1894.] Thermo-electric Properties of Salt Solutions. 359 



which I examined the current would go from hot to cold through the 

 solution. 



With the zinc salts I always used amalgamated zinc electrodes 

 coated with sealing-wax or india-rubber where they entered the solu- 

 tion. I always found these very satisfactory ; there was usually no 

 E.M.F. between them when at the same temperature. Unamal- 

 Igamated zinc is absolutely useless, as the E.M.F. flies about in a 

 fectly wild way. The following are the values found for 3 : 



Chloride of Zinc in Water, A, fig. 2. 



Concentration = 0'4888 3 = 6-609 



= 0-2444 3 = 6-868 



= 0-1222 3 = 7-295 



= 0-0611 3 - 7-89 



= 0-0244 3 = 8-14 



= 0-0122 3 = 8-38-4 



Unknown strong 3 = 6*515 



All the experiments on this salt were very satisfactory in every 

 way. 



Acetate of Zinc in Water, B. 



Concentration = 0'872 3 = 6'852 



= 0-436 3 - 6-854 



= 0-1744 3 = 7-17 



= 0-0872 3 = 7-38 



= 0-0436 3 = 7-55 



= 0-022 3 - 7-98-1 



i 



This salt gave very satisfactory results. 



Sulphate of Zinc in Water, C. 



Concentration = 0'525 3 = 8'95 



= 0-261 3 = 8'8 



= 0-0645 3 = 8-636 



= 0-0161 3 - 8-55 



This salt was not so satisfactory as those above, but the results are 

 sufficient to show that the sulphate is thermo-electrically positive in 

 aqueous solution. 



With the copper salts I came to the conclusion, after trying several 



forms of electrode, that the most satisfactory consisted of a short 



| length of fine wire projecting from the end of a glass tube, which 



1 was drawn out and stopped with shellac. The following results were 



obtained : 



