Phenomena in X-Bay Bulbs. 197 



evacuating the bulb through a side tube fused into the middle 

 of the bulb, as shown in the figure. 



The following data, taken from one set of experiments, 

 give some idea of the time the discharge must be passed 

 through the bulb before the various stages of polarization 

 are reached. The experiments began with a new bulb, and 

 after a run of about eight hours daily during three consecu- 

 tive days, the exhaustion of the bulb became high enough to 

 produce X-rays. Two days later a penetrating beam of 

 X-rays could be observed to pass through the aluminium 

 window, the pressure in the bulb being *003 mm. and the 

 p.d. about 15,000 v. After a further run for three days the 

 polarization effect became well marked, and the pressure had 

 then to be raised to *012 mm. and the p.d. increased. On 

 the following day a degree of polarization was reached where 

 a p.d. of more than 50,000 v. could not break down the resis- 

 tance of the bulb at a pressure as high as '035 mm., and 

 a few hours later the minimum pressure at which the same 

 p.d. could produce a discharge was of the order of *060 mm. 



The current supplied by the induction-coil used in these 

 experiments was comparatively small, not exceeding 3 x 10 _i 

 amp. when the bulb was well exhausted. It was noticed, 

 however, that with larger currents the same effect is reached 

 in less time ; and it seems very probable that different stages 

 of polarization are reached only after definite quantities of 

 electricity have passed through the bulb. 



If the current be reversed, it passes freely through a polar- 

 ized bulb. Experiments 'with the discharge-tube R, where 

 the anode and the cathode are identical in size and shape, 

 show that a vacuum tube may be polarized only in one 

 direction. When the current is reversed, even for a short 

 time, the polarization of the bulb becomes greatly reduced, 

 and a run in the former direction is then necessary to bring 

 the bulb back to the same stage of polarization. Attempts 

 to carry the experiments further in order to reach a higher 

 degree of polarization were hampered by the fact that with 

 increasing pressure in the bulb it becomes more and more 

 difficult to rectify the current by means of an ordinary 

 rectifying valve or a spark-gap used in these experiments. 

 The use of a thermionic valve might remove this difficulty. 



When the discharge in a polarized bulb ceases, a small 

 current of the order of 1 per cent, of the normal value 

 may still be detected by a galvanometer. This residual 

 current persists, even at the highest exhaustion of the bulb, 

 and is probably due to conduction by the walls of the 

 bulb. 



