Electrolytic Conduction of Clausius. 87 



and the arrangement put in circuit with two cells and a 

 reflecting-galvanometer. No deflexion ensued ; but as soon 

 as the upper carbon was pressed down on the precipitate, 

 there was a deflexion of 429 divisions. To estimate roughly 

 the resistance of the layer of precipitate, a resistance of 10,000 

 ohms was now interposed. This reduced the deflexion to 

 426, indicating a resistance of about three quarters of a 

 megohm. 



On cutting out the battery there was a distinct secondary 

 current. This secondary current would go to show that the 

 precipitate was a conductor of the second order, and therefore 

 perhaps unlikely to act as a metallic diaphragm. It was 

 therefore necessary to examine the nature of its conductivity 

 more closely; and since the precipitate prepared as described 

 is complicated by the presence of potassium ferrocyanide, I 

 prepared a fresh portion of cupric ferrocyanide by the method 

 of Ramelsberg for obtaining the pure substance *. I mixed 

 the aqueous solutions of hydroferrocyanic acid and pure 

 copper sulphate, filtered out and well washed the precipitate, 

 dried it in vacuo over sulphuric acid, and tested it in the same 

 apparatus as before. Its resistance appeared to be about half as 

 much again as that of the first precipitate. With an E.M.F. of 

 about '01 volt it conducted sensibly; with three volts it gave 

 a deflexion of about 700 divisions; and when this latter current 

 had passed for six minutes, on removing the battery the only 

 evidence of polarization was a mere quivering of the index 

 over about two divisions on making contact. 



There would appear to be no doubt, therefore, that cupric 

 ferrocyanide is a conductor of the first order, although of 

 somewhat high resistance; and the phenomena of Prof. Ost- 

 wald's experiment (so far as one can judge from the merely 

 qualitative data given) may be explainable on a very simple 

 hypothesis without having to attribute them distinctly or 

 solely to any semi-permeability in the partition. 



The polarization observed with the first precipitate may be 

 due to impurity, perhaps to the mixture of potassium ferro- 

 cyanide. Its persistence, notwithstanding the short-circuiting 

 by the conducting cupric ferrocyanide, may easily be due to 

 the very high resistance of the latter. 



In any case, however, the conductivity of cupric ferro- 

 cyanide must evidently be taken into account in all the 

 experiments in which it takes a part. In those on osmotic 

 pressure for instance, where a porous partition containing a 

 deposit of this salt in its pores is interposed between water 



* Pogg. Ann, lxxiv. p. Go, 184S. 



