1903.] of the Spontaneous Radiation of Radium. 199 



though we are unable to trace any cause of shift, we feel justified in 

 shifting the new spectrum on the diagram so as to bring the lines 

 into agreement with those of nitrogen. The amount of shift at the 

 position of the strongest line is nine tenth-metres, which, on the 

 photographic plate, corresponds to the space of yi^th of an inch. 



The positions of the three bands are, according to Ames,* 3576'85, 

 3371-2 and 3158-9. 



Indications of other lines, besides those which can be seen in the 

 reproduction, can be faintly glimpsed on the negative. There seems 

 little doubt that with a longer photographic exposure a more complete 

 spectrum will be obtained. We have now secured some radium 

 bromide prepared by the Socittt Centrale de Produits Chimiques, and it is 

 our intention to take photographs of this salt, as well as photographs 

 of the German salt, with longer exposures. It may then be, that 

 indications of helium, and possibly of radium itself, may be forthcoming. 



Nearly the whole of the ultra-violet radiations appear to come from 

 nitrogen, and we think it best to refrain from any discussion at this 

 moment. Have we to do with occluded, or with atmospheric nitrogen 1 

 The remarkable fact should be pointed out that in radium we have a 

 body which at the ordinary temperature, sets up radiations which are 

 similar to those which have hitherto only been obtained in connection 

 with the electric discharge. 



Description of the Plate. At the top, is placed a scale of approximate 

 wave-lengths. Immediately below is a reproduction, enlarged two and 

 a-half times, of the spectrum obtained from the radium bromide with 

 an exposure of 72 hours. As has been already explained this has been 

 shifted to bring the lines into position with those of nitrogen photo- 

 graphed from a vacuum tube. The identity of the two spectra seems 

 complete. The third band is faint in the nitrogen spectrum on account 

 of the absorption of the glass of the tube. 



Below, is a spark spectrum of radium bromide from the SocidM 

 Centrale de Produits Chimiques. The H and K lines of calcium are 

 present, as well as faintly some of the stronger lines of barium. The 

 characteristic lines of radium at 3814-59 and 3649'7 come out strongly, 

 as well as the strong line recorded by Demar^ay at 4340*6. A strong 

 line about 2710 was placed by Berndtf at 2708'6. The strong line a 

 little beyond, about 2814, is due to radium. We do not recognise several 

 lines recorded by Exner and HaschekJ in this part of the spectrum.] 



* ' Phil. Mag.,' vol. 30, p. 57, 1890. See also Deslandres, ' Comptes Eendus,' 

 vol. 101, p. 1256 ; and Percival Lewis, ' A>tropli. Journ.,' vol. 12, p. 8. 

 f 'Phys. Zeitschr.,' vol. 2, No. 12 (19001901). 

 j ' Sitzb. Ak. Wiss. Wien,' vol. 110, July 4, 1901. 



