Manchester Memoirs, Vol. xlvii. (1903), No. \%. 5 



different experiments, taken almost at random from the 

 much greater number which I have tried. Solutions of 

 cobalt of two different strengths were used. I have 

 represented in each case the actual amount of iodine 

 liberated by dissolving the oxide in hydrochloric acid and 

 potassium iodide, and, for comparison, have added the 

 calculated amounts of iodine for the two oxides men- 

 tioned above, and also for the sesquioxide. 



The cobalt solutions were prepared by dissolving an 

 indefinite amount of pure cobalt chloride in water, and 

 determining the amount of cobalt present by electrolytic 

 deposition. The electrolysis was performed by the current 

 from three Daniell's cells, the liquid, to which a consider- 

 able amount of ammonium oxalate had been added,* being 

 kept warm during the whole time, which extended to 

 about six or seven hours. The metallic cobalt was 

 deposited in a weighed platinum basin. 



Bayley {loc. czt.) recommends, for the purpose of 

 preparing a solution of cobalt of definite strength, heating 

 the pure crystallised sulphate to dull redness, in order to 

 expel the water of crystallisation, and dissolving a weighed 

 quantity of the dried salt. I have tried that method, but 

 do not consider it so satisfactory as the method of 

 electrolytic deposition. During the heating I strongly 

 suspected that some decomposition of the sulphate occurred, 

 and this supposition was confirmed by a subsequent deter- 

 mination of the cobalt in the solution by electrolytic 

 deposition, which gave slightly more cobalt than that 

 calculated from the amount of the dried sulphate. 



In the following Table of the various higher oxides of 

 cobalt which have been described, they are arranged 

 in ascending order, and, in order to show more clearly 



*The solution of the double oxalate of cobalt and ammonium is one of 

 the best for the electrolytic deposition of cobalt. 



