110 STEPPES AND DESERTS. 



of heat in America ; it will of course be understood that the ques- 

 tion respects the general hygrometric state of the atmosphere, and 

 the temperature of the New Continent as a whole. Single districts, 

 such as the island of Margarita, the Coasts of Cumana and Coro, are 

 as hot and as dry as any part of Africa. It must also be remarked 

 that the maximum of heat at certain hours of a summer's day has 

 been found, on a series of years, to be almost equal at very different 

 parts of the earth's surface, on the Neva, the Senegal, the Ganges, 

 and the Orinoco; being approximately between 27 and 32 Reau- 

 mur (93 and 104 Fahrenheit), and generally not higher, provid- 

 ing the observation be made in the shade, at a distance from all solid 

 bodies which could radiate heat to the thermometer, not in an air 

 filled with hot particles of dust or sand, and not with spirit thermo- 

 meters, which absorb the light. It is probably to fine grains of sand 

 floating in the air, and forming centres of radiant heat, that we must 

 ascribe the dreadful temperature of 40 to 44. 8 Reaumur (122 to 

 133 Fah.) in the shade, to which my unhappy friend Ritchie, who 

 perished there, and Captain Lyon, were exposed for weeks in the 

 Oasis of Mourzouk. The most remarkable instance of very high 

 temperature, in an air probably free from dust, has been recorded by 

 an observer who knew well how to place and to correct all his in- 

 struments with the greatest degree of accuracy. Riippell found 

 37.6 Reaumur (110. 6. Fahrenheit), at Ambukol in Abyssinia, 

 with a clouded sky, strong south-west wind, and an approaching 

 thunderstorm. The mean annual temperature of the tropics, or of 

 the proper climate of palms, is, on land, between 20. 5 and 23. 8 

 Reaumur (or 78.2 and 85. 5 Fahrenheit), without any considerable 

 difference between the observations collected in Senegal, Pondicherry, 

 and Surinam. (Humboldt, Me"moire surles lignes isothermes, 1817, 

 p. 54. Asie Centrale, t. iii. Mahlmann, Table iv.) 



.The great coolness, I might almost say cold, which prevails for a 

 considerable part of the year within the tropics on the coast of Peru, 

 causing the thermometer to sink to 12 Reaumur (59 Fahrenheit), 

 is, as I have noticed elsewhere, by no means to be ascribed to the 

 vicinity of the snow-covered Andes, but rather to the fogs (garua) 

 which veil the solar disk, and to a cold sea current which, commenc- 

 ing in the antarctic regions and coming from the south-west, strikes 



