of the a. Particles from Radium C and Thorium C. 541 



then removed, the time o£ removal being noted. The source 

 was inserted in the apparatus after the usual precautions * 

 against contamination had been taken, the apparatus pumped 

 out and dry air admitted to the pressure desired. Pressure 

 and temperature having been determined, two or three 

 measurements of the ionization current were then made by 

 noting the time taken for the spot of light to pass over a 

 definite poriion of the scale. The time at which each 

 observation was taken was noted. The a. rays were then cut 

 off by the movable aluminium screen and the remaining 

 ionization current measured. When this natural leak was 

 small compared with the a ray current, it was only 

 measured a few times in the course of the experiment. When 

 working near the end of the range, this natural leak became 

 comparable with the a ray current, and alternate measure- 

 ments of the two were made. 



The pressure was then varied slightly, and readings of 

 pressure, temperature, and current again made. The readings 

 were discontinued about two hours after the removal of the 

 source from the emanation. 



The sources of thorium C were obtained by rotating a 

 nickel disk for a few minutes in a hot solution of radio- 

 thorium. The subsequent procedure was similar to that 

 with radium C. 



§ 3. Observations. 



A great advantage of the apparatus used in these experi- 

 ments was that the measurements were direct. There were 

 no mica or other screens through which the a rays had to 

 pass before entering the ionization chamber. This chamber 

 was itself filled with air, so that the ionization was measured 

 in the same medium as that through which the a particles 

 had already passed. 



In the determination of the range, an accuracy of *1 per 

 cent, was aimed at. The range was inversely proportional to 

 the density of the air in the vessel. In the present investi- 

 gation, interest centred largely on the end portion of the 

 path of the a. particle, the pressure corresponding to which 

 was about 13 cm. for radium C. As the probable error in 

 the pressure measurements was less than *005 cm. the error 

 in range due to this cause was not more than # 04 per cent. 

 The probable error in the temperature measurements was less; 

 than *1° C., involving an error in the range of not more than 

 *03 per cent. The error in the measurement of the distance 



* Rutherford, Phil. Mag. xxxvii. p. 541 (1919). 



