Recoil Atoms in Ionized Air. 727 



The following explanation of this phenomenon seems 

 probable. The distance that the atoms recoil in air is only 

 a small fraction of a millimetre at atmospheric pressure. 

 Since the path is equally probable in all directions, many of 

 the atoms are projected along or close to the surface of the 

 plate ; and if the plate has any projections, however small, 

 some atoms will strike them and either lose their charge 

 or adhere to the plate, so that they fail to be drawn to the 

 negative plate. With smooth plates fewer atoms are so 

 obstructed in their path. 



Under the most favourable conditions in air at atmospheric 

 pressure, the amount of actinium D obtained by recoil was 

 less than one-half of the theoretical maximum deduced from 

 the /3-ray measurements of the active deposit. The distance 

 between the plates was 2 mm., and the potential difference 

 was 1000 volts. Exposure for the recoil was 30 minutes. The 

 value of the /3-radiation from the actinium D collected by 

 recoil was reduced to the time of the removal of the plate from 

 the field. About a half-hour later the /3-radiation from the 

 active deposit was measured, and its value was reduced to 

 the same time. The relation of the /3-ray activities of the 

 actinium D obtained by recoil to that in the active deposit 

 was 23 per cent. Since only one-half of the atoms recoil 

 into the air above the plate, the maximum amount expected 

 would be 50 per cent. Halm and Meitner (loc. cit.) obtained 

 a somewhat higher ratio, and Wertenstein (loc. cit.) working 

 on RaB obtained practically the whole amount to be expected 

 theoretically. It seems very unlikely that this low value is 

 due entirely to the imperfect surface of the plate used for 

 active deposit, and to the neutralization of the charge of the 

 atoms in the gas ionized by the radiations from the active 

 deposit and the recoil atoms. 



From a large number of decay curves taken during these 

 experiments, the half-period * of actinium D was found to be 

 4*71 minutes f. 



In conclusion I wish to express my thanks to Professor 

 Rutherford for his interest in the experiments and for 

 supplying me with the necessary radioactive materials. 



Physical Laboratories, 

 The Victoria University, 

 Manchester, England. 



* Alois F. Kovarik, Pkys. ZS. xii. p. 83 (1911). 



•j- Some experiments on the recoil atoms, thorium D, were also carried 

 out. The mean value obtained for its half-period is 3 07 minutes, which 

 is practically the value found by v. Lerch. 



3 B 2 



