3 7 6 On the Determination of the [Nov. 



is easy to see that the gazometers thus modified would act exactly 

 as before. 



It was sufficient to make these experiments on a single gas to 

 be enabled to draw conclusions applicable to all the rest. We 

 made choice of atmospheric air, as the most convenient, and the 

 least likely to be altered. This enabled us to remove tbe 

 bladders, and the globular vessels that contained them, and to 

 establish an immediate communication between the gazometers 

 and the calorimeter ; and hence to dispense with lutes, which 

 we should have found it difficult to make sufficiently tight to 

 re.-ist so great a pressure. 



We will satisfy ourselves with giving the result of a single 

 experiment, made with great care, because it was conducted in 

 every respect in the same manner as all the preceding ones. 



The gazometers were so disposed as to give in ten minutes an 

 uniform current of 1898 cubic inches of gas (at the temperature 

 of 41°, the barometer standing at 29*634 inches, bi sides a 

 pressure of 11*29 feet of water, making in all a pressure equal 

 to 41 654 inches of mercury). The experiment was conducted 

 in a manner perfectly analogous to all those that were exhibited 

 in the first table. The result was, that this current of com- 

 pressed air maintained the temperature of the calorimeter at 



28 SI 1° above that of the surrounding air, making an allowance 

 for the heat communicated by the gas tube : but this air entered 

 the calorimeter at the temperature of 206" 42°, and came out of 

 it at that of 73*098° ; so that it was deprived of 133-322° of 

 heat during its passage through the calorimeter. If we reduce 

 by calculation the results of this experiment to what they would 

 have been, supposing it the same with the first experiment made 

 on air in the first table, we find that under a pressure of 41*654 

 inches of mercury, a current of 2196*4 cubic inches of air, by 

 losing 130347° of heat, raise the temperature of the calorimeter 

 33*666°; while we see by the table that under a pressure of 



29 154 inches, the same current, in the same circumstances, 

 only raise the temperature of the calorimeter 27'76l°. The 

 ratio between these two numbers being that of the specific heats 

 of the gases, it follows that if the specific heat of air, under the 

 pressure of 29*154 inches, be 1*000 ; that of the same volume 

 of air, under the pressure of 41*6*54 inches, will be 1*2127. 



We repeated these two experiments with the greatest possible 

 care, one with air in its usual state, the other with air subjected 

 to the artificial pressure. They have given us, after having made 

 all the calculations, the ratio between the specific heat of air 

 under the pressure of 29*154 inches, to that of air under the 

 pi^sure of 41*654 inches, 1 : 12665. Takintr the mean of 

 these two determinations, we obtain the ratio of 1 : 1*2396. 



Hence it is evident that the specific heats of equal volumes of 



