756 Prof. F. Horton and Miss A. C. Davies on 



radiation produced in this space could be absorbed by helium 

 atoms and re-emitted, and thus passed on into the main tube 

 by resonance. 



For convenience in describing the experiments, tiie 

 differences of potential maintained between adjacent elec- 

 trodes will be referred to as V 2 , Y 3 , &c, as indicated in the 

 figure, but the numerical values of V l} given in the paper, 

 are 'not the actual potential differences applied from the 

 battery but the energies, measured in volts, with which the 

 electrons from the filament F nass through the gauze D. 

 Similarlv the numerical values of V 5 are the energies which 

 electrons from the side filament attain at the level of the 

 gauze K ; that is to say, the numerical values of Vi and of 

 V 5 are the "corrected" values of the respective accelerating- 

 potential differences. The corrections used in the 'expe- 

 riments recorded in this paper were obtained by finding, in 

 connexion with each set of observations, the point at which 

 ionization occurred by electron impacts on normal helium 

 atoms, a point which was usually sharply marked and un- 

 m'stakable. For this purpose 25*2 volts was assumed to be 

 the most probable value of the ionizing voltage for normal 

 helium atoms. This is the value obtained by adding 4*8 volts 

 (the voltage conuected by the quantum relation with the 

 limit or the helium principal series! to our experimentally 

 determined value of the first radiation point, viz. 204 volts. 

 We regard this mean value of the minimum radiation voltage 

 as being more exactly determined in our earlier research 

 than the value 25*6 there given for the minimum ionization 

 voltage in helium. 



The existence of two fundamental radiation voltages. 



With the experimental arrangement used by Franck and 

 Knipping in their second paper on the excitation potentials 

 of helium, it is not possible to distinguish between a current 

 which is due to the photoelectric action of radiation on the 

 collecting electrode and one which is due to ionization of 

 the gas. In order to be able to do this, w r e again used 

 the method — based on that originally due to Davis and 

 Goucher * — described in the account of our first experiments 

 with helium. In investigating the existence of the .two 

 radiation voltages, the lower filament only was employed as 

 a source of electrons, and two arrangements of the electric 

 fields in the main tube were used in different experiments. In 

 the first of these, the electrons were accelerated from the lower 



* B. Davis and F. S. Goucher, Phys. Rev. x. p. 101 (1917). 



