Phenomena in X-Ray Bulbs. 199 



The bulb in the experiment was polarized at '050 mm., 

 and this pressure raised to *072 mm. by introducing air 

 from outside. Dashes in the columns G indicate that the 

 corresponding p.d. was not sufficient to produce a discharge. 

 The table shows that the rate of recovery of the bulb, at 

 first large, decreases with time, and that in two hours 

 the bulb practically returns to its normal condition. It 

 must be pointed out, however, that the results of different 

 sets of experiments of this kind are not very consistent with 

 each other. 



The rate of recovery of the bulb, in fact, depends largely 

 on the pressure inside ; for, if an amount of gas sufficient 

 to raise the pressure to several millimetres be introduced 

 into the bulb from outside, and then rapidly pumped out, 

 the bulb returns at once to its normal condition ; while, 

 on the other hand, the same stage of polarization may 

 persist for several hours if the bulb be kept well exhausted. 



After a bulb has recovered to its normal condition, it may 

 easily be brought back to a state of polarization by a com- 

 paratively short run. 



Part played by Gases in the Bulb. 



4. It seemed at first natural to suppose that polarization is 

 due entirely to changes in the nature and the properties of 

 the gases brought while the discharge takes place. It might 

 be, for instance, that the gases evolved from the electrodes 

 consist of helium or neon whose sparking potentials at 

 sufficiently low pressure are much higher than that of air, or 

 that the gaseous molecules exhibit some properties of fatigue 

 to the process of ionization, since after a sufficiently long- 

 run the same molecules necessarily have to be ionized several 

 times. Various experiments carried out in this direction 

 have shown, however, that the gases themselves play but a 

 small role in the phenomenon. i^'or firstly, as shown above, 

 after the introduction of air from outside the bulb still 

 remains polarized. Moreover, in some experiments a minute 

 quantity of air was allowed to leak through an insufficiently 

 greased tap, and the state of polarization was still reached in 

 spite of the fact that under these conditions there was a 

 continuous flow of fresh oas from outside. The most con- 

 vmcmg experiments in this direction were made by connecting 

 the bulb to the side tube R. A state of polarization was 

 reached where the highest voltage supplied by the coil could 

 not produce a discharge through the bulb at a pressure 

 of '060 mm., and it was then found that a potential difference 



