587 



the level of the sea ; yet we cannot easily ad- 

 mit, that the clouds above the coast of Cum an a 

 and the island of Margaretta sustain themselves 

 at so small a height. The storm which growled 

 around the top of Duida did not descend into 

 the valley of the Oroonoko ; we did not in ge- 

 neral observe in that valley those strong electric 

 explosions, which almost every night, during 

 the rainy season, alarm the traveller along the 

 Rio Magdalena, in going up from Carthagena 

 to Honda. It would seem as if the storms in a 

 flat country followed the furrow or bed of a 

 large river more regularly than in a country 

 unequally studded with mountains, and exhi- 

 biting various branchings of lateral valleys. We 

 repeatedly examined the temperature of the water 

 of the Oroonoko at it's surface, while the ther- 



of a plain, you are sure of obtaining results, that are inde- 

 pendant of the local effect which has just been mentioned. 

 Gay-Lussac and Biot, in their aerostatic ascents, found the 

 inferior limit of the clouds above Paris at six hundred toises, 

 in the great heats of summer. The fogs, in which you are 

 so frequently enveloped at Xalapa, on the eastern declivity 

 of the Cordillera of Mexico, led me to admit formerly, that 

 the mean height of the clouds above Vera Cruz was only 

 seven hundred toises j but the proximity of woody and 

 damp mountains, the radiation from the soil and leaves dur- 

 ing the night in a serene sky, and the electric conductibility 

 of the rock, render those conclusions somewhat uncertain, 

 that are drawn from the measure of the height of clouds 

 adhering to mountains. 



