Ill 



of vapour corresponding to the same temperature as given in 

 the two other tables. Now, the mean of the temperatures is 

 61°*63, the quotient got by dividing their sum by twenty. 

 But the corresponding mean value of f, in column 2, 

 must be differently calculated, seeing that the temperature 

 and the corresponding tensions of the vapour augment at a 

 very different rate. For temperatures, in fact, in arithmetic 

 progression, the corresponding tensions are in geometric 

 progression ; and, although this is well known to be but an 

 approximate law, it may be considered as rigorously true 

 for the limited range of temperature within which my expe- 

 riments have been made. To calculate, therefore, the mean 

 force of vapour, as deducible from the numbers in column 

 2, and which must correspond to the temperature 61°*63, 

 it is only necessary to add together the logarithms of the 

 numbers in this column, and divide their sum by twenty, and 

 the quotient will be the logarithm of the mean. When this 

 process is gone through, the mean logarithm is found to be 

 •73699, and the corresponding number '54575. The follow- 

 ing, therefore, are the tensions of aqueous vapour at 6l°'63, 

 as deduced from my experiments, and as extracted from the 

 tables of Dalton and Kamtz. 



My Experiments. Dalton. Kiimtz. 



61°*63, . . -5457 . . . -5523 . . -5349 

 Difference between Dal ton's number and mine, n: -|- .0066. 

 Difference between Dalton's number and that of Kamtz, 

 = + .0174. 



It thus appears, that the result at which I have arrived 

 is somewhat less than the Daltonian number, but consider- 

 ably greater than that given by Kamtz ; and that, therefore, 

 my experiments, as far as they have been discussed, give at 

 least a prima facie countenance to the opinion, that the 

 values of the elastic force of aqueous vapour, as given by 

 the latter philosopher, are, at and about ()1°'63, below the 

 truth. 



