REVERSIBLE ANT) IRREVERSIBLE SYSTEMS UNDER INFLUENCE OF LIGHT. 365 
II. The System f Ag-BrAg in Light, Decinormal NaBr Solution in Light, Decinormal 
NaBr Solution in the Dark, Ag-BrAg in the Dark). 
On this system the properties of constant cells reversible in respect of the anion 
were studied in great detail, the deflections obtained being very large and suitable for 
quantitative investigation. The AgCl was removed from the Ag plates with a 
CNK solution, the Ag plates polished again, covered on one side with shellac and 
paraffin, then immersed into a solution of Br 2 in water for a short time. A very thin 
layer of AgBr was thus deposited on the same. 
These plates were first used on July 25 and 27, 1903, for the investigation of the 
system (Ag-BrAg plate in light, KN0 3 solution, Ag-BrAg in the dark), which is an 
inconstant cell, and showed polarisation. The same Ag-BrAg plates, carefully washed, 
were now placed in the quartz vessel containing a decinormal NaBr solution and solid 
AgBr, and a constant cell was obtained. Arc and acetylene light were used; the last, 
however, almost exclusively on account of its very much greater constancy. 
(1) Plate Nil of July 28, 1903 (3 Curves). —Acetylene light used. For experi¬ 
mental data, see Table III., pp. 392-393. Forty-five minutes after the circuit was closed 
the E.M.F. in the dark was still = 18 centims. to the left = 167 x 10“ 6 volt. Though 
it lias not yet reached its constant minimum, 3 curves were taken in succession to 
see the character of the results immediately after the circuit is closed. The curves 
did not return to zero deflection in the dark. During the time a curve was taken, 
the E.M.F. in the dark shifted to a new point. The deflections obtained in Nl, 2, 3 
were 60 millims., 74 to 75 millims., 74 to 75 millims., the distances between a-b and d-d 
in Nl, 2, 3 were 11 millims., 20 millims., 20 millims. Thus the results corrected for 
the variation of the E.M.F. in the dark are : 49 millims., 44 to 45 millims., 45 to 46 
millims., = average 46'3 millims. = 49'2 x 10 -9 ampere = 43‘2 x 10 -f> volt. 
(2) Plate Nl2 of July 29 («) contains 9 Curves, Table III. —-The plates from 
July 28 were left short-circuited till the next day : the E.M.F. in the dark found 
to be = 1'5 centims. = 14'2 x 10~ 6 volt to the right (i.e., in an opposite direction to 
that of the previous day). On opening and shutting the circuit, the spot of 
light always returned to the same position on the scale. The acetylene light was 
in all experiments at the same distance of 116 centims. from the quartz vessel; the 
intensity of light coming from the burner was changed by screening more and more 
the light with the chimney of the burner and measuring its intensity with the 
Rubens thermopile. The deflections obtained with Nl # , 2, 3, 4, 5, with the intensity 
of the acetylene = 71 millims. deflection, were 84, 81, 82, 80, 81 millims., 
average = 81'6 millims. = 867 x 10~ 9 ampere = 76‘1 x 10~ 6 volt. The curves 
return after the deduction period (if we wait sufficient time) to their former position 
in the dark, as is well illustrated by curves N2 and N4 ; in N3 and N5 we did not 
* After the maximum deflection was reached, screen of chronograph fell down, so that the deduction 
period was not photographed. 
