282 Scientific Proceedings, Royal Dublin Society. 



polarisation which shows a rapid recovery, and which must be due to a 

 surplus of ions over that required for equilibrium. The conditions existing 

 resemble those in an ionised gas close to an electrode. The rate at which 

 ionisation disappears in a gas is represented by 



dn 



w = - an ' 



where n is the number of ions of one sign per unit volume. This suggested 

 a similar relation for the ions in the polarisation layer, and an attempt 

 was made to see if a relation such as 



f-^ ■» 



held good for the disappearance of the polarisation of the cell. Here P must 

 be taken as representing that part of the polarisation which is due to the 

 excess of the ions, and is associated with the rapid recovery of the cell. 

 For brevity we shall term this the ionic polarisation. 



Owing to the presence of the slow recovery, it is difficult to assign a 

 precise value to P. If there were no concentration polarisation, the voltage 

 would quickly rise to a stationary value V s , but, owing to its presence, 

 this is only approximately true, as is shown by the recovery curve. The 

 E.M.F. becomes almost steady after 10 or 15 minutes, but a slow creep persists 

 for a long time afterwards. Writing 



F = initial E.M.F. at the instant of opening the circuit; 

 V = E.M.F. of cell at any subsequent time ; 

 V s = quasi-stationary value of E.M.F ; 

 P =V S -V; 



Equation (1) then gives 



(V s - V)-'=at + {V s - V a y\ (2) 



The linear graph in fig. 1 shows that it is possible to select a value V s 

 which brings the observed values of [V, - Vf 1 to a close fit with a straight 

 line over the period of rapid recovery. The most suitable value of V s can be 

 determined by trial within O'Ol volt. 



This result gives strong support to the analysis of the decay of polarisation 

 into two processes. The close agreement with the relation required by (2) 

 shows that a simple process is at work during the rapid rise, and that the 

 slow creep which persists for a long time afterwards has a very small effect 

 on the initial observations. V s should be the asymptote to the recovery 

 curve. This is, however, only approximately so, but it is of assistance in 



