'Two Ion* produced in Gases by Rdntgen Radiation. 14 3 



the potential of the initially-charged plate at 40 volts, an 

 effect was obtained which amounted to a deflexion of about 

 half a volt, and which was always of the sign contrary to 

 that of the initial charge on the opposite plate, and hence 

 corresponded to a reverse or polarization current. The two 

 charges left near the plates without any external electric 

 force simply move towards each other, and so in a direction 

 opposite to the motion of the same ions during conduction. 

 If the rays were cut off as soon as the commutator was turned 

 the electrometer deflexion became permanent, but when the 

 rays were allowed to continue the charge which the insu- 

 lated plate quickly assumed, due to the polarization, was now 

 slowly dissipated by the action of the new ions formed. 



§ 10. Dependence of Blowing Methods upon the Velocity 

 of the Blast. 



It is a fact of some importance, and one which has not 

 been fully considered by some observers, that in attempting 

 to remove moving carriers from any space by a stream of 

 gas, the number so obtained may not be dependent so much 

 on the total number present as upon the relative velocity of 

 the blast to that of the moving carriers. 



Rutherford has shown (Phil. Mag. April 1897) that in 

 blowing ionized air along a charged wire which is placed in 

 a tube a maximum value is reached for the amount of free 

 electrification blown out. as the potential of the wire is 

 gradually increased. 



A number of experiments have been made with the ap- 

 paratus shown in fig. 7, which show how the quantity of 

 electrification removed from a plate by a given blast depends 

 not only upon the electric field used, but also upon the sign 

 of the ions which are being carried out. 



In the case to be considered the plates were placed 

 •I'l centim. apart, and the slits were placed as near as pos- 

 sible to one of the plates, the other of which was always con- 

 nected to earth. A blast of the same velocity was used 

 throughout, and readings were taken of the amounts of 

 electrification obtained in a given time as the potential of 

 the plate was changed in amount and in sign. 



The following table gives the results obtained in a sample 

 case : — 



