650 Prof. C. G. Barkla on the 



mass absorption coefficient I - j in carbon of a certain X-radia- 



tion of quite ordinary penetrating power, such as certainly 



obeys the laws of scattering, was shown by Barkla and Sadler* 



to be only about '41. It is difficult to reconcile this with 



s 

 Mr. Crowther's estimate of scattering, i. e. -=1'18. 



(2) Much more penetrating beams of homogeneous X- 

 radiation have since been found by the writer, and the mass- 

 absorption-coefficient for these in carbon has been found as 

 low as about *25, or about J- -of Mr. Crowther's scattering 

 coefficient. Thus either for these rays the scattering is much 

 less than for the more absorbable rays, or there is a con- 

 siderable discrepancy between these results. Neither theory 

 nor experiment indicates that for these very penetrating rays 

 there is any diminution in scattering ; both in fact indicate 

 that the scattered radiation carries away the same fraction 

 of the energy of primary radiations differing widely in 

 penetrating power. 



(3) In aluminium itself, the mass-absorption-coefficient 



( - ) of certain penetrating rays is much less than 1*18. The 



lowest value experimentally found is about *6 for a fluor- 

 escent X-radiation characteristic of cerium. 



(4) The simple laws of absorption found by Barkla and 

 Sadler point to the conclusion that the mass scattering 

 coefficient is of the order of magnitude of *2. The absorption 

 of Rontgen radiation in a substance A bears an approximately 

 constant, ratio to the absorption in a substance B, through 

 long ranges of penetrating power. The limits to this law 

 are that the radiation used must not extend in penetrating 

 power beyond that of any radiation characteristic of either 

 A or B, and must not be near one of these characteristic 

 radiations on its more penetrating side. This proportionality 

 is evidently true only when the total absorption is great 

 compared with the portion of it due to scattering, for as has 

 been pointed out, the scattering is independent of the pene- 

 trating power of the radiation as well as the particular light 

 element which is producing the scattering. There is thus 

 a constant term in the variable mass-absorption-coefficient 



— . If then we subtract a constant quantity - due to 

 P P 



scattering from this, we expect the law to. hold even for small 



Yalues of - , for there is no obvious reason whv the law should 



P . " 



be departed from just when scattering becomes an appreciable 



* Phil. Mag-. May 1909. pp. 739-760. 



