738 Mr. J. Chad-wick on the Charge on the 



The lightest scattering atom used in the present ex- 

 periments was that o£ copper, and the value of cf> was 

 about 29°. It is easily seen that the correction due to 

 motion of the nucleus is negligible, being, in fact, only 

 one part in 20,000. 



The determination of the nuclear charge consists, there- 

 fore, in the measurement of the number of a particles in 

 the original pencil and the number in the scattered pencil 

 under known geometric conditions. The scattered particles, 

 however, are an extremely small fraction of the original 

 pencil; and, as was pointed out above, it is this which 

 provided the chief difficulty in Geiger and Marsden's 

 experiments, necessitating the use of different methods of 

 measurement in the two cases. 



In the experiments described in this paper both the 

 scattered and the direct particles were counted on the same 

 zinc-sulphide screen under the same conditions. This was 

 made possible partly by the geometry of the arrangement 

 adopted, but mainly by the aid of a new method of counting 

 the direct pencil. 



The fraction of scattered particles was increased by using 

 as scattering foil an annular ring subtending a fairly wide 

 cone at the source. There is no objection to the use of a 

 wide angle provided that the method adopted is amenable 

 to accurate calculation, for the law of scattering with angle 

 has been verified to a high degree of accuracy. 



In fig. 1 let R be the source of particles, AA' the 



scattering foil, and let the zinc-sulphide screen be placed 

 at a point 8 on the axis of the cone EAA', such that 

 RA = AS. 



The solid angle subtended at R by an elementary annular 

 ring at P is 2tt sin 0/2 . d<£/2. 



If Q is the number of a particles emitted per second by 



