536 Dr. L. B. Loeb on the Recoil of 



pressure, by coating the source with a thin layer of shellac, 

 by electric fields, and by enclosing the source in a tube with 

 a thin collodion window of 3 mm. air absorbing power, but with 

 no success. The source was finally placed in a brass tube 

 which fitted on to a ground brass stopper perforated by a 

 tube leading outside of the box. The end of the brass tube 

 towards the target was closed by a thin mica window of 1 cm. 

 air equivalent. Thinner mica was found too fragile. The 

 collodion might have succeeded had time permitted the deve- 

 lopment of a technique for attaching the collodion to the 

 tube with a gas-tight joint without injury to the collodion 

 diaphragm. 



5. Finally, precautions had to be taken with the low counts 

 to reduce and correct for the number of stray particles. 

 These were presumably alpha particles deflected through 

 small angles from the carbon walls and floor of the box. 

 They were always present even in the absence of the target. 

 They were reduced to about 1*5 per minute per 10 mg. RaC 

 by use of a slanting source and appropriate screens. 



Apparatus and Procedure. 



When work was begun, the writer decided to count scin- 

 tillations from deflexions at 120° for carbon and aluminium. 

 This angle was chosen so that the number of particles should 

 be great enough to be easily counted. This angle permitted 

 higher counts, as it permitted short distances from source to 

 target and target to screen, at the same time giving a fair 

 distance between source and screen to reduce gamma-ray 

 effects. The distance originally used had to be increased 

 because of gamma rays, and with this increase it was decided 

 that work at 170° might after all be possible. At that time, 

 however, the preliminary results on aluminium had been 

 obtained, and in the short time remaining it was felt best to 

 verify these results rather than attempt to start afresh with a 

 new apparatus. 



The apparatus is shown in fig. 1. In a brass box, B, 5 cm. 

 by 10 cm. and 40 cm. deep, was placed the source S. This 

 was a brass disk 13 mm. in diameter with bevelled edges set 

 at an angle of about. 30° with the direction of the projected 

 alpha particles. It was enclosed in the tube A covered at the 

 end towards the target by the mica window M and fitting 

 on to the ground brass plug P. The plug was perforated, and 

 the inside of the tube could be exhausted simultaneously 

 with the box B through the tube C by the same pump. The 

 target T was a plate or piece of foil of the substance to be 



