370 Mr. A. Frumkin on the 



We see that, in spite of tbe great difference in the positions 

 of the maxima, Paschen's relation always holds within the 

 limits of experimental error. 



Let us now examine two circumstances which can cause 

 an apparent contradiction with Pascheir's law. 



(1) Let us suppose that there is in the solution a very 

 active substance, which even at low concentrations displaces 

 tbe maximum of the electrocapillary curve. Every mercury 

 drop adsorbs and removes from the solution a certain 

 quantity of the active substance. Its concentration in tbe 

 immediate neighbourhood of tbe dropping electrode must be 

 less than the concentration of the original solution, and 

 therefore the potential of the dropping electrode has a less 

 displaced value than that one which corresponds to tbe 

 maximum of the electrocapillary curve, as determined by 

 the capillary electrometer. 



This discrepancy may be of importance only if the con- 

 centration of tbe active substance is very low, at any rate, 

 when the concentrations of all components of the solution 

 (ions of mercury of course excepted) are high enough, it 

 vanishes as we see from Table I. 



(2) With the capillary electrometer we measure the static 

 values of the surface-tension which correspond to an equi- 

 librium of the distribution of all components between tbe 

 surface-layer and the bulk of the solution. If this equi- 

 librium has not time to establish itself while the drop is being- 

 formed, there must be a discrepancy between the data as 

 given by the dropping electrode and the capillary electro- 

 meter. We may expect this discrepancy to be especially 

 great in the case of solutions which contain a small quantity 

 of an active substance in presence of a great excess of an 

 inactive one. Such solutions, according to Gouy *, show 

 "electrocapillary viscosity," i. e. the meniscus of the 

 capillary electrometer follows the variations of pressure and 

 potential with some delay, and there is a perceptible difference 

 between the surface-tension of a fresh meniscus and the 

 final value of surface-tension. In fact, Palm aer's "null" 

 solution 0*1 n KC1 + 0*01 n KCN, which shows electro- 

 capillary viscosity, shows also a contradiction with Paschen's 

 relation: the dropping electrode potential is here 0'574volt, 

 in good agreement with the position of the maximum of the 

 electrocapillary curve of pure 0*1 n KC1 (0*57), whereas a 

 curve plotted from the final surface-tension values of 

 0*1 n KCl + OOlnKON has a maximum corresponding to 

 0'64 volt. A fresh meniscus gives data which are included 



* Ann. chim.phys. (7) xxix. p. 239 (1903). 



