RICHARDS. — CHANGING HEAT CAPACITY. 309 



Co"" ...,,, ^ rK, - K 



A=^RT\n ^'' • • • + U- T'T^ "^" ^^^ dT-\- H T .* 



In this equation there is only one unfamiliar symbol H, which signi- 

 fies a sum of several integration constants. Since the last two terms are 

 hard to determine practically in many cases, we may combine them and 

 write 



A^R rin^V^ + U-f{K, T). 

 c ... 



In words, the available energy of a reaction is equal to the sum of the 

 concentration effect and the heat evolved, minus a quantity which is 

 partially dependent upon change of heat capacity. 



This is qualitatively the outcome of the reasoning based purely upon 

 experiment. In one respect the experimental conclusion is more detailed 

 than the theoretical, in that the former indicates the complete elimination 

 of the last term where the heat capacity of the system does not change 

 during the reaction, while the theoretical equation still contains an un- 

 known constant of integration in this case. 



A clearer insight into the matter is possibly to be obtained by writing 

 this equation in the following form, in which C represents the true 

 although as yet unknown value of the attracting affinity, and p is the 

 work needed to change the heat capacity. 



A.8 Lewis has already pointed out from the deductive side, the equa- 

 tion of Nernst for the single potential difference of a reversible electrode 

 should obviously be remodelled on the basis of these conclusions. The 

 Nernst solution-pressure P must include not only the concentration 

 effect, but also the effect of the free energy due to affinity of the 

 solid electrode for the solution. -f- 



The latter is really by far the most important part of the reaction. 

 For the present purpose it is best not to include this effect in the vague 

 and comprehensive P, but to separate the latter into its two important 

 parts, thus 



P P P 



7i7r€ = R T\n- = RT\n^ = R Tlu^ + RTlnz. 



* G. N. Lewis, These Proceedings, 35, 1 (1899), Z. phys. Chem., 32, 364. 



t Lewis introduces also a term to cover the nature of tlie solvint in the cell. 

 While this may be necessary, it seems probable that the value of A includes the 

 effect of chani^iiig solvent. 



