36 Specific Heats of Gases at constant Volume. [Nov. 22 



These, it is seen, afford a result above that theoretically assigned to 

 air at constant volume (0'1684). They differ too somewhat from some 

 experiments made by the first-described method, are somewhat lower 

 than their mean, but the consistency displayed throughout in the 

 thirteen experiments given, especially in Tables I and II, leads me to 

 give these numbers as probably a close approximation to the true 

 value. One point is at any rate brought out clearly, that is, that the 

 surmise that the specific heat of a gas at constant volume was a quan- 

 tity independent of pressure a surmise based partly on the constancy 

 of the specific heat at constant pressure would appear to be correct. 

 The values in the three tables, calculated simply on the weights of 

 gas dealt with in each set of experiments, show results quite inde- 

 pendent of the great variations of pressure and density obtaining, 

 the weight of condensation simply falling off with the decrease in the 

 weight of gas, till in the third table w is beginning to feel the errors 

 incidental to the considerable mass of the spheres and to give more 

 variable results. I have prepared very thin light spheres with a view to 

 continue the experiments to lower pressures with less danger of error. 



The cause of the excess in the value obtained above the theoretical 

 is not apparent, especially in view of the independence of pressure 

 displayed. The experiments embodied in the three tables were made 

 indeed upon the one sample of air some being liberated after the 

 first five experiments, and so on but this had been dried through 

 three calcium chloride tubes and two large U -tubes of phosphorus 

 pentoxide before passing into the pump. Between the pump and the 

 sphere it passed through a brass tube stuffed with asbestos which 

 had been previously heated to redness. The object of this is to guard 

 against oil being carried from the pump into the sphere. 



My first determination of the specific heat of air at constant volume 

 was effected on the 13th of April of this present year. It was 

 made by the method described in the beginning of this note, at a 

 pressure somewhat higher than that at which the experiments of 

 Table I were effected. This experiment gave as a result the specific 

 heat of air to be 017565. 



Note. October 18. 



Subsequent more extended experiments have shown me that this 

 condition was not absolutely fulfilled. A small reduction of the values 

 recorded for the specific heat of air is necessary on this account, but 

 insufficient to affect any remarks made in this note. Successive 

 experiments on the empty spheres, I may observe, are sufficiently 

 consistent one with another to warrant the assumption that the values 

 recorded by me are not probably affected to the extent of one per 

 cent, by errors on the calorific capacities of the spheres. 



