54Q SCIENCE PROGRESS 



perhaps be tested. It is well known that the quality of sunlight 

 is considerably modified during its passage through the earth's 

 atmosphere, which absorbs the rays of short wave-length to 

 a greater extent than those at the red end of the spectrum. 

 It might therefore be supposed (and this has been verified 

 at the summit of Mount Whitney in the Sierra Nevadas by 

 Prof. S. P. Langley) that the light in mountainous districts 

 would contain an abnormally strong blue element and that 

 the eyes of inhabitants of such regions would prove on examina- 

 tion to be specially sensitive to this part of the spectrum. In 

 the like manner we should expect to find that the vision of 

 habitual town-dwellers, accustomed to smoky and misty con- 

 ditions, would differ appreciably from that of country people 

 living under a clearer sky. 



At the present time it hardly seems probable that the 

 temperature of incandescent illuminants could be increased 

 sufficiently to bring their energy maximum anywhere near to 

 the visible region of the spectrum. There is, however, room 

 for progress in this direction, seeing that present limits are 

 set by other purely practical considerations : for example, it 

 seems at present very difficult to make trustworthy metallic- 

 filament lamps, having a reasonably long life, with a specific 

 consumption much below V2 watts per candle; yet the con- 

 sumption may easily be reduced to o'5 watts per candle or 

 less by overrunning the lamp and even maintained at this limit 

 during a short time. The higher efficiency is thus an obvious 

 possibility; the difficulty is that the filament cannot long 

 withstand such conditions. 



There is, however, another possible means of improving the 

 efficiency of illuminants of which good use has already been 

 made, namely, to choose materials departing from the black body 

 law which exhibit " selective radiation." Substances which are 

 highly polished, white, grey or coloured, may be expected to 

 exhibit this quality. What is needed for the purpose in view 

 is a material which absorbs and therefore emits visible light 

 as completely as possible but which acts as a highly reflecting 

 substance towards infra-red rays. 



It is supposed that the Nernst filament and the incandescent 

 mantle manifest selective radiation of this kind. It would take 

 us too far afield, however, if we were to enter into the many 

 interesting speculations that have been made as to the action 



