18 Water-house — Electrical action of Light upon Silver. [No. 1, 



be placed in front of the opening. The upper part of the wooden 

 case is open, but can be closed by a lid, through which, if necessary, 

 a funnel may be passed to admit of solutions being poured into the 

 cell without letting in light. The silver plates used with this cell are 

 4 inches long, and lj inches wide, other plates, such as photographic 

 sensitive dry plates or celluloid films, being about the same size or 

 smaller. 



The other cell is a modification of the form used by Becquerel 

 in his earlier experiments, and consists of a wooden trough divided 

 into two compartments by a double wooden screen which allows the 

 free circulation of the electrolytic fluid, while completely shutting off 

 light from the unexposed compartment. This trough is covered with 

 a lid, having two large openings fitted with hinged shutters, to the 

 underside of which mirrors are attached for the purpose of re- 

 flecting light at will on to one or other of the sensitive surfaces in 

 the compartments below. By this arrangement the whole of the 

 sensitive plate can be exposed to light, instead of only part of it, as 

 in the vertical cell, and at the same time the perfect protection of the 

 unexposed plate from strong light is better secured than it is in the 

 vertical cell. This horizontal trough is constructed to take two plates 

 31" x 4J" or smaller. 



In most cases, even under favourable conditions, the light-currents 

 observed, are exceedingly weak, and therefore a very sensitive form 

 of galvanometer is necessary. The one I have used is the latest modi- 

 fication of the Rosenthal micro-galvanometer made by Edelmann, in 

 Munich. It is said to be the most sensitive form of galvanometer 

 made, enabling currents of about a billionth of an ampere to be read 

 with a resistance in the coils of only 1,000 ohms. It is fitted with a 

 telescope by which direct readings are made off the mirror from a 

 millimetre scale placed at one metre from it. In this position and 

 without the directing magnet, using the 10 1 00 shunt, with a total external 

 resistance of about 60,000 ohms in circuit, the deflection caused by one 

 gravity- Danie 11 cell is one millimetre division of the scale. By using 

 the directing magnet the normal sensitiveness of the instrument can 

 be very greatly increased, though in most of the experiments it has 

 been found sufficiently sensitive without the magnet, and when used, 

 the increase of sensitiveness has been limited to about five times the 

 normal. The instrument can be set up in any position, is simple in con- 

 struction and I find it very sensitive, convenient in use and easy to ob- 

 serve with fair precision, considering the difficulty there is in obtaining 

 freedom from shake and tremor in a city like Calcutta built on a bad 

 foundation of mud. In reading the scale which is 50 centimetres long, 



