768 Prof. F. Horton and Miss A. 0. Davies on 



From the experiments described in this, and in the pre- 

 ceding section, it follows that much more ionization of 

 helium results from the action of its radiation upon abnormal 

 helium atoms, than from electron collisions with these 

 abnormal atoms. 



In connexion with the explanation just offered, of the 

 ionization detected for values of Y 1 less than 252 volts, it 

 must be mentioned that in the case of the continuous curve 

 of figure 7, when Y l was increased to 25*2 volts and beyond, 

 the beginning of the occurrence of ionization as a primary 

 effect of electron impacts with helium atoms was indicated by 

 a decrease in the measured positive current. This is shown in 

 the continuation of the curve on the right-hand side of the 

 figure. The current continued to decrease with increasing- 

 values of Vi for several volts, but subsequently gave place to 

 an increasing positive current. When direct ionization by 

 collision sets in, there is a decrease in the number of 

 20*4 volts and 21 '2 volts collisions which occur, and though 

 there is an increase of positive current due to the direct 

 ionization, this is probably partially compensated by a 

 reduction in the positive current due to the ionization of 

 20*4 volts abnormal atoms by 21'2 volts radiation. More- 

 over, some of the 21*2 volts radiation which was previously 

 used up in ionizing 20'4 volts abnormal atoms, may, by the 

 reduction in the number of these abnormal atoms, be avail- 

 able for photoelectric action on the electrodes. Such photo- 

 electric action would result, in the circumstances of the 

 curve, in an increase of negative current or in a reduction of 

 positive current. Thus, when the ionization point is reached, 

 the relative extent to which the above three effects occur 

 determines the character of the bend in the curve at this 

 point. The downward bend of the curve at 25*2 volts in 

 figure .7 is presumably due to the fact that the reduction in 

 the amount of indirect ionization, and the increase in the 

 photoelectric current, more than counterbalanced the direct 

 ionization produced. An increasing positive current is 

 obtained on further increasing Y x because eventually the 

 number of 25*2 volts collisions predominates over the number 

 of 20*4 volts and 21*2 volts collisions. That an increase of 

 photoelectric current does sometimes occur at the ionization 

 point was shown by certain radiation (K) curves, obtained 

 with a fairly high pressure of helium, in which a marked 

 increase of negative current took place at this point. 



