LAKES OF GUIANA. 23 



barbour of refuge spacious enougb for whole fleets, by deepeniug the chanuel a 

 few miles long with which it communicates with the Araguari, and dredging tbe 

 estuary which has silted up, presenting in some places depths of little over three 

 feet. 



The subsidence and entire disajjpearance of the Guiana lakes is a process 

 which is accomplished very rapidly. The reeds and other plants which wither 

 in summer are deposited in a bed of floating humus in which various vegetable 

 species and even shrubs strike root. These verdant carpets are from time to time 

 rent by the storms and driven to the surrounding margins. Here they are soon 

 again massed together and thickened. Thus the lake becomes gradually filled in, 

 or transformed to a floating quagmire, firm on the surface, boggy in the lower 

 depths. In a short time nothing remains of the lagoon except the navigable 

 channel, the iyarapé, or " canoe track." Coudreau even, hazards the theory 

 that the lakes are emptied by a kind of see-saw movement of the banks.* 



In several of the lacustrine beds have been found huge trunks, whose origin 

 it is difficult to explain, except on the supposition that the lakes were at first dry 

 land, which afterwards subsided through some sudden disturbance of the ground. 

 Another explanation of the phenomenon, however, is suggested by the form and 

 trend of the seaboard. The alluvial promontories at the Oyapok, Approuague, 

 and Cachipour estuaries are all disposed in the direction of the norlh, and in 

 their lower course these rivers also follow the same direction, evidently under 

 the influence of the marine current, which deposits its sedimentary matter along 

 ihe shore. It may therefore be assumed that at contact with this current the 

 Araguari was also deflected northwards, and that the chain of lakes which have 

 the same trend are the remains of the old fluvial bed. 



The Carapaporis Strait, which flows between Maraca Island and the mainland, 

 and which is clearly distinguished by its greater depth from all the surrounding- 

 shallow basins, would on this hypothesis be the old mouth of the Araguari, 

 scarcely modified since the time when the river reached the sea more to the east. 

 If so much be allowed, there would be nothing surprising in the fact that, like the 

 Amazons, the copious Araguari should float down large trees and deposit them 

 along its lower winding course, which afterwards became a system of lagoons con- 

 nected together by tortuous channels. In the same way the marine current itself 

 intercepts the snags washed down by the Amazons, depositing them along its 

 muddy course, where they afterwards become embedded in the alluvial coastlands 

 of more recent formation. Such ligneous deposits have been found at depths of 

 78 or 80 feet. 



But however this be, great changes have been in progress even during the 

 contemporaneous period. A mere glance at the map suSices to show that the sea- 

 board of the contested Franco-Brazilian territory between the Araguari and the 

 Cachipour presents a striking contrast to the section of the coast running east and 

 west between Cayenne and the Corentyne estuary. This section develops a 

 regular convex curve, indicating the incessant deposit of sediment by the marine 

 ♦ La France Equinoxiale, Voyage à travers les Guyanes et V Amazonie. 



