860 



SCIENCE. 



LN. S. Vol. XII. No. 310. 



tween the fresh-water faunas of tropical 

 Africa and South America than the hypoth- 

 esis of an ancient land bridge ; supposing 

 that these faunas were only the remains of 

 an ancient cosmopolitan tropical fauna, the 

 paleontological evidence should be totally 

 different. 



In regard to the geological distribution of 

 the mammals of South America, the opin- 

 ions of the respective authors are very diver- 

 gent. There is, however, one point of which 

 there can be no doubt, i. e., the Pliocene 

 exchange of North and South American 

 types. It must be decided by North Ameri- 

 can zoologists whether this interchange has 

 commenced at the close of the Miocene or 

 only in the Pliocene. We may therefore con- 

 sider as Pliocene the Argentine Araucanian 

 formation, where the northern Artiodactyla 

 and other North American immigrants first 

 appear ; the Entrerian formation, containing 

 the neotropical forms must then be Miocene. 

 This formation was considered by Ameghino 

 Eocene (1889), or Oligocene (1898), and by 

 the writer (1898), Miocene. In favor of 

 his opinion Ameghino quotes the result of 

 the study of G. Alessandri on the fossil 

 selachian teeth of Entrerios which he be- 

 lieves to be Eocene. Mr. A. Smith-Wood- 

 ward, to whom I have sent the material 

 of our museum, writes me: "I conclude 

 that the formation cannot be earlier than 

 Miocene and is probably Pliocene." I have 

 called attention to the fact that in the En- 

 trerios deposits occurs Monophora darwini, 

 a Scutellid with perforated disk which is 

 common in the corresponding formation of 

 the north Patagonian coast. No Scutellidse 

 with perforations of the disk are known 

 earlier than the Miocene. On the other 

 hand, the Mollusca of this formation are 

 almost all extinct species and therefore I 

 cannot believe it Pliocene. 



Zittel in his ' Manual ' has well explained 

 the relations between the two American 

 mammal faunas. I am, however, disposed 



to believe, contrary to him and to Ame- 

 ghino, that the genus Didelphys in South 

 America appears as a member of the North 

 American immigration. If derived from 

 the Patagonian Microbiotheriidse, as sug- 

 gested by Ameghino, this genus may have 

 issued in the earlier Eocene time from 

 Patagonia and Archinotis and, after having 

 reached Europe in the Eocene and North 

 America in the Miocene, turned to South 

 America in the Pliocene. If Ameghino is 

 right, the Proboscidia are derived from the 

 Eocene Patagonian Pyrotheriidse and, after 

 having appeared in Europe and North 

 America, returned to Argentina during the 

 Pliocene in the form of the Mastodon. 



If this migration is a relatively well es- 

 tablished fact, it is quite doubtful in what 

 manner Patagonia received its rich mam- 

 malian fauna in the Laramie period. Flor- 

 entine Ameghino pointed out that this 

 must have occurred by means of a land- 

 bridge which united both Americas at the 

 beginning of the Tertiary period. On this 

 matter there has been a discussion between 

 Ameghino and the writer in the Revista 

 Argentina de Historia Natural,Yol. I., Buenos 

 Ayres, 18^1, p. 122 fif. and p. 281 ff., in 

 which I have combated this hypothesis. 

 The fresh- water faunas of the two Americas, 

 as I have shown, are so completely differ- 

 ent that only a prolonged and absolute 

 separation can explain the fact; the geo- 

 logical history of both North and South 

 America demonstrate an enormous devel- 

 opment of the Cretaceous Ocean, separating 

 the two Americas, and in the Tertiary 

 period the North American territory in- 

 creased but slowly. This presumed prim- 

 itive connection of the two Americas is not 

 at all supported by facts, but only based on 

 the predominance of wrong ideas of the 

 history of the Australasian Territory. The 

 Eocene mammals of Patagonia and North 

 America certainly do not justify this hy- 

 pothesis . The Eocene faunas of Reims and 



