406 Experimental Researches on the Depressions [July, 



the indefinitely small decrements of caloric from evaporation are 

 balanced by the indefinitely small increments arising from conduction 

 and radiation in the equally small moments of time." Now as Messrs. 

 Dulong and Petit have shewn that the rate at which a body cooled be- 

 low the temperature of the air (by conduction and radiation) reacquires 

 heat, is proportional not to the simple difference of temperature, but to 

 that difference raked to the 1.233 power; hence it should follow that 

 the amount of evaporation should increase in the same ratio ; " but," 

 says he (page 191), " how determine the rate of evaporation ? One of 

 the most striking phenomena of evaporation is the cold produced by it ; 

 the consequence of the absorption of heat attending the conversion of 

 water into vapour. This depression of temperature must evidently 

 be as the evaporation ; or rather the momentary depression will be 

 in proportion to the rapidity of the evaporation. The momentary 

 depression is equal to the momentary increment of heat which would 

 take place were the cooling power of evaporation suspended, and the 

 moistened bulb thermometer allowed to assume the temperature of 

 the air. This is known to be as the 1.233 power of the total depres- 

 sion : the evaporation will then be as the 1.233 of the depression." 

 But the evaporation is (according to Dalton), as the tension of the 

 evaporating surface minus the tension of the vapour in the air (= 

 in dry air:) then finally this tension will be as the 1.233 power of 

 the depression : or 



1.233 



dm = ^/f'—f" 

 m being a co-efficient depending on the latent heat of air and the 

 ratio of the evaporation to the weight and surface necessary to produce 

 a fall of one degree ; which Captain Herbert deduced from the experi- 

 ments made at Benares. The complete formula, at 30 inches, for dry 

 air becoming 



1.233 



6.056 

 in which L (proportion of mass of water to the vapour required to be 

 evaporated to produce a fall of 1°) is derived from a table published 

 in the Oriental Magazine, September 1827 ; it varies from 898 at 40° 

 to 1005 at 90° and 1250 at 1800. The divisor 6.056 would require to 

 be diminished to 5.4 to suit the present experiments, but neither 

 would the formula then agree so well as the more simple one of 

 Leslie and others. The fact is that the experimental curve is of so 

 simple a nature, that; any geometric series of moderate divergence may 

 within limits be accommodated to it by proper co-efficients : thus my 



