733 



at the two periods of their maximum and mini- 

 mum *, that during- the floods it is often fifteen 

 or twenty times as much as at the season of 

 drought. 



When we have doubled the Punta Barima^ 

 and entered the bed of the Oroonoko, we find it 

 to be only three thousand toises in breadth. 

 Greater estimations have arisen from the error 

 of pilots in measuring the river in a line not 

 perpendicular to the direction of the current. 

 It would be useless to fortify the island of Can- 

 grejoSj near which the water is from four to five 

 fathoms deep. Vessels there would be out of 

 gun-shot. The labyrinth of channels that lead 

 to the little mouths {hocas chicas) changes daily 



* Mr. Girard found the volume of the Nile, at the port of 

 Syout, in the time of low water, 678 cubit metres in a se- 

 cond, while the gauges gave him during the inundations 

 10247 cubic metres [sur la Valtte d'Egypte, p. 13). We may 

 judge by analogy of the enormous increase of the Oroonoko, 

 if we recollect, that it rises 25 feet in places where I found 

 it's mean breadth to be 1000 toises. The following is a com- 

 parative table of some of the great rivers of the New World, 

 calculating the length of the eourse, according to the most re- 

 cent maps, and adding one third for the sinuosities : 



The Amazon, 980 leagues, of 20 to a degree. 



The Missisippi, 560 leagues, in going up by the principal 

 branch to the Chippeway, but 815 leagues in going up to the 

 sources of the Missouri. 



The Rio de la Plata, 530 leagues, in going up by the Rio 

 Paraguay. 



The Oroonoko, the known part 420 leagues. (The Indus 

 ha^a course of 510, and the Ganges of 426 leagues.) 



