1896.] X Rays on the Contact Electricity of Metals. 335 



sulated quadrants of the electrometer. This plate is movable in a 

 direction perpendicular to its plane, and can thus be drawn away 

 from the tinfoil. If there is any electric potential difference 

 between the opposing surfaces of the two plates, further separa- 

 tion causes a change in it which, reacting on the electrometer, 

 deflects it. 



§ 6. When everything was in position, before starting the rays I 

 measured the contact potential of the plates by the method mentioned 

 above, and found it to be 



+ 0-44 volt, 



the zinc being positive to the tinfoil. The rays were now turned on, 

 &o as to pass through the tinfoil plate on to the zinc. The contact 

 potential, measured while the rays were passing, was about 



+ 0*43 volt. 



But before the plates could be separated to see whether the counter 

 potential applied had annulled the contact potential or not, the charge 

 leaked away as if the insulation of the tinfoil plate were bad. Also I 

 observed that when the plates were not in connexion with one 

 another, but only with the electrometer, they seemed to act as though 

 they were connected together by a bad conductor or an electrolyte. 



§ 7. The X rays were now turned on so as to pass perpendicularly 

 through the tinfoil plate on to the zinc. In a few minutes the deflec- 

 tion indicated 



—0-50 volt. 



and remained steady there for several minutes. While in this con- 

 dition I separated the plates. The deflection increased by a small 

 amount, showing that the electrolytic power of the air while the rays 

 are passing through it had somewhat more than counterbalanced the 

 •contact potential of the plates, and had established a slight difference 

 of potential in the direction opposite to that which exists between 

 the surfaces of the plates when connected metallically in air. This 

 was confirmed by a measurement, made immediately after, by the 

 null method, which gave 



-r-044 volt 



for the potential difference between the plates ; the rays not being on 

 at the time. 



§ 8. In the experiments described above, the rays had fallen per- 

 pendicularly on the plates, passing through the tinfoil (see diagram). 

 I now placed them so that the rays should pass between them in a 

 direction approximately parallel to their surfaces. On breaking the 

 metallic connexion between the plates and turning on the rays the 

 electrometer rose to 



-0-39 volt, 



