hy t/w Motion of Positive and Negative Jons. 359 



I may theretbre be permitted to make some remarks on 

 the points which Dr. Stark has brought forward, as some of 

 his statements are apt to be misleading. 



In the first paper which I published on ionization by 

 collision, in ' Nature,' 9th August, 1900, I called attention 

 in the first paragraph to the cathode rays and Becquerel 

 rays which have l)een known to have the property o£ ionizing 

 the gases through which they pass. The particles consti- 

 tuting these rays travel with a very high velocity, and I 

 pointed out that the chief result of my experiments was the 

 definite proof of the fact that negative ions produce others 

 by collision with neutral molecules when they are travelling 

 with comparatively small velocities. I also quoted experi- 

 ments which showed that negative ions produce others by 

 collision with neutral molecules much more readily than 

 positive ions. 



As far as I am aware, the only other experiments from 

 which a clear proof can be deduced of the genesis of ions by 

 collision when small electromotive forces are acting, are those 

 of Stoletow ^ on Ultra-Violet light. 



I called attention to these experiments in my first paper 

 as there is no publication previous to August 9th, 1900, in 

 which it is suggested that Stoletow^s experiments can be 

 explained by a collision theory. 



My first experiments with Rontgen rays were made at the 

 Cavendish Laboratory, and at that time Prof. Thomson 

 maintained that the best explanation of these experiments 

 was to be found in the theory of surface layers which he had 

 given in explanation of Stoletow's experiments f, as he con- 

 sidered that the potentials which I used were too small to 

 give rise to ionization by collision. In support of this view 

 Prof. Thomson quoted a paper by Rutherford and McClung 

 (which he communicated to the Royal Society on the 15th 

 June, 1900), from which it could be deduced that it would 

 require a fall of potential of at least 175 volts in order to 

 produce ions by collision. 



In dealing with the subject of ionization by collision I 

 have from the first considered the possibility of the action of 

 the positive ions. I have recently made some experiments 

 in order to obtain some definite information on this subject. 

 An outline of the method which I have used is given in the 

 ' Electrician ^ (3rd April, 1903). The investigations establish 

 all the essential points which the theory demands, and give 



* Stoletow, Journal de Physirjue, Series 2. Tome ix. 1890. 

 t J. J. Thomson, Phil. Mag. Dec. 1899. 



