IX AQUEOUS SOLUTION, AND THE EXISTENCE OF COMPLEX IONS. 
1 
2.3 
resulting lithium chloride solution shows throughout the entire length of the column 
a uniform blue colour. In the latter case, when Cu is the following ion, it 
encroaches on the intermediate solution and advances ahead of the refraction margin 
from 0'5 to I'O centim., the colour gradually fading away. 
The only case in which no boundary whatever has been obtained occurred in the 
attenij)t to measure the velocity of the NO 3 ion in nitric acid."^^ Three indicators (the 
acetic, oxalic, and phenylsulphonic ions) were tried in different experiments, always 
without success. It may be mentioned that with such excellent conductors as the 
acids the use of a high voltage is practically impossible on account of the great heat 
development. An apparatus was, however, constructed by means of which a potential 
fall of 2'0 to 3'4 volts was obtained at the anion boundary ; under these conditions 
the whole column of liquid was seen to be mixed by rapid convection currents. 
It has been occasionally noticed that with the same boundary the inferior limit 
apparently differs according to the direction from which it is approached ; thus for the 
^ NO 3 margin, when an experiment is started with a low voltage and this is 
gradually increased, no margin can he detected at 0'29 volt, whereas starting with 
0’4 volt and diminishing, the boundary does not entirely disappear until a potential 
fall of 0'17 volt is reached. This is due to the fact that with a lower potential fall a 
larger time is required before the stationary condition is reached than with a higher 
one, and if before an experiment is started a little diffusion takes place at the point 
where gelatine and aqueous solutions are in contact, it is possible that the margin 
may have travelled right through the tube before it has become visible. 
The production of a refraction boundary in the case that the slower ion precedes 
the faster, has been noticed for a few cases, e.c/., Cl, j;- Cl, and^ Cl. In the 
’ ^ ’ Mg Cu Cu 
majority of instances, howevei', no boundary is obtainable under these circumstances. 
But if at the beginning of such an experiment the concentrations of the two solutions 
are proportional to the transport numbers of the respective cations (or anions), then a 
stable boundary results which travels with the current. Such a ratio of concentration 
is automatically brought about during electrolysis, when the slower ion follows the 
fastei', and once this condition is established the margin so produced may be made to 
move backwards by altering the direction of the current, and this may be repeated a 
number of times without losing the boundary, althougli its velocity has been proved 
by experiment to l)e less iji the backward than in the forward direction. The 
stability of this margin is easily understood from the fact that the potential slope in 
* For the investigation of acids and bases it is impossible to employ gelatine solutions in the cathode 
and anode cells I'espectively, as these are immediately destroyed by the H and OH ions; for these, there¬ 
fore, a partition of earthenware was substituted at the one end, and in order to prevent as far as possible 
movements of the solution through the membrane, the open limb of the apparatus (E, fig. 1) is dispensed 
with. 
