* 
58 C. F. Hartt—Tertiary Basin of the Marafion. 
vary in age and greatly in the conditions under which they were 
, deposited. 
Ase Pebas shells do not shed one ray of light on the ee 
question of the glaciation of the Amazonian valley. I hav 
owever, shown that the supposed facts on which Prof. Aegean 
founded his theory, viz: the assumed identity of structure of 
the Serras of Ereré and Parti (Almeyrim); the occurrence re 
erratics of diorite at Ereré, etc., were no facts at all. Hrer 
a monoclinal ridge of sandstone which no geologist would ever 
think of calling drift, and the supposed drift clays at its base 
contain lower Devonian trilobites and are traversed by trap 
dykes; the supposed erratics of diorite are boulders of decompo- 
sition; the Serras of Parti* are composed of horizontal beds of 
soft rocks undoubtedly more modern than the coast of Kreré 
and offering not the first evidence of glacial origin ; the gigan- 
tic moraine which Prof. Agassiz thought to have pete 
across the mouth of the Amazonas does not exist. Moreover 
I have failed in finding, during many months of careful search, 
anything like drift in the province of Para; and therefore, hav- 
ing no evidence whatever of the former existence of glaciers in 
the Amazonas, or question of the glacial origin of the valley 
need not be rais 
eI do sok ‘believe in the glaciation of the Amazonas, I 
still saThies to the belief that glaciers have existed in the cen- 
tral and southern portions of the Brazilian plateau. Prof. O. H. 
St. John, who, as one of the geologists of the Thayer expedi- 
tion, made a journey through the interior of Brazil from Rio de 
Janeiro to Maranhao, assures me that he has found not only the 
superficial deposits, but also the topography characteristic of a 
glaciated country in Minas Shea while these phenomena are 
not visible in Piauhy and Mar 
ough the Pebas shells thick no light on the question of 
the glaciation of the Amazonian valley, they aid in establishing 
the fact that the Upper Amazonas or “Maraiion, from Iquitos to 
ae a distance of some 240-250 miles, flows through a 
‘cca Be asin, _ we of the river being deepl cut through 
this age. The width of this basin is undetermined, as 
is also the atest age of the beds, for the nature of the fauna is 
such that it is impossible to say to which division of the Ter- 
tiary they are to be referr The fauna indicates an 
estuary formation. That at the time of the deposition of Pebas 
beds there was water communication between the basins of the 
Amazonas and the Orinoco is scarcely probable. 
* IT visited the Serra of Paraudquara in 1871. 
