Mercury-dropping Electrodes. 385 



is allowed to run out in small drops under the surface of a 

 liquid, the surface of each drop forming out of the interior of 

 the pure metal which has not yet been in contact with either 

 air or liquid. According to all Quincke's* observalions, the 

 positive current in this case always goes in the direction of 

 the falling drops of mercury. 



3. " That is to say, the mercury which collects on the 

 bottom of the containing vessel, and which concentrates on 

 its surface the films that have been produced by the changes 

 referred to, has greater positive potential than the upper, 

 continually renewed surface. Such a difference of potential 

 implies an electric double layer, whose positive half is inside 

 the lower mercuiy and its negative on the anion of the fluid 

 in contact. 



4. " This double layer forms gradually, since it has been 

 shown by Quincke that slow dropping produces weak differ- 

 ences of potential ; while, as the speed of dropping increases, 

 a maximum difference of potential is reached, after which 

 increasing the speed of dropping has no further effect. The 

 maximum will occur as soon as the new portions of the upper 

 mercury form into drops so quickly that they cease to be 

 charged sensibly before they break away, and therefore the 

 upper surface remains in a perfectly unchanged state. 



5. ''According to Faraday's law, the passage of positive 

 electricity into the metal can only occur by electrolysis, in 

 which is concerned some substance having less attraction for 

 positive electricity than the mercury. This may probably bo 

 atmospheric oxygen dissolved in the liquid, to which we may 

 ascribe sufficient affinity for negative electricity to draw it out 

 of the mercury and replace it with positive electricity. The 

 slowness of the chareing would then be due to the smallness 

 of the quantity of oxygen present and its slow renewal by 

 diffusion. Further experiment is needed to prove this 

 hypothesis. Quincke shows that boiling the electrolyte does 

 not stop the phenomenon, but a very small quantity of oxygen 

 would be enough. 



6. " If the effect were due to any component of the elec- 

 trolyte which is present in large quantity, the charging of the 

 surfaces in contact would take place in an insensibly short 

 time. 



7. " Although this is to be understood merely as a hypo- 

 thesis, it is sufficient for what follows that, under the received 

 conditions, mercury in contact with liquid charges itself, only 



* Pogg. Ann. cliii. p. 161 (1874). 

 PIdL Mag. S. 5. Vol. 27. No. 168. May 1889. 2 G 



