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



nesium, aluminium, zinc, and iron), hydrogen is liable, in some 

 cases certain, to be liberated in place of the equivalent of 

 metal. In other cells, such as lead or silver in contact with 

 their sulphates or free sulphuric acid, on making either of 

 those metals the anode in the cell, a peroxide (Pb02 or AgO) 

 is apt to form in spots on the surface of the metal plate and 

 entirely alter both the E.M.F. and resistance of the cell, 

 besides obviously altering the chemical work of electrolysis. 



The desiderata in a cell intended for this work — namely 

 tolerable freedom from polarization, perfect reversibility, low 

 and constant internal resistance — are only even approximately 

 fulfilled by a very few of a large number of cells examined 

 by me, and therefore the experimental results are greatly 

 restricted. 



Mercury. 



After dismissing aluminium and magnesium as being un- 

 suitable, for reasons already stated, the metal which has the 

 next highest " the rmo voltaic constant ^^ positive in sign is 

 mercury, and being the metal which is replaced by most 

 other metals, as copper, zinc, &c., it yields electromotive 

 forces considerably in excess of the values deduced from the 

 heat of formation of its salts. The insolubility and sparing- 

 solubility of the mercurous chloride, bromide, and sulphate, 

 and the inadmissibility of mercuric salts, owing to their 

 reduction in contact with metallic mercury, rendered it 

 necessary to employ mercurous nitrate, and, accordingly, 

 a copper-mercury nitrate cell was prepared, in which each 

 metal was in a strong solution of its own nitrate, with a 

 small proportion of free nitric acid, which is necessary to 

 keep the mercurous salt in solution by preventing the forma- 

 tion of basic salts, and also diminishes the internal resistance. 

 I found this cell gave an E.M.F. of "43 volt, while from the 

 thermochemical numbers of Julius Thomsen, viz. 



[Cu, 0, N2O5 Aq] = 52410 and [Hgg, 0, N2O5 Aq] =47990, 



one would deduce an electromotive force of only "095 volt. 



(Wright and Thompson, adopting the older standards of 

 measurement, give "433 volt for the experimental and "097 

 volt for the calculated value.) 



When a current of '2 ampere was passed in alternate 

 directions through the cell, it was found that if the current 

 was not continued for too long a period in one direction the 

 resistance was moderately constant, though by the prolonged 

 action of the current the solution became either nearly 

 deprived of mercury, or saturated with mercurous nitrate in 



