332 



Dr. G. Gore. On some Relations of [Feb. 21, 



of gas or formation of solid coating npon the plates, and of any- 

 other circumstance which appeared likely to affect the speed of 

 corrosion, or weight of the plates, or of the deposited silver; and in 

 any case where any solid coating was found, it was entirely removed 

 after the experiment, and previous to ascertaining the losses of weight 

 by corrosion. After the experiment the plates were well washed and 

 carefully dried between hot sheets of filter-paper before weighing. 

 The metals and liquids employed were as pure as could conveniently 

 be obtained, and distilled water was used for all the solutions. The 

 sample of potassic cyanide usually employed was found by analysis to 

 contain 89*14 per cent, of the actual substance. Nearly all the 

 liquids were at the atmospheric temperature, and in nearly every case 

 the comparison sheet was employed. 



As the proportion of silver deposited to the amount of external 

 current generated, and therefore also to the amount of positive metal 

 corroded, varied to a very small extent with the density of the current 

 at the cathode in the depositing solution (see " Electrolytic Balance 

 of Chemical Corrosion," " Proc. Birm. Phil. Soc," vol. iii, 1883, 

 pp. 278-304), the amount of cathode surface was usually varied 

 according to the apparent strength of the current and rate of 

 deposition, being smaller as the strength of current was less. 



The degrees of electromotive force were measured by the aid of 

 two thermoelectric batteries, each consisting of about 300 pairs of 

 iron and German silver, together with a Clark's Standard cell (see 

 " Proc. Birm. Phil. Soc," vol. iv, Pt. I). 



The following table exhibits the results. In the table the number 

 of square inches given includes in each case the immersed portion of 

 both surfaces of the positive metal and comparison plate. The losses 

 of the plates, and the weight of deposited silver, are in every instance 

 expressed in grains, and the rates of loss of the positive plate are 

 given in column 6 in grains per square inch per hour. If the whole 

 of the corroded portion of the positive metal produced external 

 current, consistent in amount with the atomic weights and valencies 

 of the metals as given in the table, and if that amount be called 100, 

 then the proportions of such current actually obtained in the 

 experiments are those given in column No. 9. The quantities of 

 substance mentioned in column No. 1 were in each case per ounce of 

 water. 



Remarks. 



Only those results which appeared reliable are inserted in the table ; 

 many uncertain ones were omitted. In some cases coatings of solid 

 matter formed either upon the positive or comparison plates, or gas 

 was evolved by the plates. Cases in which the coatings could not be 

 perfectly removed were rejected. With magnesium in solutions of 

 amnionic, potassic, or sodic chloride, potassic bromide or iodide, a 



