with the Mercury Cathode. 385 



Under the most favorable conditions, determined empiri- 

 cally for one definite amount of sodium chloride, results show 

 fair analytical accuracy. The chlorine tends to be low, aver- 

 aging — "0005 grms. between the extremes of + - 0008 grms. 

 and — "0030 grms. ; and the sodium high, averaging + *0005 

 grms. hetween the extremes of +'0016 grins, and —•0018 

 grins. The direction of the errors is just what would be 

 expected from the preceding work of this paper. 



Summaey 



This work shows then, that in the electrolysis of sodium 

 chloride, under the conditions descrihed, with the anode of 

 silver, or silver plated, and the mercury cathode, silver is 

 always transferred from the anode to the cathode mercury; 

 although, under conditions determined by experiment to be 

 most favorable for a given amount of chloride, the amount of 

 silver transferred may, for analytical purposes, be neglected. 



The favorable conditions used in this work for the electroly- 

 sis of 50 emS of *1N sodium chloride solution, containing *2923 

 grams of salt, with the apparatus described, were obtained by 

 using a current l - 2-15 amps, and allowing it to fall to 0*1 

 amp. This required 18-20 minutes. 



It is recommended that the anode covered with silver chlo- 

 ride be first heated below the fusing point of the chloride to 

 decompose all of the silver oxide and that then the heat be 

 increased so as to fuse the chloride, five or ten minutes at a 

 temperature 400-500° being sufficient. 



Sodium hydroxide is always present in the inner cell after 

 the beginning of the electrolysis. 



The best method of treating the anode covered with fused 

 silver chloride to prepare it to be used in a subsequent electroly- 

 sis is by heating about 20 minutes at about 500° or a little over 

 in a current of hydrogen. 



