148 Prof. E. Rutherford on Uranium Radiation and 



This result shows that the ions in Eontgen and uranium 

 conduction move with the same velocity and are probably 

 identical. The velocity of an ion in passing through a gas 



is proportional to — , where e is the charge carried by the ion, 



and m its mass. Unless e and m vary in the same ratio it 

 follows that the charge carried by the ion in uranium and 

 Eontgen conduction is the same, and also that their masses 

 are equal. 



It was found that the velocity of the negative ion was 

 somewhat greater than that of the positive ion. This has 

 been shown to be the case for ions produced by Eontgen 

 rays *. The difference of velocity between the positive and 

 negative carrier is readily shown. The rate of leak of B is 

 observed when charged positively and negatively. When B 

 was charged positively the rate of leak measured the number 

 of negative ions that escaped the electrode A, and when 

 charged negatively the number of positive ions. The rate of 

 leak was always found to be slightly greater when B was 

 charged negatively. This is true whether A is charged 

 positively or negatively, and shows that there is an excess of 

 positive ions in the gas after passing by the electrode A. 



The difference of velocities of the ions can also readily 

 be shown by applying an alternating electromotive force to 

 the electrode A sufficient to remove a large proportion of the 

 ions as the air passes by. The issuing gas is always found 

 to be positively charged, showing that there is an excess of 

 positive over negative ions. 



A large number of determinations of the velocities of the 

 uranium ions have been made, with steady and alternating 

 electromotive forces, w T hen the air passed between concentric 

 cylinders or plane rectangular plates. In consequence of the 

 inequality of the velocity of the current of air over the 

 cross-section of the tube, and other disturbing factors which 

 could not be allowed for, the determination could not be 

 made with the accuracy that was desired. For an accurate 

 determination, a method independent of currents of air is 

 very desirable. 



§ 15. .Potential Gradient between two Plates. 



The normal potential gradient between two plates is altered 

 by the movement of the ions in the electric field. 

 Two methods were used to determine the potential gradient. 



* Zeleny, Phil. Mag. July 1898. 



