230 



on which this rests is the only tenable one. The first equation is 

 of this form : 



Log.F 

 t + I Log. m 



whence the experiments of Dr Dalton are accurately represented 

 by the expression 



(12.) 1±J^= 1.1320^^' 



^ ' 337 



But it is remarkable that, if one series of experiments below 

 212" be taken, and any other above 212^, the expression which ac- 

 curately represents the one will widely deviate from the other. 

 So that, though a formula of the same nature represent the expe- 

 riments of the French Academy and of the Franklin Institute very 

 accurately, it requires diflfereut constats. If we average Dr Dal- 

 ton's and Dr Ure's below 21 2^ and those of the Academy and 

 Institute above that point, e find they coincide more nearly with 

 the following formula than with each other : 



Log.F 



333= 

 From the whole investigation, the authof arrives at the follow- 

 ing conclusions : — (1.) That the discrepancies between the most 

 recent and extensive experiments we possess are so great, as to 

 leave this branch of experimental science in a disgraceful state of 

 uncertainty, and that it is desirable that those who are engaged 

 in experiments on heat, would revise this useful and interest- 

 ing branch of inquiry ; and it is suggested that, for this pur- 

 pose, the same apparatus and methods of experiment which are em- 

 ployed in the experiments below atmospheric pressure, should also 

 be those which are employed at high temperatures above atmosphe- 

 ric pressure. (2.) That the most accurate experiments we are in 

 possession of at pressures below 212"^ are Dr Dalton's recent ones, 

 which accurately coincide with the hypothesis that the intervals of 

 the common scale are to those of true temperature, in a geometri- 

 cal register, reckoned from the zero point of the mercurial ther- 

 mometer 175° below the zero of Fahrenheit, and of which the com- 

 mon ratio is 1.1320, the corresponding elastic forces forming the 

 logarithmic progression to the true temperatures, whose index is 

 2.602. That the mean of the experiments of Dr Dalton, Dr lire, 

 the French Academy, and the Franklin Institute, give 1.11401 as 

 the ratio of progression of the temperatures, 2. being the index 



