Electromotive Forces from Thermochemical Data. 217 



forms one of the most serious difficulties in the experimental 

 work, for if the resistance change with change of direction of 

 the current, the difference of heating-effects might be due, 

 at least in certain cases, to such change rather than to the 

 valueofE— Ea. 



This difficulty was endeavoured to be met by using a high 

 resistance galvanometer on a circuit of several thousand ohms 

 as a measurer of the difference of potential between the poles 

 of the experimental cell, when on open circuit and when 

 direct or inverse currents are passing through it. 



Taking E as the E.M.F. of the external battery and R its 

 resistance, and E^ the E.M.F. of the experimental cell, and 

 r, its resistance, and letting r^ denote the total external re- 

 sistance, including that of the current-galvanometer, we get 

 for the difference of potentials between the poles of the 

 experimental cell : 



V= '^ T, — — ^with du'ect currents, 



U + r, + ri, ' 



and 



„ E,(R-\-r,.)+Er. .,, . , 



V, = — ^-^=i — L with inverse currents. 



' R + r, + r^, 



Since, however, the value of Ey (or indeed E) is subject to 

 possible variation due to polarization, the above method fails 

 to distinguish between change of difference of potential due 

 to change of resistance and that due to polarization ; and 

 although the latter, if present, may be usually detected by 

 suddenly interrupting the current and noting the value of E^, 

 the difficulty still remains to some extent and is liable to 

 vitiate the results. 



From the causes just indicated it wiU be readily seen that 

 extremely few cells are suited for thermal experiments when 

 currents of moderate strength have to be sent through them 

 in alternate directions. Consequently a large number of 

 voltaic combinations were tried and abandoned as unsatis- 

 factory, for some of the following reasons : — Certain cells are 

 found to give marked variation in their internal resistance 

 with different direction of current, commonly ovving to the 

 sparing solubility of at least one of the salts formed and the 

 consequent coating of one or both plates with insoluble and 

 badly conducting films {ex.gr. cells involving the sulphate, 

 chloride, or bromide of lead, mercury, or silver, three typi- 

 cally exceptional metals). In other cases cells are rendered 

 useless by the imperfect reversibility of the chemical change ; 

 as in cases in which it is attempted to deposit a metal which 

 can replace hydrogen from water at all readily {ex. gr. mag- 



