86 NATURAL HISTORY OF 



trial soil.* The multitude of enormous volcanoes in 

 the Andes, do not appear to have depressed the east 

 coast to a perceptible submersion ; or rather, to what is 

 more than fully replaced by the deposits of the vast 

 and numerous rivers which intersect the whole surface. 

 It is, moreover, stayed by the mountain systems of 

 Brazil, Guyana, and Venezuela ; from whence, and from 

 the basins at the foot of the Quindiu Cordillera, and 

 the Pacaraima mountains, have been effected man^ 

 entire discharges of elevated lakes, such as the Amucu 

 and Savanhas of Dutch Guyana, while the swamps of 

 the Parana, and the lagoons on the coast, remain un- 

 changed. But at the northern extremity of South 

 America, where the Andes present an interruption in 

 the direct chain, a branch turning eastward, and then 

 to the north, shows a connection from volcanic Trinidad, 

 through the West Indian Islands, till the mountain 

 character, but not the volcanic connection, is lost in the 

 island of Cuba. All this enormous surface, from Bar- 

 badoes to Yera Cruz, forming the two distinct basins 

 of the Caribbean Sea and Gulf of Mexico, present many 

 indications of a violent disruption belonging to the pre- 

 sent geological superficies of the earth, and perhaps not 

 remote in date from the submersion of Atlantis on the 

 African coast. A series of volcanic craters, still in 

 violent ignition, may have worked on the single moun- 

 tain ridge, of no great breadth of base, pressed by 



* In most volcanic upheavings, there follows a subsi- 

 dence, nature endeavouring to return to its anterior equi- 

 librium j but the result is rarely down to the former level. 



