ON CLIMATES AND WINDS. 547 



reason, the most elevated tracts of land, which are the most prominent, and 

 the most exposed to the effects of the atmosphere, are always colder than 

 situations nearer the level of the sea. The northern hemisphere is somewhat 

 warmer than the southern, perhaps because of the greater proportion of 

 land that it contains, and also in some measure on account of the greater 

 length of its summer than that of the southern : for although, as it was 

 long ago observed by Simpson, the different distance of the sun compen- 

 sates precisely for the different velocity of the earth in its orbit, with respect 

 to the whole quantity of heat received on either side of the equinoctial 

 points, yet Mr. Prevost has shown, that in all probability the same quan- 

 tity of heat must produce a greater effect when it is more slowly applied ; 

 because the portion lost by radiation from the heated body is greater, as 

 the temperature is higher. Since, therefore, on account of the eccentricity 

 of the earth's orbit, the north pole is turned towards the sun 7 or 8 days 

 longer than the south pole, the northern winters must be milder than the 

 southern : yet the southern summers, though shorter, ought to be some- 

 what warmer than the northern : but in fact they are colder, partly per- 

 haps from the much greater proportion of sea, which in some degree 

 equalises the temperature, and partly for other reasons. The compara- 

 tive intensity of southern summer and winter is not exactly known ; 

 but in the island of New Georgia the summer is said to be extremely 

 cold. 



The northern ice extends about 9 from the pole : the southern 18 

 or 20 ; in some parts even 30 ; and floating ice has occasionally been 

 found in both hemispheres as far as 40 from the poles, and sometimes, as 

 it has been said, even in latitude 41 or 42. Between 54 and 60 south 

 latitude, the snow lies on the ground, at the sea side, throughout the sum- 

 mer. The line of perpetual congelation is three miles above the surface at 

 the equator, where the mean heat is 84 ; at Teneriffe, in latitude 28, two 

 miles ; in the latitude of London, a little more than a mile ; and in latitude 

 80 north, only 1200 feet. At the pole, according to the analogy deduced 

 by Mr. Kirwan,* from a comparison of various observations, the mean 

 temperature should be 31. In London the mean temperature is 50 ; at 

 Rome and at Montpelier, a little more than 60 ; in the island of Madeira, 

 70 ; and in Jamaica, 80.f 



There are frequently some local causes of heat and cold which are inde- 

 pendent of the sun's immediate action. Thus, it has been observed, that 

 when the weather has been clear, and a cloud passes over the place of obser- 

 vation, the thermometer frequently rises a degree or two almost instanta- 

 neously. This has been partly explained by considering the cloud as a 

 vesture, preventing the escape of the heat which is always radiating from 

 the earth, and reflecting it back to the surface : the cloud may also have 

 been lately condensed, and may itself be of a higher temperature than 

 the earth. Mr. SixJ has observed that in clear weather, the air is usually 

 some degrees colder at night, and warmer by day, close to the ground, 



* Anjgstimate of the Temperature of different Latitudes, Lond. 1787. 

 f O;i Isothermal Lines, see Humboldt, Fragments Asiatiques, ii. 398. Mem. 

 d'Arcueil, Hi. 462. + Ph. Tr. 1784, p. 428 ; 1788, p. 103. 



2 N 2 



