the Absorption of X- Rays. 203 



bv the Barkla and White determinations of - for Al 



s P 



and H 2 0, the assumption that - for all light elements 



has the same value cannot be considered to be in harmony 

 with reality; and still less, I think, can this be the case 

 for the heavier elements, as was assumed by Hull and 

 Rice. So it follows that all evidence of the assumption 

 that the number of electrons in the lightest elements should 

 be the same as the atomic number, which is based on the- 

 estimate of mass scattering coefficients, cannot be thought 

 conclusive except perhaps for the elements H, Li, Be, B,. 

 and N. 



Glocker has pointed out that - in case of Al must be 



p 

 less than 0*2, because already at the wave-length X = 0*16 



the uncorrected mass absorption coefficient ( — J is 0*18, 

 and -, as we know, can only be a part of the whole 



absorption coefficient. Now, lor the lightest elements 

 (Li, Be, B, N) the value 0*2 must undoubtedly be nearly 



correct ; thus - cannot have the same value for all the 



P 

 lighter elements. As I have shown before (Table XII I.) , 

 the value 0*080 may well agree with the determinations of 



- by Barkla and White for AL 



r 



As was previously pointed out (p. 183), K -pyn n rapidly 

 increases with decreasing wave-length. This would, 

 according to what has been said above, mean that absorption 

 dependent on scattering is much greater in case of Pb 

 than of Cu, i. e. that in an atom of Pb there would be 

 present a considerably greater number of outer electrons 

 than in an atom of Cu. Yet whether the simple law that 

 defines the scattering power in case of the lighter elements 

 be in force in case of the heaviest must for the present 

 be left undecided. Presumably the electrons in case of 

 the heavier elements are not active individually but in 

 groups, as is the opinion of Barkla and White. By direct 



determinations of — , Barkla and Dunlop * have shown 



( £ )pb • ' (-)o a 



that ~~ — is about 2 or 4 times greater than -fi . 



(i) i\ 



V/o/Al V/Al 



* Barkla and Dunlop, Phil. Mag. xxxi. p. 222 (1916). 



