or from independent stock, we need not here inquire. The 



