450 DE. E. B. DENISON AND DE. B. D. STEELE ON THE 



development of this idea reference must be made to the original literature ;* but the 

 following is a brief account of the method in its present form : 



It has been shown, both theoretically and experimentally, that if two salt solutions 

 containing a common ion are placed one above the other, and an electric current 

 passed through the system, a stable margin will be formed between the two solutions, 

 provided that the specifically slower non-common ion follows the faster one. Under 

 these circumstances it has been shown that the boundary moves in the same direction 

 as the non-common ions and measures their velocity. In order to observe and measure 

 the velocity of the boundary, advantage is taken of the slight difference in the 

 refractive index of the two solutions, which difference renders the margin quite 

 visible when viewed through a telescope. The method of determining the ratio U/V, 

 and hence the transport number. U/(U + V), for a given salt will be rendered clear by 

 the following considerations :- 



Let us suppose that a current flows through the system, 



Anode solution of lithium chloride solution of potassium chloride solution of 

 potassium acetate cathode. 



At the margin between the lithium and potassium chloride solutions the slower 

 lithium ion follows the specifically faster potassium ion, and there results a stable 

 margin, the velocity of which is that of the lithium and potassium ions at that point. 

 This, velocity, however, depends on the potential gradient, and it has been proved 

 that the concentration of the lithium chloride becomes automatically adjusted, so that 

 the potential gradient is just sufficiently increased to make the lithium ions keep pace 

 with the potassium ions. In the same way the specifically slower acetate ion follows 

 the faster chlorine ion, at the margin potassium acetate, chloride, and the motion of 

 this margin gives the velocity of the acetate-chlorine ions ; but whilst the lithium 

 and acetate ions are moving under the influence of an unknown potential gradient, 

 that which is driving the potassium ion is the same as that driving the chlorine ion 

 the potassium chloride solution being homogeneous and, therefore, the velocities of 

 these two ions are strictly comparable. Moreover, the potential gradient in the 

 middle electrolyte can easily be calculated, and hence, also, the average mobility of 

 the ions, or their velocity under a driving force of one volt per centimetre. 



The ratio U/V and the average mobilities of the ions of a number of salts have 

 been determined in this way,t and whilst the agreement between the results so 

 obtained and those obtained by the indirect method is, on the whole, fair, considerable 

 deviations occur in the case of many salts. In the method used for these measure- 



* WHETHAM ('Phil. Trans.,' A, 1893, p. 337; A, 1895, p. 507); MASSON ('Phil. Trans.,' A, 1899, 

 p. 331); KOHLRAUSCH (' Wied. Ami.,' 1899, LXII., p. 209); WEBER ('Sitz. Berliner Akad.,' 1897); 

 STEELE ('Phil. Trans.,' A, 1902, p. 105; 'Zeit. fiir Phys. Chem.,' 1902, XL., p. 689); ABEGG and GAUS 

 ('Zeit. fur Phys. Chem.,' 1902, XL., p. 737); DENISOX ('Zeit. fiir Phys. Chem., 1 XLIV., p. 575). 



t MASSON (he. tit.) STEELE (loc. tit.). 



