158 Prof. E. Rutherford on Uranium Radiation and 



case of uranium conduction the phenomenon is still further 

 complicated by the want of uniformity of ionization between 

 the plates and the resulting disturbance of the electrostatic 

 field due to the excess of ions of one kind between the 

 plates. 



The ionization of the gas is greatest near the uranium sur- 

 face, and falls off rapidly with the distance. The rate of 

 recombination of the ions thus varies from point to point 

 between the plates, being greatest near the surface of the 

 uranium. 



The equations which express completely the relation between 

 the current and electromotive force for the rate of leak between 

 two parallel plates, one of which is covered with uranium, are 

 very complex and cannot be expressed in simple form. The 

 disturbance of the electrostatic field between the plates, due to 

 the movement of the ions, has to be considered as well as 

 the variable rates of recombination at the different points, and 

 the difference of velocity between the positive and negative 

 ions. 



The great difficulty in producing complete saturation, i. e. to 

 reach a stage when all the ions produced reach the electrodes, 

 may be due to one or more of three causes: — 



(1) Rapid rate of recombination of the ions very near the 

 surface of the uranium. 



(2) Presence of very slow moving ions together with the 

 more rapidly moving carriers. 



(3) An effect of the electric field on the production of 

 the ions. 



The effect of (3) is probably very small, for there is no 

 experimental evidence of any such action unless the electro- 

 motive forces are very high. That the slow increase of the 

 current in strong fields is due to (1) rather than (2) receives 

 some support from an experiment that has been recently 

 tried. Instead of measuring the current with the uranium 

 covering one electrode, the air which had passed over uranium 

 was forced between two concentric cylinders between which 

 the electromotive force was acting. The rate of leak was 

 found to only increase 2 or 3 per cent, when the E.M.F. was 

 increased from 16 to 320 volts. This increase is much smaller 

 than in the results previously given. Since the effect of (2) 

 would be present in both cases, this experiment seems to show 

 that the difficulty in removing all the ions from the gas is not 

 due to the presence of some very slow-moving carriers. 



