ON THE SILVER VOLTAMETER 589 



bowl surrounding it. The glass funnel largely prevented such a distribution of 

 current. 



The experiments recorded in Part I. indicate further that, under normal conditions 

 of working, the exclusion from the kathode vessel of the anode liquid by means of a 

 porous pot is without influence on the deposit, and we therefore conclude that the 

 cause of the abnormal values in our voltameter (Observations 1 to 13) is to be sought 

 in the contamination of the solution, but that this is not due to any change which is 

 inseparably connected with the conditions of electrolysis. 



F. Examination of the Mother- Liquors from Recovered Silver Nitrate. 



Although it was not found possible to obtain appreciably higher values for the 

 electrochemical equivalent by repeated or prolonged electrolysis, or by bringing the 

 anode liquid into contact with the kathode, it was known that high values could 

 readily be obtained by using recovered silver nitrate that had been cleared with acid 

 and rendered neutral by fusion, but not otherwise purified. It was therefore thought 

 to be desirable to investigate these solutions in order to determine the nature of the 

 impurity which they contained. For this purpose the mother-liquors left behind 

 during the purification of the recovered nitrate, as described in A II. and III., were 

 collected and examined. After recovering in a pure state the greater part of the 

 850 grammes of nitrate there remained a yellowish liquid containing about half its 

 weight of silver nitrate. On dilution with water the liquid became turbid and a 

 thick brown cloud was formed which ultimately settled down as a black precipitate.* 

 The diluted solution, which contained 14'4 per cent. AgNO a , was then electrolysed 

 and gave the extraordinary value 



T12141 milligrammes per coulomb (27b) 



for the electrochemical equivalent. A second electrolysis gave the value 



1-12055 (30c), 



and a third electrolysis, after the addition of 1 gramme of crystallised ferric nitrate to 

 about 400 cub. centims. of the mother-liquor, gave the value 



1-12171 (74d). 



As it appeared that the impurities were largely precipitated by diluting to 15 per 

 cent., an electrolysis was carried out with a more concentrated mother-liquor con- 

 taining 43 per cent. AgNO 3 . The value obtained for the electrochemical equivalent 

 was 1 '12252 (476), which is no less than 0'00425, or 0'36 per cent., higher than the 

 normal, and is possibly the highest yet recorded. 



* A slight cloudiness had already been noticed when the fused nitrate referred to at the beginning of 

 the section was dissolved in water and diluted after filtration instead of before. 



