the Natural Ionization in a Closed Vessel. 



223 



originally at 21 atmospheres. The pressure was reduced 

 from time to time and readings of the natural ionization 

 taken. The leak decreased with pressure till four atmospheres 

 was reached, when it commenced to rise. This effect appears 

 to be due to an emanation given off by the walls of: the 

 pressure-chamber, for on filling to 15 atmospheres and then 

 allowing the air to expand, filling again, and again allowing 

 to expand, it was found that the ionization decreased by 

 about 75 per cent, and fell in with the general trend of the 

 curve. For the higher pressures, the emanation does not 

 diffuse from the walls of the pressure-chamber into tne 

 electroscope so quickly on account of the great density of 

 the gas. For the remaining experiments then the bomb was 

 filled and emptied several times before taking each reading ; 

 this process taking so long that any active deposit from the 

 emanation would (assuming it to be that from radium) have 

 all disappeared. 



The air used was always drawn from the outside of the 

 laboratory, since it was found that air drawn from the inside 

 was more active, due to the fact that experiments with 

 radium were being carried on in the room in which the 

 liquid-air machine was situated. It was found that leaving 

 the air compressed in the cylinder for two weeks did not 

 make any observable difference in the readings. 



The insulation was tested from time to time by connecting 

 the cup C to the same potential as the gold-leaf by means of 

 the terminal K'. The rate of motion of the gold-leaf was 

 never altered by more than 2 or 3 per cent. This small 

 variation is to be expected, since when C is charged some 

 ions will be drawn to it which before went to the gold-leaf 



system. 



Fiar. 4. 

























\ 

























" t * 





















4---^: 



z^- — 

















..-. 



"11^ 



~~-~z" 

















,„. 



,--'~^S- 























*^~ 





















f 



.' 



■' 























1 / 



■— 

















Pbcssur 



r av Atm 



OSPHEPCS 



.„ 



The final curve obtained connecting ionization with 

 pressure is given in fig. 4. It is seen to be a straight line 

 which curves over somewhat towards the pressure-axis at 

 low pressures. 



