538 Dr. L. B. Loeh on the Recoil of 



ground glass plate. It was lined throughout with graphite. 

 The screen o£ ZnS was made on the inside of the glass 

 window Z by sitting fine crystals of zinc sulphide on to it 

 when slightly moistened with a dilute solution of collodion 

 in amyl acetate. The lead block B faced with thin graphite 

 and with the corners bevelled, so that the alpha particles 

 deflected by them could not, reach the ZnS screen, served to 

 reduce gamma radiation. The distance from the centre o£ S 

 to the centre of T was 2" 5 cm. while that from the centre of 

 T to the point on Z, where counts were made, was about 

 4*5 cm. 



The counting was done as follows. The source after 

 removal from the emanation had its activity quickly deter- 

 mined by a gamma-ray electroscope. It was then placed in 

 position in the tube A, and the latter was placed in position on 

 the plug P. The target T was not in the box at this juncture. 

 The box was closed by the glass plate and carefully exhausted 

 to about 8 mm. pressure, so as not to break the mica window 

 by sudden pressure differences. A series of six to eight 

 counts was made to determine the number of stray particles 

 and to eliminate contamination in case it was present. This 

 control count was made without a target, as the number of 

 stray particles had to be determined, and a graphite target 

 deflected particles to the screen to a greater extent than the 

 box behind it. The stray particles were but few in number, 

 e. g. about 1*5 per minute per 10 mg. RaC, which number 

 even then was as great as the number of particles deflected 

 from an Al target. The stray particles had a characteristic 

 fragmentary sort of appearance similar to that produced by 

 alpha particles near the end of their rangv. They had a 

 range of about 4 mm. of air. Contamination due to a break 

 in the mica was at once recognizable by the brilliant scintil- 

 lations in contrast to the faint ones due to the stray particles. 

 The counting actually began as a rule within 17 minutes of 

 the removal of the source from the emanation. 



The air was next let in gradually, the glass plate removed, 

 and the target T placed in position. The glass was replaced 

 and the air again slowly exhausted to the same pressure as 

 before. Eio-fit to twelve counts were made with this 

 arrangement. Air was then again let in and the source 

 removed. After again exhausting a series of six counts was 

 made to eliminate the possibility of contamination having 

 taken place during the other operations, or through a conta- 

 mination of the target. The count was as a rule negative, or 

 gave one particle in two or three counts. By this time the 

 .source had decav^d to about one-third its initial activity. 



