524 



SCIENCE 



[N. S. Vol. LIV. No. 1404. 



age is to reduce the number of electrons 

 caught and to increase the sharpness of the 

 pattern. On decreasing the bombarding poten- 

 tial without altering the ratio of retard- 

 ing to bombarding potential the lateral maxi- 

 mum moves away from the primary beam to- 

 ward the plane of the target, and the ratio 

 of the intensity of this maximum to that of 

 the other becomes greater. 



In attempting to interpret these results we 

 have been led to consider the scattering of 

 electrons by a positive nucleus of limited 

 field, one for which the central force on an 

 electron is Ee/r'' for values of r less than 

 p, and zero for all values of r greater than p. 

 Such a field would exist for a concentrated 

 positive charge E surrounded by a spherical 

 shell of uniformly distributed charge — E and 

 of radius p. The field of a system compris- 

 ing a central positive nucleus of n electronic 

 charges surrounded by n electrons uniformly 

 distributed over the surface of a sphere of 

 radius p will also be roughly of this nature, 

 provided n is not too small. Neglecting the 

 change of mass of the bombarding electrons 

 while traversing the field within the shell it 

 turns out that when such a system is under 

 random bombardment by electrons approach- 

 ing on parallel lines, the number of these 

 emerging per unit solid angle in a direction 

 making an angle * with the path of the inci- 

 dent beam is given by 



'* ==^'((2^-1)^1" 

 where 



;— 1 



-f- cos ^ ) -t- ( 1 — COS 



*)j 



Vp 



' E ' 



V being the potential drop through which 

 the bombarding electrons have acquired their 

 speed. 



An examination of this expression shows that 

 when p is very large the intensity of scattering 

 will be small in all directions except in and 

 near the direction * ==. '^, that is, in the direc- 

 tion of motion of the incident electrons. As 

 /? decreases the emerging electrons are less con- 

 centrated in this direction. For /3^1 the 

 distribution becomes entirely independent of 

 angle. As /? decreases from unity to the value 



one half the scattered electrons become more 

 and more concentrated in and near the direc- 

 tion * = 0, the intensity in this direction be- 

 ing infinite for jS = 1/2. For values of ^ less 

 than 1/2 the distribution curves for the range 

 1 > ^ > 1/2 are identically repeated, the dis- 

 tribution approaching uniformity in all direc- 

 tions as ^ approaches zero. 



For a neutral system of two or more concen- 

 tric shells the distribution will be broken up 

 into various beams or lobes corresponding to 

 groups of electrons whose trajectories pass 

 through one, two or more of the shells. In par- 

 ticular a system comprising two shells will 

 give, in an appropriate range of bombarding 

 potentials, distribution curves similar to that 

 shown in Fig. 1. 



All of the main features of the distribution 

 curves so far observed for the scattering from 

 nickel seem reasonably accounted for on the 

 supposition that a small fraction of the bom- 

 barding electrons actually do penetrate one or 

 more of the shells of electrons which are sup- 

 posed to constitute the outer structure of the 

 nickel atom and, after executing simple orbits 

 in a discontinuous field, emerge without appre- 

 ciable loss of energy. 



If the theory of the scattering here pro- 

 posed proves to be the correct one, there seems 

 no reason why the careful study of such dis- 

 tribution curves as shown in Fig. 1 may not 

 reveal much of interest concerning the dis- 

 position of electrons within the atom. It is 

 hoped to report more extensively on this work 

 in the near future. 



C. Davisson, 



C. H. KUNSMAN 



Eesearch Laboratories or the American 

 Telephone and Telegraph Company aot) thb 

 Western Electric Company, Inc. 



the atomic weight of boron 

 The application of positive-ray analysis by 

 Aston^ has yielded the evidence of existence of 

 two isotopes of boron with atomic weights 

 10 and 11, in accordance with the prediction 

 of Harkins.- Although the result of Smith 

 iPhil. Mag., 40, 628 (1920). 

 2 Jour. Amer. Chem. Soc, 4S, 1988 (1920). 



