OF SILVER BETWEEN ZINC AND LEAD. / 



two layers from the results of analysis gave very fliictuatiug 

 values. Thus in some cases this ratio, or the partition co- 

 efficient of silver between zinc and lead, was found to be as low 

 as 45, while in others it was something like 150 or more. Not 

 only was this discrepancy to be found with different samples, 

 but pieces from the same lump cut at different distances from 

 the line of demarcation gave different percentages of silver. 

 Thus it was found in the lighter layer that the pieces nearer 

 the line were richer in silver than those much removed from 

 the same ; while the heavier layer showed an utter irregularity 

 of distribution. After many trials in the case of the heavier 

 layer it was found that the alloy was much contaminated with 

 particles of zinc, especially on the surface, and that at some 

 points the contamination had penetrated to some extent into the 

 mass of the alloy. This may have happened owing to the 

 particles of zinc oxide, already existing in the metal used or 

 formed during the process of heating, persistently adhering to 

 the sides of the glass tube and helping to retain particles of 

 zinc, which would thus refuse to rise to the layer above even 

 after vigorous stirring. The discrepancies in the zinc layer may 

 have been due to the formation of minute crystals of a silver-zinc 

 compound which tended to accumulate in the lower part of the 

 layer even during the short time required for the mass to solidify. 



To remove these irregularities the zinc layer was analysed 

 as a whole mass. The whole of the lighter layer and a small 

 part of the heavier layer next to it was cut with a saw, dissolved 

 in nitric acid and titrated for silver as usual. In the solution 

 the lead was determined as peroxide deposited electrolytically. 

 The amount of zinc in the mass was determined by difference. . 



Previous to making these determinations, the solubility of 



