14< Col. R. Wright's Meteorological Observations made 



atmosphere, so that several years frequently pass without any 

 regular fall of rain. The vfcinity of the lake, no doubt, acts 

 sli^rhtly as a refrigerant; but the city is built on the border of 

 its"outiet to tlie sea, where it is both narrowest and shallowest, 

 anil is consequently heated nearly to the temperature of the 

 incumbent atmosphere. Add to this, the small sandy eleva- 

 tions to the north, which intercept the partial effect of the sea- 

 l)reezes, so that they are scarcely felt, except in the months of 

 December and January, when the thermometer sometimes sinks 

 to 73°; yet the medium even of these two months is not less than 

 Sl^; while that of La Guayra from November to December at 

 noon, is, according to Humboldt, 75°-8, and at night 70°-9. 

 (Personal Narrative, vol, iii. p. 387.) Rio Hacha is situated 

 on a sandy beach ; the sea-breeze blows with such violence that 

 boats can scarcely land between ten in the morning and four 

 in the afternoon. These winds, however, sweeping over the 

 hot plains of Coro and Maracaybo, have but a partial effect 

 in lowering the temperature, the annual mean of which is l°-98 

 less than that of Maracaybo, I never saw the thermometer 

 lower than 75°, nor above 89°. In Santa Marta the average 

 of the coolest months is 82°'25. The thermometer, however, 

 never rose during my residence there above 87°. The soil is 

 sandv, and the city is surrounded by bare rocky heights to the 

 north and south, which counterpoise the cooling influence of 

 the Sierra nevacla (snowy mountains), from which it is but a 

 few leagues distant. The temperature of Barranquilla, a vil- 

 lage situated on the river Magdalena, about eighteen miles from 

 its mouth, is nearly the same with that of Santa Marta; for if, 

 on the one hand, the air is refreshed by the evaporation from 

 a damp soil covered with luxuriant forests and the vicinity of 

 a large river, on the other, it is beyond the reach of the sea- 

 breeze, and the influence of the mountains which operate in 

 Santa Marta. The annual mean is 82°'20. That of Cumana 

 is, according to Humboldt, 81°. The breezes which sweep 

 from the gulf of Paria over the wooded Brigantine chain pro- 

 bably contribute to lower the temperature. 



We iiave thus, on a calculation of six points on the Atlantic 

 coast of Colombia, a mean annual temperature of 82°*S6*. 

 The shores of the Pacific, as far as the latitude of Payta, are 

 subjected to other influences, being almost entirely covered by 

 (lani]> luxuriant forests; while the ocean itself is cooled, as 

 Humboidt observes, by the winds which blow continually 



• I have not incliulcd Cartagena, bccai:se the number of observations 

 is perhaps too limited to draw a conchision as to the yearly tem[)crature. 

 If we take them into the calculation, the annual mean would l)e 8.'5°*8G, 

 which is probably too high. 



