TO W. D. Matthew — Mammalian Migrations. 



tive genera {Cynohyo&nodon, Quercyrtheriunx) in Europe, 

 which may be of antochthonic origin. 



(4) No Hysenodonts are found in the Middle Eocene of 

 Africa. 



These facts seem to indicate an outside region, probably Asia, 

 for the center of development and dispersion of the Hysenodonts. 

 Thence they reached North America in Lower Eocene, 

 developed till Middle Eocene and became extinct. They 

 reached Europe in Middle or* Lower Eocene and Africa in 

 Upper Eocene. In Upper Eocene higher stages in the evolu- 

 tion of the race {Pterodon, Hyamodon) invade Europe and 

 Africa and spread to America* in Lower Oligocene. The dis- 

 tribution of this family is more or less completely paralleled 

 by several other faunal groups. 



Inference from the above as to early Tertiary continental 

 connections : || = separation ; < — > — union permitting in- 

 termigration ; ? = doubtful.] 



Middle Oligocene, North America Asia< > Europe ? Africa 



Lower Oligocene, North America^ > Asia< > Europe '? Africa 



Upper Eocene, North America || Asia-^ — > Europe | Asia< — > Africa 



Middle " North America Asia< > Europe fl Africa 



Lower " North America < — >Asia< — > Europe || Africa 



Basal " Asia North America < >> Europe || Africa 



This is of course only a working hypothesis which, in my 

 opinion, accords best with the data as far as I know them. In 

 brief, it is that Asia is and has been the great center of evolu- 

 tion and dispersion of the dominant mammalian types ; in the 

 other continents, the course of evolution has been — aside from 

 a few well-known exceptions — alternately an autochthonic fau- 

 nal development and a series of waves of migration from the 

 highly progressive faunas of the great Asiatic land mass, 

 according as the continents were separated from or connected 

 w T ith it. The principal exceptions are the Proboscidea, of Afri- 

 can origin, the true Edentates of South American develop- 

 ment and doubtful origin, the Camels, of North American 

 origin, — probably other groups, if we knew something about 

 the fauna of the early Tertiary of Asia. 



* "Or Lower " because, although Sinopa has not been found in strata of 

 Sparnacien time of Europe, several of its associated genera of the Wasatch 

 do occur there and the fauna is very imperfectly known. 



