On the Existence of a Compound of Gold and Tin. 95 



the zinc will polarize the whole surface of the copper, and the 

 cell will give very nearly the same deflexion on an electro- 

 meter as a Darnell's cell. 



We have here, then, a delicate method for detecting the 

 presence of very small quantities of a more positive metal 

 when merely mixed with a more negative metal. 



If, however, these two metals have entered into combination 

 with the evolution of heat, energy must be absorbed in the 

 cell to break down this compound, and we should expect the 

 E.M.F. to be correspondingly lowered. 



Let us now suppose that using the same Daniell's cell we 

 replace the zinc plate by an alloy formed by melting together 

 copper and zinc, for instance a piece of ordinary brass wire. 

 We find that this brass wire behaves like a piece of copper in 

 the cell, that apparently the zinc it contains is not free to 

 enter into combination. 



On now making up a series of alloys, each containing more 

 zinc than the last, and testing their E.M.F. in the cell, an 

 alloy is at last reached which gives almost the same E.M.F. 

 as zinc, while the alloy before it behaved like copper in the 

 cell, though only containing 2 or 3 per cent, less of the more 

 positive metal. 



We have evidently, then, just passed over the alloy con- 

 taining exactly the right proportions of the two metals to form 

 the compound, and are now dealing with alloys which consist 

 of the compound mixed with an excess of free zinc, while all 

 the alloys before that point had consisted of the compound 

 mixed with an excess of free copper. 



In the case of alloys, however, where the metals are merely 

 mixed, the introduction of a very small percentage of the more 

 positive metal would at once be indicated by the electrometer. 



In this way I have examined the alloys formed between the 

 metals tin, zinc, lead, and cadmium, and the results have 

 already been published. In the case of these alloys, a very 

 small quantity of the more positive metal at once caused an 

 increase of E.M.F*, thus confirming Matthiessen's result that 

 these alloys are merely mixtures of the two metals one with 

 the other. In the case, however, of the copper-zinc and 

 copper-tin alloys, combination had evidently taken place, the 

 rise of E.M.F. of the copper-tin series taking place at the 

 same composition at which the maximum point of the curve 

 of conductivity occurs, as redetermined by Prof. Roberts- 

 Austen. Whether a redetermination of the conductivity of 

 the copper-zinc alloys would show a similar discontinuity 

 agreeing with the point at which the rise of E.M.F. occurs 

 remains to be seen. 



