Absorption of Hard X-Rays in the Lightest Elements. 741 



average for of A], have found 0*15, which is in close 



agreement with the value above. 



According to the atom model of Rutherford-Bohr the 

 electrons are imaged as discrete particles, which are moving 

 in rings around the nucleus. In the model, recently worked 

 out by Lewis and Langmuir, the electrons, however, are 

 thought to form coaxial rings of electricity, " ring- electrons." 

 For the purpose of obtaining material to decide between the 

 two assumptions, Schott tS has worked out the theory of 

 scattering on the basis of the first theory. As a matter of 

 fact the paper gives the results for both assumptions, for the 

 ring-electron is nearly a limiting case of discrete electrons,.. 

 Yet the paper shows that the mass scattering coefficient of 



hydrogen (-h) reaches the value of 0*398, and is quite 



independent of the wave-length. As is shown in formula (6) , 



it follows from my experiments that - H for greater wave- 



P 

 lengths has a value 0*387, which thus is in good agreement 

 with the said value and the value earlier calculated by 

 Thomson. The mass scattering coefficient nevertheless is not 

 independent of the wave-length but is decreasing with 

 increasing wave-length (formula 5), which does not agree 

 with the said result of the investigation of Schott. 



In two papers on "The Size and Shape of the Electron ""f,. 

 Compton, on the basis of the atom model of Lewis and 

 Langmuir, has made a theoretical investigation for finding 

 how the scattering effect is dependent on the wave-length. 

 If the electron is thus taken as a flexible ring of electricity, 

 Compton finds that the scattering coefficient will be expressed 

 by the formula : 



. r . (1-29-61 gy + 524-2 (0-5398 g)V..) (7) 



where cr stands for the value of scattering coefficient at great 

 wave-lengths, and a means the radius of the electron as calcu- 

 lated by Compton and found to be (1*85±0 05) . 10~ 10 cm. 

 In the following table, for the effective wave-lengths (X e ) 



used by me, the values for -- derived from my experiments 



°"o 

 (<7 = 0'387) have been combined with those calculated by 



* Schott, Proc. Rov. See. xcvi. p. 410 (1920). 

 t Compton, Phys. Rev. xiv. p. 20 (1919). 



