

1904.] produced by the Action of Light. 371 



{analogous to van't HofFs " principle of moveable equilibrium " for heat) 

 is thus proved.* 



(2) The phenomena of reversal of the current, etc., observed by 

 Becquerel and Minchin, are due, not to surface phenomena or different 

 " thicknesses of sensitive layer," but to the fact that their combinations 

 all give inconstant cells showing polarisation, a phenomenon which they 

 a,nd other observers entirely overlooked, since they made no study of 

 the induction and deduction periods. Whenever a consideration of the 

 composition of the heterogeneous system, and of the reactions going on 

 in the same under the action of the current, shows that an inconstant 

 <^ell should be formed, the curve illustrated by fig. 2 is obtained, the 

 peculiar course of the induction and deduction periods being caused by 

 the principal E.M.F. and the E.M.F. of polarisation being set up 

 simultaneously under the action of light, with different velocities, and 

 diminishing on removal of the light, with different velocities, vanishing 

 almost simultaneously. Whenever a consideration of the composition 

 of the heterogeneous systems, and of the reactions going on in the same 

 under the action of the current, shows that the cell should be reversibly 

 constant, then, provided the action of light upon the plate is such as to 

 allow of the formation of a constant cell, and, provided the plates are 

 so prepared as to avoid the formation of " gas batteries " (for this the 

 E.M.F. in the dark must be made very small a few millionths of a 

 volt) we get constant cells, the induction and deduction period having 

 a normal course, as illustrated in fig. 1. 



(3) The law governing the induction period in constant reversible 

 cells is evidently 



ar 



i.e., the speed with which the system approaches, on exposure to light, 

 its new state of equilibrium in light (given by the line d e, and 

 characterised by the value of constant E.M.F., 7r ') is directly propor- 

 tional to the remoteness of the system at the given time, r, from the 

 point of equilibrium in light (given by the value TTQ - TT, if TT is the 

 E.M.F. at the time T, while TT O ' is the maximum E.M.F. in light), and 

 to the amount of variation already produced by light in the system up 

 to the time T (given by TT - TTQ, if TTQ is the initial E.M.F. of the system 

 in the dark) plus a constant K, which is also of the character of an 

 instability constant, i.e., the more the system is removed from its point 

 of equilibrium in light, the quicker it approaches it, and the more the 

 molecules and atoms have been shaken up by light (exposed to light 

 previously to the time T), the quicker the system approaches the point 

 of equilibrium in light. This gives also the fundamental law of 

 photography, the connection between the amount of silver salts 

 decomposed and the time of exposure, since the E.M.F. is, during this 

 * See ' Zeitech. physik. Chemie,' 1902, p. 332 ; ' Phil. Mag.,' 1903, vol. 5, p. 223. 



