4 Makower, Scattering during Radio-active Recoil. 



can therefore only be explained either if the quantity of 

 radium B reflected at O is very small, or if the chances of 

 recoil are greater under the present experimental con- 

 ditions than in the previous experiments. The latter 

 explanation seems to be the correct one, for we have seen 

 that radium B and radium C reach the surface S in almost 

 equal proportions, and the a-ray activity of the plate S 

 was found to have about -^^^ the activity of the surface Q 

 when tested 20 minutes after the recoil from the wire P 

 had ceased. Now it can be calculated from these facts 

 that, if a small fraction x of the radium B recoil-atoms 

 reaching Q are reflected on to the plate S, the fraction 

 of radium C recoil-atoms subsequently reaching S to the 

 total number formed on O must be about lo.r. Taken in 

 conjunction with the fact that the activity of O was only 

 twenty times that of S, this result leads to the conclusion 

 that the proportion of radium C atoms which succeed in 

 recoiling from the surface O is greater than the fraction 

 (one thousandth) previously obtained. Though it is not 

 possible to be quite sure of this deduction from the above 

 evidence, the conclusion is not unreasonable since the 

 atoms of radium B deposited from the wire P by recoil 

 are lightly distributed over the surface Q without any risk 

 of being covered by surface films as might easily be the 

 case with any other method of deposition. The whole 

 question of the scattering of recoil-atoms is at present 

 receiving more careful examination. 



