Electromotive Forceafrom Thermochemical Data. 221 



data give for zinc-mercury nitrate cells Efc= 1-673 volt, the 

 experimental value being 1'5 volt; and for zinc-mercury 

 chloride cells E;i= 1*028, the experimental value being 1-043 

 (Helmholtz), or from 1*123 to -988 (Wright and Thompson). 

 Hence it is evident that mercury cells do not evolve electro- 

 motive forces of about half a volt in excess of their calculated 

 values, the various statements to that effect being based on 

 inaccurate thermochemical *data. On the contrary, it will be 

 noticed that the value of E;^ for the nitrate cell is greater 

 than the experimental value E, which may be due to an error 

 in my determination of the heat of formation of mercurous 

 nitrate, but more probably to the same cause acting as in the 

 cases of lead and silver, both of which metals have negative 

 " thermovoltaic constants/^ and which greatly resemble mer- 

 cury in many chemical characteristics. 



Silver. 



The " thermovoltaic constants '' for silver, given by Wright 

 and Thompson, vary from '395, in the case of the nitrate, to 

 •02, in the case of the iodide, but are all negative in sign. It 

 is noticeable that the larger values are attached to the 

 soluble salts, while the insoluble chloride, bromide, and iodide 

 have much smaller " constants," and these depend largely 

 on the nature of the liquid in which the haloid salt is sus- 

 pended ; thus AgCl in solution of ZnCl2 has a thermovoltaic 

 constant of "112 to -062 volt, while a copper-silver chloride 

 cell in which the AgCl is suspended in CuCl2 solution has a 

 difference of only '001 volt from the computed value. The 

 causes producing the large divergence in the case of the 

 nitrate or sulphate are, therefore, probably different in nature 

 from those producing the small and very variable " con- 

 stants " (?) in the case of the chloride, bromide^ or iodide. 



In order to test whether a cell consisting of silver in a 

 solution of its nitrate, opposed to some other metal, gives rise 

 to reversible heat-effects or not, a copper-silver nitrate cell 

 was prepared and used in the manner previously described. 

 It was found that when small currents (-1 to '2 amp.) were 

 sent in alternate directions through the cell by means of 

 another battery, apart from the heat due to the square of the 

 current and the resistance, there was an evolution of heat 

 when the cell was worked forwards,, and an absorption of heat 

 when worked backwards so as to dissolve silver and deposit 

 copper. 



The following numbers, taken from one set of experiments, 

 show the difference of heating with synergetic and opposed 

 currents (E.M.F. of the ceU -431 to -428 volt):— 



