Derby.] 1"0 [Feb. 21, 



of Pvainha, Monte Alegre, Santarem and Obydos, in the midst of which 



there appears occasional]}' a conical or flat-topped peak, to attest the orig- 

 inal character of the plain and the extent of the denudation. The table- 

 lands and their slopes are generally wooded, while the lower undulated 

 plains are open and grassy, covered with a barren soil of loose sand. In 

 the interior, on both sides of the river, these table-lauds appear to rise 

 gradually in height, until they become united with the more elevated plains 

 of central Guiana and Brazil. 



The last division of the highlands, that of the hilly or mountainous 

 country, is represented, near the northern bank of the Amazonas, by an 

 isolated group of mountains, in the vicinity of Monte Alegre and Erere. 

 These rise abruptly in the midst of a plain to a height of 1,000 feet, and are, 

 in general, rocky or sandy and barren. Associated with these mountains 

 and having the same geological structure is a low, stony campo. Ascend- 

 ing the tributaries on both the northern and the southern side, there is 

 found, in the regions of the rapids, at a distance varying from 50 to 200 

 miles from the main river, a hilly country, whose highest points are, in 

 general, lower than those of the Erere group of mountains. These hilly 

 regions are usually well-wooded, with many valuable kinds of timber, the 

 Brazil and sapucaia nut trees (Bertholhtia excclsa and Lecythis grandi- 

 flora) being very abundant and characteristic. To these hilly regions 

 succeed, on the north, the high mountains of Guiana and, on the south, 

 the table-lands of central Brazil. 



The differences above noted in the different regions of the highlands or 

 terra firma depend on the geological structure of the valley, and before 

 describing minutely the different formations, it may lie well to present a 

 general sketch of the geology of this part of the valley, and indicate the 

 relations of the regions above described. 



Prof. Hartt has well described this structure as follows ;* "The Ama- 

 zonian valley first appeared as a wide strait between two islands or groups 

 of islands, one now forming the base and nucleus of the Brazilian plateau, 

 the other, on the north, the plateau of Guiana. These islands first, appeared 

 at. or shortly after, the beginning of the Silurian Age." 



In this canal, before the elevation of the Andes, were deposited a series 

 of beds, representing the Upper Silurian, Devonian, Carboniferous and 

 Cretaceous, which appeared successively in dry land on each side, narrow- 

 ing the strait between the two islands. Prof. Hartt continues: "Before 

 the rise of the Andes the valley of the Amazonas consisted simply of two 

 gulfs united hy a narrow strait. The Andes were thrown up across the 

 mouth of the western gulf, converting it into a. basin, though it probably 

 had an outlet both to the north and south. fiThe whole continent was 

 afterwards depressed, so that the waters covereuwidely the Guayanian and 

 Brazilian plateaux, and the Tertiary beds were deposited there, varying in 

 thickness, coarseness or fineness, according to the conditions under which 

 they were formed. * * * * 



* Journal of the American Geographical Society, Vol. Ill, p. 231, 1872. 



