IN ITS RELATION TO THE EYE. 449 



namely, the brightnesses of the reflectors, are in rooms of moderate 

 height dropped well into the field of view. It is obvious also that 

 the effect will depend on the number and size of the bright surfaces 

 in the field of view as well as on the angle of presentation to the 

 eye. For rooms of the size of the one in which we worked, an ap- 

 proximation of a maximum brightness may be gotten from the fol- 

 lowing data based on the testing of 52 lighting situations. For the 

 indirect installations the eye fell off 8.6 per cent, in power to meet 

 the standard imposed by the test as the result of 3 hours of con- 

 tinuous reading with the maximum brightness in the field of view 

 of 0.138 cp. per sq. in. For the direct installation the loss was 6.6 

 per cent, for a brightness of 0.0129 cp. per sq. in. ; 8 per cent, for a 

 brightness of 0.1815 cp. per sq. in.; and 32.9 per cent, for a bright- 

 ness of 0.66 cp. per sq. in. For the semi-indirect installations the 

 loss was 15 per cent, from a brightness of 0.264 cp. per sq. in.; 48 

 per cent, for a brightness of 0.361 cp. per sq. in. ; and 60 per cent, 

 for a brightness of 0.614 cp. per sq. in. We would not feel inclined 

 to recommend a maximum brightness greater than 0.15-0.2 cp. per 

 sq. in. with the grouping of distribution factors ordinarily found in 

 the lighting of rooms. In contrast with this, the brightness of the 

 gas flame and oil lamp is from 3-8 cp. per sq. in. ; the Welsbach 

 mantle from 20-50 cp. per sq. in. ; the carbon filament from 375- 

 480; the filament of the vacuum tungsten lamp from 875-1,000; the 

 filament of the gas-filled tungsten lamp 10,271-16,433; and the open 

 arc lamp from 10,000-50,000. 



6. A marked characteristic of the effects produced by the dense 

 and completely opaque direct reflectors was the low illumination of 

 the ceiling and upper part of the room, and the high and in some 

 cases almost glaring illumination of the floor and objects in the 

 working plane. So far as the effects on the eye of the kind regis- 

 tered by our tests are concerned, however, these irregularities of 

 illumination and of low surface brightness extraneous to the lamp 

 and reflector seem to be of comparatively little consequence, so 

 long as the higher brilliancies of lamp and reflector are themselves 

 properly taken care of. With the direct reflectors, translucent and 

 opaque, we have had quite wide variations in the distribution of 

 illumination ranging from the well-illuminated ceilings and the com- 



