REVERSIBLE AND IRREVERSIBLE SYSTEMS UNDER INFLUENCE OF LIGHT. 373 
The use of ice or hot water and stirrer enabled us to cool or heat up the bath 
quickly, and the 70 litres water of the same enabled us to keep the temperature 
constant for longer periods. Experimental data, see Table V. 
(«) The System Ag-BrAg Plates in OH normal KBr Solution (8th, 9th, and 13 th 
July, 1 904). 
July 8.—The deflections (200 ohms shunt of galvanometer) were : 
In X1 = 75'4 millims., in N2 = 75‘4 millims. at 38 C T C. (at 3 p.m.). 
In N3 = 81'7 millims., in N4 = 81'7 millims. at 20 o, 8 C. (at 5 P.M.). 
July 9.—The E.M.F.’s in the dark = 6 millims. to the right (no shunt) in N l and 
N2, = 5 millims. to the right (no shunt) in N3 and N4. The deflections obtained (200 
ohms shunt of galvanometer) were : 
In Nl = 107 millims., in N2 = 108 millims. at 2 ’8 C. (at 12.30 p.m.). 
In N3 = 102 millims., in N4 = 101 millims. at 27 0, 1 C. (at 1.30 P.M.). 
July 13.—-The E.M.F.’s in the dark (200 ohms shunt) = 5‘5 centims. to the right 
in N1 and N2, = 1*7 centims. to the right in N3 and N4, = 1'5 centims. to the right 
in N5 and N6. I he deflections obtained (200 ohms shunt of galvanometer) were : 
In Nl = 133 millims., in N*2 = 133 millims. at 48°*5 C. 
In N3 = 136 millims., in N4 = 133 millims. at 33 0, 5 C. (at 12.30 p.m.). 
In No = 140 millims., in NG = 136 millims. at 2 0, 3 C. (at 3.40 p.m.). 
II the results are taken for the intensity of light of the previous experiments = 48'3 
to 48'5 millims., 6 per cent, must be added, and we get : 
In Nl = 141 millims., in N2 = 141 millims. at 48° C. 
In N3 = 144 millims., in N4 = 141 millims. at 33 Q, 5 C. 
In No = 148 millims., in NG = 144 millims. at 2 0, 3 C. 
The results obtained with the decinormal KBr solution thus are the following:— 
(1) On the 8th, 9th and 13th of July the three series of results differ from one 
another, the E.M.F. of the combination (with freshly covered plates) becoming again 
greater when it was left in the dark for longer periods after illumination. 
(2) In each series the deflections obtained, differ only by a few millimetres, i.e., 
the variation of the P.M.F. created by light with temperature can only be small 
(&) The System Ag-BrAg Plates in Off normal LiBr Solution (July 20, 1904). 
The Ag-BrAg plates used here were those used above, but subjected between the 
15th and 19th of July to the action of the solutions of HBr (to be dealt with later 
