230 Prof. E. F. Herroun on the Divergence of 



Hence in tin cells, while I recognize a negative " thermo- 

 voltaic constant" which is capable of producing reversible 

 thermal effects in the cell, I am inclined to refer them to two 

 causes : (a) a negative heat of dilution of the salts, and (/3) to 

 effects of differences of concentration of the acid necessarily 

 employed to keep the salt in solution*. 



General Conclusions. 



A review of the published measurements of the thermal 

 values of chemical reactions, in particular those directly re- 

 lating to electromotive forces, namely, the replacement of one 

 metal by another from its dissolved salt, will, 1 think, show 

 that however valuable they may be as guides they cannot be 

 implicitly relied upon as furnishing absolute and unimpeach- 

 able data by which to judge the truth or fallacy of a given 

 theoretical conclusion. In the first place, in the numbers 

 representing the heat of formation of the same body, deter- 

 mined by different experimenters, there are frequently dis- 

 crepancies of considerable magnitude. Thus Thomsen finds 

 for the heat of reaction of lead acetate and zinc 34,950 caL, 

 Favre finding only 31,200 c, while Andrews gave 37,710 c.; 

 a total variation between the extremes of 6,510 calories. For 

 the heat of formation of CuO, and therefore of all salts 

 derived from it, Thomsen gives 37,160, Favre and Siiber- 

 mann 43,770, and Andrews 38,300, while [Fe, Clg, Aq] is 

 variously estimated by Thomsen at 99,950, by Favre and 

 Silbermann at 106,700, and by Andrews at 102,060. 



Besides these experimental differences there is always the 

 fact that, as Berthelot has clearly laid down in his introduction 

 to his Essai de mecanique chimique, the quantity of heat 

 evolved in a reaction measures the sum of the phi/sical and 

 chemical changes which occur in the reaction ; some of which 

 quantities may be positive or negative, and which may or 

 may not directly influence the transformable energy of the 

 chemical change. 



Adopting the view of Helmholtz that a chemical process 

 gives rise to a total amount of energy, part of which is 

 freely transformable and part of which is " bound energy ,^^ 

 while the calorimeter measures the total energy, the elec- 

 tromotive force of a voltaic cell is a measure of the actual 

 transformation of free energy. In cases, therefore, in which 

 a portion of the total energy is to be reckoned as negative 



* In my previously published measurements of the E.M.F. of tin cells 

 the value given for the cadniium-tin sulphate cell was only "189 volt ; but 

 I have since found that, using more dilute and less acid solutions, a per- 

 sistent value as high as •234 volt may be obtained. 



