1S13.] Specific Heat of the different Gases. 431 



vat ons on the rate of cooling, a term, whose excess above the 

 temperature of the surrounding air, supposed likewise increased 

 by 1'5° (to render the circumstances similar) is nearly the same. 

 We lind 74'165°. Neglecting the terms immediately following, 

 let us take the on£ observed 20 minutes after, namely, 68-91,5. 

 If we subtract it from the other term, we get 5-252°, a quantity 

 which expresses the cooling of the calorimeter in 20 minutes. 

 We find by a proportion that if the excess of the temperature of 

 the calorimeter above that of the surrounding air, augmented 

 always by 4-5°, had been in the case which we have chosen 

 exactly 23-321, the cooling in 20 minutes would have been 

 5' 1I>7°, and of course this excess would have been found at the 

 end of 20 minutes equal to 23'12i°. 



This diminution of temperature, given by experiment, is not 

 the same with that which the calorimeter would have'experienced 

 in the same time, if it had preserved the rate of cooling which 

 it had at first. It would have been more considerable ; but wc 

 may determine it by means of the preceding data, and by making 

 use of the following formula : — 



S = A log. hyp. j * 



in which S signifies the quantity wanted ; A, the excess of the 

 temperature of the calorimeter above that of the surrounding 

 air in the first instant of the experiment ; and B, that excess at 

 the end of 20 minutes. 



If we apply this formula to the numbers stated above, we have 

 A = 28-321°, and B = 23-124°, which gives us S = 5-7411°. 

 We find S = 5*7760°, if we make use of the second series of 

 observations of the cooling of the calorimeter. The mean of 



s 

 these two values gives us S = 5*7586°, and of course - = 



D' 



2'8793" = the quantity of heat lost in ten minutes, supposing 

 it to preserve always the initial rate of cooling. This is the 

 {lumber which we employed to calculate the relation between the 

 specific heat of air and water. 



§ III.— Third Method. 

 The experiments which we made, according to the process 

 contrived by Count Rumfbrd, gives us a direct and mote ,-imple 

 method of determining the specific heat of air compared with 

 that of water. We we by the table of these experiments that, 

 under a pressure of l'9-922 inches of mercury, and at the tem- 

 pciaturc of 32°, 6077'5 cubic inches of atmospherical air, or 

 1672-9 grains, by undergoing h diminution of 164*192 of 

 temperature, raised the temperature of the calorimeter 7 • But 



of (bii formula. 



