177 



Fig. 6. 



In the second experiment the proportion of ga^ to air, by 

 pressure, is as 88'j to IIT/ : but, according to the energy spent 

 by the a particle, as 885 to ll'5J2'2, i.e., as 94*5 to 5'5. 

 Hence the true value of HI is found from the equation 



•945 lU +'055x213 = 289 

 whence it!Z = 294 

 Hence specific ionisation of 082=294/213 = 1*38. 



The results of this particular experiment are recorded 

 in the second line of the results for carbon bisulphide in § 2. 



It seems probable that the determination of the ionisa- 

 tion in various cases due to the a rays may be of consider- 

 able importance, and I therefore attach a drawing of the 

 apj^aratus which I have used (pi. vii.) in the hope that it may 

 save the time of any other workers in this direction. 



In the figure, P is cne of three glass pillars which sup- 

 port the high potential plate. I have also used a glass plate, 

 as shown by the dotted surface, to insulate the upper plate 

 of the ionisation chamber. Sulphur and ebonite do not 

 stand the heat. The upper gauze, gg, is the lower wall of 

 the chamber, (/'</' is the lower gauze and is earthed ; it is sup- 

 ported by three brass pillars, only one of which is shown. The 

 vertical tubes are shown at TT, and the radium plate at RF. 

 The semaphore, xn, is made of thin sheet copper, and can be 

 turned round so as to uncover the radium. It may be worth 

 while mentioning that I have found it better to keep the 

 plate, QQ, 'out of sight" of any insulating material connected 

 with the high potential plate : if this is not done, then the 

 creep of electricity over the insulators which is apt to occur 

 when the potential is changed exerts a troublesome electro- 

 static effect. DEFG is the outline of the electric oven. The 

 tube A goes to the manometer, B to a bulb used in the deter- 

 minations of stopping power, and C to a bulb which contains 

 the liquid whose vapour is being treated. 



