Temperature on the Specific Heat of Aniline. 63 



Let the potential-difference of a Clark cell be e, and let n 

 be the number of cells used. 

 Then 



JRiCSiM + Wj) l rv l " n dt ' 

 hence 



J n l 



\dtjn* J U^JSjM + ?(;,) 



71" 



Let us assume for the time that # N is independent of n, 

 then, when ^, = ^ N , we get 



tf# N 1 e* 



dt n* ' JR^SjM + Wi) ' 

 hence 



rffty _1__ <70k; _L-^ — -&C 

 dt n^ dt 1%2 2 dt n 3 2 



where n l5 w 3 , t? 3 &c. are the number of cells used in each 

 case. 



If, therefore, we plot the curves (which in this case are 

 straight lines since the variables, viz., p(O l — o ) and R 1? S 1? 

 and Wi may for small changes of temperature be considered 

 as linear functions of iy see paper J, pp. 442-448), which 



we obtain by taking -~ X — ^ as ordinate, and 6\ as abscissa, 



they will intersect where the abscissa is 0-$, and there only, 



(10 

 and since the value of —j^ can be experimentally determined 



for different values of 1? the observations themselves can 



be made to give the value of # N , and we can thus find 



e 2 

 T ,» /c , ,, - 1 the rise due to the electrical supply alone. 



A small deviation from the true value of &s is of little 



consequence, for the resulting value of —^ x — 2 will only 



■ft** n d0$ 1 , ,, ... P\0\ — 0~S) 1 



differ from —rr X — l by the quantity § , and since 



p is small as compared with the other magnitudes, 0i — 0n 

 must be considerable before the error becomes appreciable. 



As this is an important point, I will select one of the three- 

 cell experiments (No. 37) in order to show the relative mag- 

 nitude of the various quantities, and the probable limit of 

 error arising from the assumption that 0^ is independent of n. 

 The differences in temperature are expressed in terms of the 



