ON TIIK SILVER VOLTAMETEIi. 575 



the highest point of the anode to tin- lowest. This, however, only holds good for a 

 few seconds after the cessation of the current in the main circuit. The siijici position 

 of tin- i-IVrct MM tlir main current effect results in the current density l>eing greatest 

 at the |>oint where the anode enters the electrolyte, and this is, in general, the first 

 portion of the anode which becomes noticeably thin. If the current is very feeble, 

 < [illusion tends to keep the liquid more homogeneous. Observations show that when 

 silver rods are used as anodes there are other effects of electrolysis which produce 

 vertical grooves in them. 



If the concentration of the electrolyte is diminished, the ratio of the concentration 

 of the anode film to that of the main electrolyte is increased and that of the kathode 

 film to the liquid is probably diminished. There is, however, a similarity with respect 

 to the action of the current on the surfaces of separation of these films and the 

 electrolyte. In l>oth cases the current in its passage through the voltameter flows 

 from a liquid of high concentration into a mass of liquid of lower concentration, and 

 hence, if there is a marked surface of separation of anode liquid and electrolyte, and 

 kathode liquid and electrolyte, any effect of the current on these surfaces will be 

 similar. CORK* has shown that when a solution of small concentration rests on one 

 of high concentration and a current is passed downwards, the surface of separation of 

 the liquids becomes indistinct, but if the current is reversed the surface of separation 

 becomes more marked. By using "silver anodes and kathodes of platinum foil in 

 electrolytes contained in glass vessels we were able to see the heavy anode liquid in 

 its descent from the anode and the light liquid in its ascent from the kathode. Even 

 when at a distance of a few millimetres from the electrode these liquids appeared to 

 be quite distinct from the main body of the electrolyte, and we are justified therefore 

 in assuming the existence of even more distinct surfaces of separation around the 

 anode and kathode when the current is flowing. 



An interesting question is whether the properties of these films of liquids are very 

 different from the main portion of the electrolyte. Olraervation shows that as they 

 leave the electrodes they break up into cylindrical columns, but whether or not they 

 are in the form of uniform thin films when in contact with the electrodes direct 

 observation does not show, but a number of experiments with currents of different 

 intensities aud electrolytes of different concentrations enable an opinion to lie formed. 

 When weak electrolytes are used (e.g., l per cent, solutions) a current of O'l ampere 

 produces a deposit having a matt surface, but with a 15 per cent, solution the deposit 

 is markedly striated (figs. 9t and 10, Plate 9). We interpret these results in the 

 following manner. The film of liquid in contact with the kathode has a greater mean 

 thickness in the l per cent, solution than in the 15 per cent, solution. This follows 

 because the rate of deposition of silver is the same in each voltameter. If the film is 

 very thin, it is unstable and breaks up into cylindrical columns of liquid. Hence in 



* GORK, ' Roy. Soc. Proc.,' No. 203, p. 332, 1880, and No. 212, p. 56, 1881. 



t Fig. 9 represents a portion of a deposit which was stripped from the side of a platinum bowl. 



