24 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 94 



the American Indians in variable degrees, Asiatic (Mongoloid and non-Mongo- 

 loid), Melanesian, Polynesian, Caucasoid, and Australian affinities. I believe 

 the first to be the clearest, the last most attenuated. There have also been men- 

 tioned [in connection with America] Pygmies, or Negritos. This problem awaits 

 solution." 



Regarding the Australia-Antarctic-South America route for human 

 immigration — " the topographic, biogeographic, paleontological, and 

 anthropological considerations render the utilization of this route very 

 probable " ; though he is aware of the conjectural nature of his hy- 

 pothesis and of the fact that its adoption demands objective proofs.^' 



In the same year Rivet (1926 c) discusses the problem of the way 

 of the supposed Australian migrations toward the American continent. 

 He states that during the last few years it has been possible to " es- 

 tablish ", by anthropological, ethnographic, archeological, and lin- 

 guistic means, immigrations of Malayo-Polynesian, Melanesian, and 

 Australian contingents into the New World, but that the mode of 

 coming of the Australians offers still considerable difficulties. They, 

 as well as the Tson (the Patagonians, the Onas), their South Ameri- 

 can " relatives," knew but primitive means of navigation which 

 were insufificient for any long sea voyage, and there is no indication 

 that they possessed anything superior in this respect in the past. 

 The coming of the Australians by way of the Bering Strait must be 

 excluded — there are no traces of them on the way. There remains the 

 southern route suggested by Correa, over the islands and waters of 

 the Antarctic, which may have presented a different configuration in 

 the past, due possibly to a greater ice extension. The material evi- 

 dences of such migration remain to be determined. 



*' " L'analyse anthropologique, I'ethnographie, la linguistique, nous permettent 

 de croire a la genese des populations precolombiennes de I'Amerique en ad- 

 mettant I'intervention de diverses stratifications ethniques. On pent demeler 

 parsemees, en degres variables, chez les Indiens de I'Amerique des affinites 

 asiatiques (mongolo'ides et non-mongoloides), melanesiennes, polynesiennes, cau- 

 casoides, australiennes. Je crois que les premieres sont les plus nettes, les 

 dernieres les plus attenuees. On a parle aussi de Pygmees, voire de Negritos. 

 C'est un probleme a resoudre." (P. 107.) 



^ " Mais les considerations topographiques, biogeographiques, paleontologiques 

 et anthropologiques rendent tres vraisemblable I'utilisation de cette voie d'acces. 

 .... Certes I'adoption definitive de mon hypothese exige prealablement des docu- 

 ments objectifs de I'utilisation de cette voie par I'homme. II faut cependant 

 esperer qu'on obtiendra dans I'avenir ces documents, dont un grand nombre sont 

 peut-etre ensevelis sous les glaces mysterieuses des terres antarctiques." (P. 116.) 



" Je ne conteste pas le caractere conjectural de plusieurs de mes considerations. 

 On n'oserait pas donner comme deja etablie la voie de transit dont j'ai presente 

 I'hypothese. Mais je n'hesite pas a consi^erer celle-ci comme legitime et tres 

 vraisemblable." (P. 118.) 



