of the /3 Rays excited in Metals by Soft X Bays. 279 



the tube by way o£ the thin aluminium window W. The 

 beam, after being limited by a narrow slit S, which is closed 

 airtight with thin mica, enters an evacuated chamber and 

 impinges on a thin sheet R of the metal under examination 

 (which will, for brevity, be referred to as the " radiator "). 

 The box is placed between the pole-pieces of a large electro- 

 magnet, and the excited /3 rays, after leaving the radiator 

 and passing through a 25 mm. slit, are bent round on to the 

 photographic plate P in the manner indicated in the diagram. 

 This arrangement is essentially similar to that previously 

 employed for the analysis of the primary j3 rays of radium B 

 and radium C *, the radioactive source being replaced by the 

 radiator. The principal advantage of this arrangement — that 

 of concentrating a considerable fraction of the toial radiation 

 into a fine line on the photographic plate — has been fully 

 discussed in the paper mentioned above : in spite of this, 

 however, and of the fact that the X-ray bulb was run at the 

 highest power of which it was capable, it was found that a 

 net exposure of twelve hours to the rays was required to 

 obtain an appreciable photographic effect. As under the 

 most favourable conditions an exposure of twelve hours to 

 X rays involves a continuous experiment of at least 18 

 hours' duration, and as the X-ray bulb has only a limited life, 

 this inefficiency of the radiator seriously increases the ex- 

 perimental difficulties. We intend to continue the experi- 

 ments with a water-cooled rhodium anticathode, and as 

 rhodium concentrates a large proportion of the energy of its 

 X radiation into a predominant line, this may appreciablv 

 shorten the exposures required. We are at present using 

 the j3 rays which are emitted from the radiator in a direction 

 approximately at right angles to that of the incident X rays. 

 This latter direction makes an angle of about 20° with the 

 normal to the radiator, and unless the number of /3 rays 

 emitted by a target struck by soft X rays varies greatly with 

 the direction, little improvement in experimental conditions 

 can be expected from an alteration of these angles. 



The radiators so far examined are iron and lead, and we 

 have obtained definite evidence of well-defined maxima in 

 the /3-ray spectra of these elements ; from the imperfect 

 homogeneity of the X radiation employed, these are probably 

 superposed on a heterogeneous /3 radiation, but in view of 

 the proof given recently by Chadwick t of the efficiency 

 of: the photographic plate as a detector of minute variations 

 in /3-ray intensity, our photographs cannot be expected to 

 offer any evidence on this point. 



* Rutherford and Robinson, Phil. Map:, xxvi. p. 717 (1913\ 



t Chadwick, Verh. d. Deuischcn Fkt/s. Gesell. xvi. p. 383 (1914). 



