1819.] P>'°f- Leslie on Heat and Climate. 25 



d.* It appears, therefore, that the formula which we derived from 

 experiments made on the collapse of air in the receiver of an 

 air-pump, will apply likewise to the gradation of cold in ascend- 

 ing the atmosphere. The density at the surface being reckoned 

 the unit, let its diminution at the given elevation be denoted by 

 x, then the diminution of temperature on the ascent will be in 



degrees of Fahrenheit = [ * ] 125 x, which may, perhaps, 



2 j.f 



This formula corresponds, as well as might be expected, with 

 the very few facts which I have been able to collect. Dr. 

 Hunter reports, that at the elevation of 1400 yards on the Blue 

 Mountains, in the island of Jamaica, the springs were 18° 

 colder than on the level of the shore. The difference of density, 

 at such an altitude and such a climate, may be computed to be 



•137, whence ( 4 3 * J" g ~ '^ ' ) 1000 = 19°, which differs 



only one degree from observation. Lord Mulgrave mentions, in 

 his voyage, that the air at the top of a mountain in Spitz- 

 bergen, 1503 feet high, was 8° colder than at the bottom. But 

 the decrease of density may be estimated at - 057 ; whence 



1000 ( 3 * _ 3 3 2~ x -057 ) = 7 ' 5 °- I perceived the same agreement 



in some trials which I lately made on the springs in Fifeshire. I 

 must confess, however, that though I am fully convinced that 

 the formula expresses the true law of the diminution of tem- 

 perature at different elevations, I still entertain a suspicion that 

 the coefficient 1000 requires some correction. The glass re- 

 ceivers which I employed were not so much disproportioned as 

 I should have desired ; and the difference is so minute, on 

 which the analogy turns, that I must have hesitated to publish 

 the result without an appeal to observation. 



Resuming the first formula ( _ x ) 125 x, the part — ~ * x 



may be regarded equal to unit in small heights above the 

 surface ; and since the diminution of density is in this case 

 nearly proportional to the ascent, the gradation of cold must, 



• If greater accuracy be required, let this equation be converted into an analogy 

 2000 + d : 2000 :: 2000 : 2000 - t, then 2000 + d: d ;: d:d - J ; whence d - J = 

 d 



20° 



2000 , . When d = 20°, the correction will be = —— ; and when d «= 50°. 

 — — +1 51 ' 



a 



50° 

 there should be an abatement of -—- , or somewhat more than a degree. 



+ The formula, which I have finally adopted from more perfect experiments, 



——J, in which the expression is a little modified, and the coefficient 



er multiplier diminished. 



