SOUTH AMERICAN GEOLOGY 
211 
fauna, and the evidence of emergence in the greater Antilles dur¬ 
ing the early Eocene times. 
During the Eocene and Oligocene periods lignite-bearing, con¬ 
tinental, or lagoonal, (non-marine) deposits are found in southern 
Chile, Tierra del Fuego, and Graham Land (Latitude 42 ° and 
southward). The flora contained in these beds is so similar that 
a land-connection with Antartica is imperative. Terrestrial faunas 
of this time are found in the so-called Notostylops and Pyrotheri- 
um beds of Argentina. 
Much naturally remains to be learned about the Late Tertic 
(Miocene and Pliocene) formations of South America. In particu¬ 
lar the extent of the Carribean submergence in Venezuela and Co- 
ombia is imperfectly known, as is the extent to which the now cov¬ 
ered basins of the Orinoco, the Amazon and La Plata were invaded 
by the sea. There is a strong presumption that there was interoce- 
anic connection from the Carribean Sea across Colombia and 
Ecuador to northern Peru. Vague records of travellers hint at 
this, and the fauna of the Zorritos formation of northern Peru is 
overwhelmingly Carribean in its affinities and Central America 
was emergent at that time, according to Vaughan. 
The early Miocene, marginal invasion of southern Chile (Navi- 
dad beds), and the similar Atlantic invasion of southern Argen¬ 
tine (Patagonian), have much in common; and both, but especially 
the former, contain a strikingly different fauna from that found 
in the Zorritos formation of Peru. This lack of community be¬ 
tween the Zorritos and Navidad faunas suggests a former greater 
extent of the land at this time somewhere between Latitude 12® 
and 35° South. If the geography has been like that of the present 
the Miocene Humboldt or Peruvian current would inevitably have 
carried the Navidad fauna northward as it does the present fauna. 
Emergence seems likely in the down-faulted regions which are now 
the Peruvian and Chilean deeps lying close to the present shores. 
This conception receives a measure of support from the fact of 
the identity of the two associated floras, their Brazilian affinities, 
and the fact that neither is a coastal flora. 
At this time the site of the present Andes was not sufficiently 
elevated to modify the climate of the present desert belt of the 
West Coast, or to prohibit the free spread of the flora southward 
12 See Spieker, E. M., Johns Hopkins Studies in Geology, No. 3, 1922. 
