546 SCIENCE PROGRESS 



For example, the latest forms of yellow flame-arcs owe their 

 efficiency largely to the fact that the carbons are impregnated 

 with materials giving several vivid lines in the yellow-green. 

 These lines are superimposed upon the continuous spectrum 

 of incandescent carbon and naturally distort to some extent the 

 colour of objects illuminated by them. 



It is to be expected that line spectra judiciously used may 

 lead to efficiencies much in excess of those obtained by the use 

 of incandescent solid materials, which are thrown into a mere 

 jumble of vibrations, some of which are valuable for the purpose 

 in view but the great majority non-luminous. 



But when " luminescence " is set up by the electrical stimu- 

 lation of rarefied gases or metallic vapours in a vacuum tube, 

 free natural vibrations come into play and if these are such 

 as to produce the sensation of light the luminous efficiency is 

 correspondingly high. 



As explained above, most artificial illuminants which utilise 

 luminescence also depend for their light upon incandescence 

 to some extent and we therefore obtain a line-spectrum super- 

 imposed upon a continuous spectrum. Possibly this also takes 

 place to some extent in the case of the Moore vapour tube 

 and the quartz tube mercury vapour lamp, the spectra of which 

 exhibit a considerable number of lines. We may naturally 

 suppose that the spectra obtained from such vapours would 

 depend on the pressure within the tube and that the dominant 

 colour would be determined by the nature of the gas. Nitrogen 

 yields a pinkish light, neon a bright reddish-orange, whilst 

 carbon dioxide is said to afford a white light so closely re- 

 sembling daylight as to have proved very suitable for silk 

 mills and in other cases in which exact colour-matching is 

 essential. 



Perhaps the best-known example of a source of light de- 

 pending upon luminescence is furnished by the mercury vapour 

 lamp, which yields four groups of bright lines in the visible 

 spectrum, situated respectively in the yellow, green, blue and 

 violet. In addition there are also a considerable number of 

 lines in the ultra-violet. It has, in fact, been estimated that 

 whereas only a fraction of i per cent, of the total energy from 

 the electric incandescent lamp takes the form of ultra-violet 

 rays, as much as one-third of the energy produced in the 

 spectrum of luminescent mercury is in the invisible ultra-violet 



