578 MR. F. E. SMITH, MR. T. MATHER, AND DR. T. M. LOWRY 





 deposited in the form of a thin ring where the electrolyte was at its highest point, and 



one or two millimetres below this ring the edge of the main deposit was formed. A 

 photograph of such a ring deposit is given in fig. 13 (Plate 9). 



Another interesting question is whether the kathode and anode films can be easily 

 scattered by agitation of the liquid. We have made experiments with the kathode 

 only, for there were no means of detecting whether the anode film was disturbed or 

 not. Our solutions were such (solutions used in Observations 1 to 13) that striae 

 were readily produced, and the method of investigation was to rotate the kathode 

 and insert stationary glass vanes in the electrolyte to prevent a uniform motion. We 

 were astonished to find the striae as distinct as ever, but instead of being vertical they 

 were in the form of a spiral. This is well shown in figs. 12 and 13, which are from 

 photographs. Examination of the inclination shows ( 1 ) that the kathode liquid moved 

 upwards ; (2) that on the assumption that the kathode alone rotated and the whole 

 of the electrolyte remained stationary, the vertical velocity was 21 centims. per 

 second.* The electrolyte was not stationary, however, and we judged the kathode 

 film to rotate at very nearly the same rate as the kathode ; had it been at exactly the 

 same rate the striae would have been vertical. It is evident, therefore, that the 

 vertical velocity of the kathode film was not very great ; it was possibly of the 

 order of 1 centim. per second. The stability of the liquid columns is, however, 

 astonishing. 



It is well to point out that the Richards voltameter is a concentration cell after 

 electrolysis commences. The anode liquid within the porous pot is of higher 

 concentration than the kathode liquid, and in consequence any short-circuiting of the 

 voltameter after the main circuit has been broken will diminish the mass of silver on 

 the kathode bowl. A steady current may thus be produced for some time, and is 

 easily measured by an ammeter. 



Another point investigated by us was the possibility of silver being deposited from 

 a concentrated solution of silver nitrate at the bottom of a platinum bowl when a 

 second solution of much lower concentration rests on it. We employed solutions 

 containing 50 per cent, and 1 per cent, of silver nitrate, but neither in platinum bowls 

 nor in silver ones was any increase observed. 



Other effects have been observed, but it is not easy to suggest an explanation of 

 them. The most remarkable occurs when a kathode bowl is half filled with an 

 electrolyte, left for an interval of about 10 minutes, and the remaining portion of the 

 electrolyte added just prior to the completion of the electric circuit. If about 

 5 grammes (or less) of silver are deposited, the level of the electrolyte when the bowl 

 was half filled is clearly indicated in the deposit, the density of the latter being 

 appreciably greater below the original level of the solution, and the change in the 

 density is marked by a clear line running round the bowl. 



h The bowl rotated clockwise at the rate of 40 turns per minute : its maximum diameter was 10 centims., 

 and the inclination of the striae was almost exactly 45 C. 



