Electromotive Forces from Thermochemicat Data. 227 



These considerations, therefore, show that the heat of 

 formation of the anhydrous salt is the most important factor 

 in determining- the chemical reaction, and suggest that the 

 kind of energy set free in the formation of a dry salt differs 

 from that due to the hydration of that salt, the latter being 

 only partly " free energy," while the negative heat of simple 

 solution may be supplied entirely by the " bound energy " of 

 the chemical change. 



A similar explanation will account for the fact that an 

 iron-cadmium chloride cell gives an E.M.F. opposite in 

 direction to that anticipated from the heats of formation 

 of their salts in solution, since Thorasen gives [Cd,Cl2] 

 = 93,240, [Cd, Cl2,Aq] = 96,250, and [Fe, Cl2]= 82,050, 

 [Fe, Cl2,Aq]=99,y50. 



Thus, taking the anhydrous salts we get a difference of 

 11,190 on the side of cadmium, while taking the dissolved 

 salts gives 3700 calories in favour of iron being attacked and 

 depositing cadmium, which is known not to occur. Thus the 

 first action in point of time — the formation of the anhydrous 

 salt — directs the course of the chemical change, which, how- 

 ever, in the present case is not a replacement of iron by 

 cadmium, but if a current flow, hydrogen is evolved on the 

 iron plate ; so that this cell does not reverse the expected 

 reaction and furnish a current at the expense of sensible 

 heat, but, owing to the initial difference in the heats of forma- 

 tion of the dry salts, an entirely different reaction is started 

 and continued. 



It is owing to the fact that iron is not deposited electro- 

 lytically from neutral or acid solutions of its salts that its 

 large positive constant was referred earlier in this paper to 

 a false reaction, though the above consideration shows at least 

 one cause for its anomalous behaviour. 



Tin. 



Measurements of the electromotive force of cells in which 

 tin in a solution of its chloride is opposed to other metals 

 having shown * that it possesses an apparent negative " ther- 

 movoltaic constant,^^ it was resolved to employ a cell in which 

 tin is the metal attacked, and v/hich should consequently give 

 an E.M.F. in excess of the heat evolution due to the chemical 

 change. Copper, which naturally suggests itself as the 

 opposing metal, is, however, very unsatisfactory for thermal 

 measurements when used in a solution of its chloride, owing 



* Phil. Mag. ser. 6, vol. xxi. p. 13. 

 Q2 



