[ 5^ 1 



fuppofition on which it is grounded. On examining a confider- 

 able number of obfervations of this fort, I find that the differences 

 of heat at different elevations are to the differences of the 

 logarithms of the mercurial heights at thofe elevations, very 

 nearly, as i6o to i ; and the difference of the logarithms may be 

 found fufiiciently near, by dividing the elevation in feet by 

 60000. If then, in the inflance above given, the S. wind reached 

 the height of 5000 feet, the difference of the logarithms fhould 

 be aoooo =>o83333, which multiplied into 160, gives i3°,33 for 

 the difference of temperature at the furface of the earth and at 

 that height j and the former being 49", the latter mufl have been 

 35°,67, and confequently the mean temperature of the fouthern 

 air 42,33. Again, fuppofing the N. wind to have obtained the 

 fame height, the temperature at the furface of the earth being 

 30°, that at the height of 5000 feet fhould be i6°,67, and the 

 mean temperature 23,33; o^' ''^ round numbers, the mean 

 temperature of the fouthern current was 42°, and that of the 

 northern 23°, and the difference of both temperatures 19°. 

 Now 5000 feet— 60000 inches ; and by General Roy's expcri- " 

 ments*, it appears that 1000 parts of air at the temperature of 

 42° lofe about 40 of their bulk by 19*" of cold, and confequently 

 60000 fhould lofe 2400. But as in this cafe the height of the 

 column of air is fuppofed to be the fame after condenfation as before 

 it, its mafs or denfity mufl be encreafed by the addition of 

 2400 inches, which when the barometer is at 29,89, and the 

 thermometer at 23", weigh 762,36 grains ; and as one-tenth of 

 an inch of ?, even at the temperature of 62% weighs 344,32 



• Phil. Tranf. 1777, Part 11. 



grains, 



