ON ELECTKOLTSIS IN ITS PHYSICAL AND CHEMICAL BEARINGS. 321 



The chlorate deposit after heing heated to redness was dissolved in pure nitric 

 acid and the solution tested for chloride, but none was found. 



The silver cell being thus shown to be capable of working quite satisfactorily, the 

 arrangement and comparison of copper cells were proceeded with. It was found 

 in the previous experiments on copper that the deposits obtained on platinum 

 wires gave much more concordant results than any other form of cell. This form 

 of cell had been already used by Soret. It was accordingly decided to adopt that 

 form. The cathodes were platinum wires about 15 cm. long and 1 mm. in dia- 

 meter, and the anodes were wires of electrolytic copper whose diameter was some- 

 what greater. It is, however, well known that the density of the solution alters 

 very considerably in the neighbourhood of the cathode, and if the electrodes are 

 vertical this effect is not so much interfered with by convection cm-rents as when 

 the electrodes are horizontal. The wires were therefore placed horizontally in 

 the solution ; for this purpose they were bent at right angles, and the one end 

 fastened to a binding screw attached to a thick wooden rod that lay on the edges 

 of a porcelain dish that contained the electrolyte. A number of cells were pre- 

 pared respectively for 1, 2, 3 up to 16 wires. The wires were arranged 



about 2 inches apart, and between them lay the anode wires attached to a similar 

 piece of wood that rested on the other end of the dish ; the wires in the same cell 

 were put in multiple arc by a strip of copper laid along the wood, through which 

 the binding screws passed. Some difficulty was met with in arranging that the 

 anodes and cathodes did not touch, but with care it was surmounted. The solution 

 was an approximately saturated solution of Hopkin and Williams's pure copper sul- 

 phate. No acid was added. The wires -with their deposits were washed by rinsing 

 them in several changes of distilled water and then dipped in alcohol and dried in 

 an air-bath at a temperature of about 100° C. 



In the experiments given below the copper deposits -were brilliantly clean and 

 generally showed no trace of oxidation on drying. The reason for this was never 

 made clear ; it is, however, easily recognised that under certain circumstances, 

 defined perhaps by current density or the state of the solution, the copper deposits 

 do not easily oxidise on drying, whereas in other experiments it seems almost 

 impossible to dry them without considerable oxidation. 



In all the experiments a rough tangent galvanometer was included in the 

 circuit, so that any considerable variations of the current might be detected. A 

 water or dilute copper sulphate cell, with copper electrodes whose distance apart 

 could be adjusted, was also included to serve the purpose of a rheostat. 



In a first series of experiments a silver cell was compared with each one of the 

 copper wire cells in turn ; the results are given in the following table : — 



1886. 



