728 Dr. J. A. Crowther on "J" Radiation. 



approximately the same wave-length but of a considerable 

 number of lines comparatively widely spaced in the spectrum. 

 The " J ,} radiations from elements of low atomic number 

 are very weak compared with the K radiations. In the case 

 of copper for example the intensity of the hard fluorescent 

 radiation was only about one-thirtieth of that of the cha- 

 racteristic K radiation from the radiator. This would indicate 

 that these hard fluorescent radiations are not easy to excite. 

 To excite any given line of the series, therefore, it is probably 

 necessary that the wave-length of the primary radiation 

 should not be much shorter than that of the line to be excited. 

 In other words, a given primary radiation would only excite, 

 to any appreciable degree, the particular Tine whose fre- 

 quency was nearest its own. This would explain why the 

 secondary radiation from elements of low atomic weight is 

 never very different in quality from that of the incident 

 radiation. It would also explain the difficulty of disentang- 

 ling in a satisfactory manner these hard fluorescent radiations, 

 as to the reality of which there seems to be little doubt among 

 those who have worked with the secondary X radiation from 

 elements of low atomic number. 



On this hypothesis the hard secondary radiations obtained 

 from aluminium, copper, and wax correspond to different 

 lines in the " J " spectrum of the elements. In the case of 

 the two former the lines fall well within the limits of the 

 series whose higher members have been tabulated by 

 Rutherford and Robinson. The radiation from the wax 

 radiator (which corresponds to the carbon absorption band 

 of Barkla) would correspond to a 7 radiation of rather higher 

 frequency. * 



The absorption bands discovered by Barkla for aluminium 

 and copper unfortunately represent frequencies rather greater 

 than that of the primary radiation used in the present experi- 

 ments, and the corresponding fluorescent radiations would 

 therefore not have been excited by the primary rays employed. 

 The curves given by Barkla and White in the paper referred 

 to show only one definite absorption band for each substance, 

 whereas on the suggestions we have advanced a con- 

 siderable number might have been expected. They call 

 attention, however, to certain deviations and irregularities 

 in their data which they state to be certainly real. A further 

 analysis of the curves or a still more accurate series of 

 experiments might easily resolve these into absorption bands 

 corresponding to other lines in the spectrum of the " J " 

 series. 



