439 



the middle of the stream. Our canoe touched 

 several times during the morning. These shocks, 

 when violent, are capable of splitting a light 

 bark. We struck on the points of several large 

 trees, which remain for years in an oblique posi- 

 tion, sunk in the mud. These trees descend 

 from Sarare $ at the period of great inundations. 

 These so fill the bed of the river, that canoes in 

 going up find it difficult sometimes to make their 

 way over the shoals, or wherever there are ed- 

 dies. We reached a spot nea,^ the island of 

 Carizales, where we saw trunks of the locust- 

 tree of an enormous size above the surface of the 

 water. They were covered with a species of 

 plotus, nearly approaching the anhinga, or whit e 

 bellied darter. These birds perch in files, like 

 pheasants and parrakas. They remain for hours 

 entirely motionless, with the beak raised toward 

 the sjiy, which gives them a singular air of stu- 

 pidity. 



Below the island of Carizales we observed a 

 diminution of the waters of the river, at which 

 we were so much tjie more surprised, as, after 

 the bifurcation at la Boca de Arichum, there is 

 no branch, no natural drain, that takes away 

 water from the Apure. The loss is solely the 

 effect of evaporation, and of filtration on a sandy 

 and wet shpre. We may form an idea of the 

 magnitude of these effects, when we recollect, 

 that we found the heat of the dry sands, at dif- 



