Conducting Gases on the Direction of the Electric Field. 225 



of ions near one electrode, the ratio of the velocities is equal 

 to the ratio of the currents in the two directions, and that the 

 current is proportional to the square of the voltage. 



Explanation of the Results. 



The theory given above suffices to offer a general expla- 

 nation of the results given when the ionization is produced 

 by radium or Rontgen-rays. For the thin stratum of Rontgen- 

 rays, the current varies roughly as the square of the P.D. 

 between the plates, and the currents vary directly as the 

 velocities of the ions. It is probable that with 1 15 volts 

 between the plates the ionization was not sufficiently intense 

 to fulfil the condition that the potential-gradient was very 

 small near the piate A. 



In the experiments where radium-rays are used to produce 

 ions, the ionization is not confined to a definite layer, but falls 

 oft' gradually between the plates. In consequence of this the 

 relations between current and P.D. occupy an intermediate 

 position between the results obtained in the case of uniform 

 symmetrical ionization and those obtained with strong surface 

 ionization ; i. e., the current increases faster than the P.D., 

 but not so fast as the square of the P.D. 



With a thin layer of intense ionization confined to the sur- 

 face of one plate, the value of the velocity K 1 of the positive 

 ion is given by 



T . _ 32tt d*t\ 



Since i, is independent of the intensity of the rays when 

 this intensity exceeds a certain limit, it can be accurately 

 measured, and the other quantities on the right-hand side of the 

 equation easily determined. All that is necessary for deter- 

 mination of the velocity of the ion is to send a very strono- 

 beam of Rontgen-rays close to one plate, and then determine 

 the current between parallel plates for a definite P.D. With 

 a bulb giving out strong rays, there is no practical difficulty 

 ia producing sufficiently intense ionization near the surface 

 of one electrode, such that with 30 volts between the plates the 

 potential-gradient close to the electrode may be small enough 

 to satisfy the theoretical conditions. 



Practical Drawbacks of the Method, using Rontgen-Rays. 



The method given above seemed at first sight a very satis- 

 factory one for determination of the velocities of the positive 

 and negative ions produced by Rontgen-rays in different 



Phil. Mag. S. fi. Vol. 2. No. 8. Aug. 1901. Q 



