SECT. XXV. AVERAGE ANNUAL TEMPERATURE. 239 



ture, they would be nearly parallel to the equator between the 

 tropics, and would become more and more irregular and sinuous 

 towards the poles. These are called isogeothermal lines. A 

 variety of local circumstances disturb their parallelism, even 

 between the tropics. 



The temperature of the ground at the equator is lower on the 

 coasts and islands than in the interior of continents ; the warmest 

 part is in the interior of Africa ; but it is obviously affected by 

 the nature of the soil, especially if it be volcanic. 



Much has been done to ascertain the manner in which heat 

 is distributed over the surface of our planet, and the varia- 

 tions of climate, which, in a general view, mean every change 

 of the atmosphere, such as of temperature, humidity, varia- 

 tions of barometric pressure, purity of air, the serenity of the 

 heavens, the effects of winds, and electric tension. Tem- 

 perature depends upon the property which all bodies possess, 

 more or less, of perpetually absorbing and emitting or radiating 

 heat. When the interchange is equal, the temperature of a body 

 remains the same ; but, when the radiation exceeds the absorp- 

 tion, it becomes colder, and vice versa. In order to determine 

 the distribution of heat over the surface of the earth, it is neces- 

 sary to find a standard by which the temperature in different 

 latitudes may be compared. For that purpose it is requisite to 

 ascertain, by experiment, the mean temperature of the day, of 

 the month, and of the year, at as many places as possible through- 

 out the earth. The annual average temperature may be found 

 by adding the mean temperatures of all the months in the year, 

 and dividing the sum by twelve. The average of ten or fifteen 

 years will give it approximately ; for, although the temperature 

 in any place maybe subject to very great variations, yet it never 

 deviates more than a few degrees from its mean state, which con- 

 sequently offers a good standard of comparison. As a standard, 

 however, much greater accuracy is required. 



If climate depended solely upon the heat of the sun, all places 

 having the same latitude would have the same mean annual 

 temperature. The motion of the sun in the ecliptic, indeed, 

 occasions perpetual variations in the length of the day, and in the 

 direction of the rays with regard to the earth ; yet, as the cause 

 is periodic, the mean annual temperature from the sun's motion 

 alone must be constant in each parallel of latitude; for it is 



