Phenomena in X-Ray Bulbs. 195 



anode A and the cylinder B are surrounded by water-cooling 

 jackets W. The bulb is connected to a discharge-tube R, 

 14 cm. long and 5 cm. diam., with two large aluminium 

 electrodes sealed into tfie tube and provided with water 

 jackets. These electrodes are identical in size and shape, so 

 that the current in the tube may be reversed without 

 changing in any way the conditions of the discharge. The 

 whole is connected to a Gaede mercury pump and McLeod 

 gauge. 



The apparatus always contains some vapour from sealing- 

 wax and tap grease, which, if not constantly pumped out, 

 slowly accumulates, reaching in one or two days a maximum 

 value of about *015 mm. ; but it could be made so airtight 

 that this maximum pressure was never exceeded, even when 

 the bulb was not pumped out for several weeks. The 

 accumulation of the vapour is fortunately a slow process, as 

 is shown by the fact that with the pump constantly working, 

 a pressure could be easily reached which is too small to 

 detect by the McLeod gauge. The presence, however, of 

 mercury vapour in the apparatus was unavoidable. 



Under the discharge large quantities of gases are constantly 

 being evolved from the glass and metal parts. The evolution 

 of these gases, although diminishing with usage, always 

 remains appreciable, necessitating the constant use of the 

 pump during the discharge. 'By varying the speed of the 

 pump and controlling the stop-cocks, the pressure in the bulb 

 could, however, be regulated and kept sufficiently constant. 

 Before taking pressure readings the discharge was inter- 

 rupted, and the pump stopped in order to ;ivoid the effect of 

 spark-pressure * and to ensure an even distribution of the 

 gas throughout the apparatus. 



After the discharge through the bulb had been maintained 

 for a sufficiently long time, the pump constantly working at 

 full speed, the pressure, which at first is considerable, begins 

 to diminish, finally reaching a value of some thousandths of 

 a millimetre, and at this stage of exhaustion a beam of X-rays 

 of increasing penetrating power begins to pass through the 

 aluminium window. At the same time, the sparking 

 potential across the bulb gradually increases, and in order 

 to maintain the discharge, it becomes necessary to regu- 

 late the gas-pressure, keeping it constant above a certain 

 lower limit. 



If, now, the discharge is further maintained for a long time, 

 the following effect is observed : the sparking potential 

 -begins to increase markedly, the current through the bulb 



* J. J. Thomson, 'Electricitv through Gases,' 2nd ed. p. 516. 



'0 2 



