and a Theory of Contact- Potential. 325 



by an iron wire, or if this was dipped in the mercury collected 

 in the cup E. 



In this latter case there were attached to both electrodes 

 iron wires dipping in mercury, exposed to the air, and in 

 almost identical conditions — the only difference being in the 

 capillary relations of the mercury contained in the cup and 

 of that in the funnel and neck. 



The question arose, Is the formation of drops at the end of a 

 very fine tube accompanied by electrical phenomena such as to 

 cause a positive charge to be gradually accumulated in the mer- 

 cury-reservoir f 



The effect observed was not due to the electrification pro- 

 duced by friction against the glass; for this effect is of much 

 smaller dimensions, and the mercury would be negatively 

 charged. 



The case of these falling drops is in many respects analo- 

 gous to that of mercury advancing in a capillary tube. I 

 tested therefore the behaviour, in such a tube, of a column of 

 mercury broken by an air-bubble, by connecting the cups in 

 which dipped the ends of the tube to the two electrodes of an 

 electrometer, then suddenly raising one cup. 



I found that the advancing column was always positive to 

 the retreating mercury; and by multiplying the effect by in- 

 serting many air-bubbles a deflection was obtained, when the 

 rate of flow was sufficient, corresponding to *7 D. (The fric- 

 tional charge, as shown by a low-resistance galvanometer and 

 an undivided column of mercury, was extremely slight.) In 

 such an apparatus the retreating end is oxidized, the coating 

 of oxide appearing after prolonged action, while the advancing 

 end is reduced and acquires a positive charge. 



Similarly, the advancing drop in the apparatus first described 

 beomes positively charged, so long as any reduction takes place. 

 This depends on previous exposure to the air and other cir- 

 cumstances which favour oxidation, but also on an action 

 which takes place more or less continuously in the following 

 way: — The end of the thread, after the drop has detached itself, 

 suffers a momentary retraction ; and this therefore causes a 

 slight oxidation of the surface, which, however, is not of suffi- 

 cient duration to lower the potential of the mass of mercury 

 in the column to any extent; for, the forward motion at once 

 setting in, the oxide locally deposited is reduced, this reduc- 

 tion being equivalent to a positive charging ; and, since the 

 drop takes time to form , the positive charge is communicated 

 to the column, which thus becomes more and more charged 

 by each forward motion of the column, so long as oxidation 

 can occur immediately after each drop has fallen. 



Phil Mag. S. 5. Vol. 12. No. 76. Nov. 1881. 2C 



