24 



PROCEEDINGS OF THE AMERICAN ACADEMY. 



TABLE II. 



% of energy 

 cut off by euphos 



2.6 

 4.5 



Source 

 100 Watt G. E. M. 

 Acetylene Flame 



Nernst (with globe) 5.2 



Tungsten (100 wt.) 6.0 



Graetzin Gas Lamp 8 .4 



Sunlight 21 .0 



Magnetite Arc (glass globe) 22.5 



Magnetite Arc (quartz window ) 28 .0 



Carbon Arc (quartz window) 30.0 



Mercury Arc (glass) 41.7 

 New Quartz Lamp (with Alba globe) 42.5 



Old Quartz Lamp (no globe) 55.7 



New Quartz Lamp (no globe) 65 . 1 



Candle power 



{horizontal) 



39.25 



27.35 



68.0 



79.4 



75.0 

 272 . (equivalent) 

 700.0 

 760.0 

 720.0 

 100. 

 820. 

 415. 

 348. 



It will be noted that the smallest percentage of ultra violet is shown 

 again by the G. E. M. lamp, with the acetylene flame standing second. 

 The Welsbach mantle of the Graetzin lamp runs materially higher 

 than any of the electric incandescent lamps in spite of the fact that 

 this lamp was tested with its globe on. Next higher than the Graetzin 

 lamp, and approximating the arc lamps, comes sunlight, standing 

 between the incandescent sources which give a continuous spectrum 

 and the arcs of various sorts which give highly selective radiation. 

 At the other end of the list is the quartz lamp worked intensively 

 without its globe. These ratings of the various illuminants are 

 instructive as showing the distribution of the energy as between 

 ultra violet and the remainder of the spectrum, but they are not 

 significant as regards the extremely practical matter of illumination. 

 If the ultra violet component of artificial light involves any risk of 

 injury to the eyes the one important thing to find out in comparing 

 sources of light is how much ultra violet they deliver for a given 

 illumination. In other words if one desires to light a room, say to 

 an intensity of five foot candles, with what illuminant can he obtain 

 this intensity while receiving the minimum amount of ultra violet 

 radiation? It is not of the slightest practical consequence from the 

 standpoint of good and safe illumination whether a given light source 

 produces much or little ultra violet per watt, provided it produces an 

 insignificant amount per foot candle, hence the luminous efficiency^ 



