I 42 
ME. B. D. STEELE ON THE MEASUREMENT OF IONIC VELOCITIES 
The ap])Hi‘atu,s previ(ju.sly desciil>e(l, ^uid shown in fig. 2, is filled with the solution 
to be measured, and the cells C and C' Avitli the indicator jellies are placed in position 
tor the investigation of the anioii marc’iin Tlie cell C' contains the anion indicator. 
O O 
After the current is started the resistance of tlie solution Ijetween the points a and h 
is measured."^' 
If the concentration is the same right n]3 to tlie margin, the resistance will remain 
constant u]) to the time when the margin reaches the wire h, when it will suddenly 
increase, and continue to do so until both the jioints a and h are well within the 
indicatfjr. 
In fig. 5 are drawn the curves for the conductivity between the points a and b and 
for the neighbourhood of the anion margin for solutions of magnesium and copper 
sulpliates. As abscissae are plotted the position of the margin at the time of obser¬ 
vation, the position of the platinum points lieing indicated by two small circles on 
each curve. As ordinates are plotted the conductivity of the column of licpiid 
lying between the two points. 
The similarlv olitained curve fitr potassium chloride is also shown. Following tliis 
curve from left to rigid, it is seen that the conductivity, and so also the conceidration, 
is constant as long as the two points are within the s(.>hddon, a sudden and great fall 
in conductivity taking place after the |)oint b is ])assed by the boundary. This fall 
continues, and the concentration is not constant again until the two points are both 
well within the indicator solution. The curves for the magnesium and copper 
sidphates show no such constancy in conductivity within the iDOundary. On the 
contrary, it is seen tliat just as the margin apj)r(:)aches the point h a very considerable 
change is indicated. 
The cation l)oundary of magiiesium sulphate has been examined by the same 
metliod, bid ouly a very slight conductivity change was shown; this was, however, 
in the opposite direction to that detected at the anion end. 
Further and very striking evidence of the dilution that occurs at the anion 
margin for the solution of coppei' sulphate, is olitained liy simple observation of the 
solution during an experiment. Using sodium acetate as indicator tliere results a 
system in which a dark-ldue solution of copper acetate lies over the lighter blue 
solution of the suljihate ; a perfect lioundary is produced, and one at which the 
refraction and colour margins are absolutely coincident ; but the solution immediatelv 
lieneath the indicator is much holder in colour than that a T centim. lower, thus 
indicating a diminution in co})])er concentration. This difference in colour has heen 
* It is necessary that the iilatinnm points should I)e so near together that the difference in potential 
between them is not sufficient to overcome the decomposition tension of v'ater and so to cause an evolution 
of hydrogen and oxygen gases, which would result in a mixing of the two solutions. A very curious 
phenomenon has been noticed in measuring the resistances in this apparatus : in all cases the resistance as 
measured during the passage of the curi'ent is less than that found when it is measured with no current 
l)assing. and this is the case whether the latter is sent in the one direction or in the opposite. 
