MATTHEW, PATAGONIA AND PAMPAS CENOZOIC 
155 
The author considers that the Equidse reached South America from 
the Old World by way of a South Atlantic land bridge. 
Roth’s argument does not appear convincing to the reviewer. Against 
it may be briefly noted the following: 
1. Any other source than North America for the invading fauna in¬ 
volves geographic changes of a highly improbable character. 
2. The existence of a land bridge between Africa and South America 
in the late Tertiary would almost certainly involve a community of fauna 
between the two continents which does not exist. 
3. The counter-migration from South America took place to North 
America and to North America only. It occurred chiefly at the end of the 
Pliocene, as recorded in the North American faunal succession; but doubt¬ 
ful precursors are found in the Middle Miocene (Mascall x ) and Lower 
Pliocene (Snake Creek 1 2 ) of North America. 
4. There is no difficulty in the derivation of all the South American 
Equidse from the more primitive North American Equidse of the late Mio¬ 
cene and Pliocene ( Protohippus, Pliohippus, Neohip par ion). 
5. The Siwalik fauna is more or less composite and includes Pleistocene 
as well as Pliocene species. It is doubtful whether any considerable part of 
it is Miocene. 
6. The Pampean Canis, Felis, Smilodon, Mastodon etc. are most 
nearly related to late Pliocene and Pleistocene species of the north, and 
especially of North America. The so-called Canis and Mastodon of the 
northern Miocene are much more primitive forms, well separable generically. 
7. All the genera of northern origin recorded from the Middle Pampean 
are identical with or equivalent to the Pleistocene genera of North America, 
and decidedly more advanced than the Upper Miocene and Pliocene genera 
of this continent. Equus, Tapirus, Cervus, Mastodon, Arctotherium, Canis 
and Smilodon first appear in North America in the Pleistocene. Hippi- 
dion and Onohippidion are equivalent in specialization to Equus, and are 
derivable from the much more primitive Pliohippus and related genera of 
the Upper Miocene arid Pliocene. The so-called Listriodon and Catagonus 
of the Pampean are closely related to Platygonus of the Pleistocene and 
Prosthennops of the Upper Miocene and Pliocene of North America. Au- 
chenia and Palceolama are equivalent in specialization to the Pleistocene 
camels of North America, decidedly more advanced than the Upper Miocene 
and Pliocene Procamelus, Alticamelus and Pliauchenia of this country. 
The same relations appear in the Pampean Cervidse. Among the Felidae 
Smilodon is found only in the Pampean and in the North American Pleisto- 
1 Sinclair, 1906. 
2 Matthew and Cook, 1909. 
