789 



The most violent showers produced parti- 

 ally 14 lines of water, which falls in drops of 

 an enormous size ; and this characterizes the 

 small rains of the tropics, that they fall in drops 

 which remain at a great distance from each 

 other. There have been years (1798 and 1799), 

 when during nine months, from December to 

 September, the rains did not yield two inches 

 of water. In the New Continent, the drought 

 of Cumana, PuntaAraya, and the island of Mar- 

 guerita, can be compared only with the pro- 

 vince of Ciara in Brazil, where sometimes 

 (1792-1796) it does not rain during several 

 years. (Corogr. Bras, ii, p. 221.) The vege- 

 tation, notwithstanding the drought, is fresh at 

 Cumana, for instance, near the Chora de Ca- 

 yuchinos. The dew is almost null ; the little 

 water that fails at Cumana descends in showers 

 with extraordinary rapidity ; these showers last 

 in general but from fifteen to twenty minutes. 

 I saw 4 J lines at the maximum, fall in six mi- 

 nutes. All my measurements were made in 

 cylindric vases, and in such a manner that the 

 evaporation could not lead to error. During 

 the great storm of September 16th, 1802, at 

 Cumana, I placed two cylindric ombrometers, at 

 heights which differed only twenty-two feet per- 

 pendicularly. It rained with violence from 

 3 h 25' to 4 h 5' ; I found in the most elevated 

 ombrometer, 6^ lines of water, and in the 



