Energy Distribution in Spectra. 55 



electroscope * e, so that a negative current of electricity 

 flowed to the leaf when light entered the cell. 



The electroscope was used as a null instrument. To effect 

 this, arrangements were made to balance the photo-current 

 by a current of positive electricity, so that the leaf should 

 remain at zero potential. The method used is shown in fig. 1 . 

 A rectangular glass tank 11 100 cm. long, 6 cm. deep, and 

 1 cm. wide was filled to a height of 25 cm. with a mixture 

 of saturated copper sulphate and glycerine. (Jopper elec- 

 trodes were inserted at each end, that on the left being 

 earthed, the other kept at a potential of +40 volts. An 

 insulated vertical copper tongue £ dipped in the liquid and was 

 mounted on horizontal rails so that it could travel the whole 



length of the tank. This tongue was connected to the 

 © © 



gold-leaf through a high resistance r ot: 6'87 x 10 n ohms f. 



Hence if xbe the distance in cms. between the tongue and 



© 



the left-hand electrode, the potential of t is rArz x 40 volts, 



and the current flowing to the gold-leaf (which we suppose 

 earthed for the moment) is 



IM x40X 6W7I(F arapereS: 



that is, the current is proportional to x. 



Hence if the tongue be moved so that this current just 

 balances the photo-electric current, the latter amounts to 

 x x 5*83 X 10~ 13 ampere. Observations were taken by moving 

 the tongue till the gold-leaf was at zero potential, the photo- 

 current being then proportional to the reading of the tongue 

 in cms. 



Intensity Ratios in the JB aimer Series of Hydrogen, 



Pure H 2 vapour was admitted to the Geissler tube and 

 the pressure measured by an oil-gauge. An alternating 

 current of 15 milliamperes was passed through the tube, and 

 the photo-currents due to H a , H^, H y , H$ measured. The 

 observations were repeated at various pressures with the 



* Beatty, Phil. Mag. Nov. 1907, p. 606. _ 



t This resistance was made by depositing platinum by cathode dis- 

 charge on a quartz rod 1 cm. long, whose ends were copper plated and 

 soldered to copper wires, the whole being contained in an evacuated 

 glass tube. The apparatus was made three years ago (Beatty, Proc. Roy. 

 Soc. Al lxxxix. 1913, p. 318), and its resistance has increased since then 

 from 5-57 X 10 10 ohms to 6*87 X 10 u ohms. 



