j3 and 7 Rays from Radioactive Substances. 455 



product radium E, which, as Gray * has shown, emits relatively 

 an exceedingly weak 7 radiation. It may prove significant 

 that only those products which emit well-defined groups of 

 /3 rays emit also a strong 7 radiation ; for, as far as obser- 

 vation has gone, the /? rays from uranium X and radium E 

 give a continuous spectrum in a magnetic field. 



In order to account for the emission of groups of homo- 

 geneous rays from a single product, it is necessary to suppose 

 either that the atom breaks up in a number of distinct ways, 

 each of which is characterized by the emission of /3 particles 

 of definite velocity, or that the {3 rays are altered in velocity 

 in some definite way during their escape from the disinte- 

 grating atom. On the first hypothesis, it might be anticipated 

 that the different modes of transformation of a /3 ray product 

 would give rise to a series o£ new products, but only one is 

 observed. In addition, the energy emitted during the 

 transformation from one type of matter into another would 

 vary widely for different atoms of the same substance, and 

 this seems improbable. On the second hypothesis, it is 

 supposed that the disintegration of each atom takes place in 

 exactly the same way with the emission of the same amount 

 of energy, but that the energy of the /3 particle may bo 

 decreased by definite but different amounts due to trans- 

 formations of its energy in its passage through the atomic 

 system from which it originates. Since it is known that 

 /3 rays in escaping from an atom give rise to 7 rays, it is 

 natural to suppose that the loss of energy of the ft particle 

 in escaping from the atomic system is connected in some 

 way with the excitation of 7 rays. 



The work of Barkla and others on the X rays have 

 brought out clearly that under suitable conditions of exci- 

 tation, each element emits one or more definite types of 

 X radiation which are characteristic of the element. Barkla f 

 has determined the coefficient of absorption //, in aluminium 

 of the characteristic X rays for elements up to atomic 

 weight 140. The value of fi/D for aluminium, where D is 

 the density, decreases rapidly with the atomic weight and 

 varies between 435 for calcium of atomic weight 40*1 to 

 06 for cerium of atomic weight 140*25. Plotting the 

 logarithms of the values of /m/D against the logarithm of the 

 corresponding atomic weights, the points for the heavier 

 elements are found to lie nearly on a straight line. If this 



* Gray, Proc. Roy. Soc. A. lxxxv. p. 131 (1911) ; lxxxvi. p. 513 

 (1912). * 



t Barkla, Phil. Mag. xxii. p. 396 (1911). 



2H2 



