10 Professor Apjohn upon a new Method of 



by the second of the above experiments ; and as both air and gas are dry, and 

 must have been, with at least a high degree of probability, proportionally affected 

 by variations of pressure, the precise influence of these, about which, indeed, phi- 

 losophers are not agreed, do not require to be taken into consideration, nor is 

 there any thing farther necessary for rendering the result thus obtained strictly 

 comparable with those of other experiments, than to reduce it by the Rule of 

 Three to what it would be if the specific heat of air were .267» the number by 

 which it is usually represented in books, at the mean altitude of the barometer. I 

 shall now, before proceeding to the tabular view of my experiments, and their 

 results, exemplify the method of calculation which has been just described. 



On the 4th of August, 1835, the following observations were made, first on 

 hydrogen, and subsequently upon air. 



t t' d p 



Hydrogen . . . 68 48 20 30.114. 



Air 68 43 25 30.114. 



By applying to these results the equation a =r -~- x — > we get 



Specific heat of air r= .2767 '=■ c. 



Approximate specific heat of gas . zz .409 = «• 



But the gas, upon analysis, was found to contain 5 per cent, of air. Hence the 

 specific heat of the hydrogen supposed pure, as deduced from the equation 



{a—c)n 

 a—a-\-Y\ 



(lOO-w)' 



becomes, .4151. And as, .2767 : .4151 : : 2670 : .4005, the specific heat of 

 hydrogen compared to that of air under a pressure of 30, when water Is repre- 

 sented by unity, or what amounts to the same, when air is .267- 



The following tables include the particulars of the first series of experiments 

 I performed on this interesting subject. In order that they may be perfectly un- 

 derstood, the reader should recollect that t Is the temperature of the dry, and t' of 

 the wet thermometer ; that dz=.t — ^ ; that j> is the existing pressure, as measured 

 by the barometer ; a the specific heat of gas, as deduced from the formula 



_ p/ 30 



