the Intensities of Lights of different Colours. 173 



the daylight appears a feeble blue decidedly tinged with violet. 

 The side of the ring facing the Welsbach lamp is coloured 

 greenish-orange. This greenish orange was matched in hue, 

 but not in luminosity, by a rotating disk having 41*5 parts of 

 red-lead, 41*5 of chrome-yellow, and 17 of emerald-green, 

 viewed illuminated by daylight. This appears to be the hue 

 of the Welsbach light when compared with daylight. 



Experiments in which are compared the Hues of the Light of 

 an Incandescent Welsbach Lamp and of Candle-light.— -When 

 "a white-light incandescent Welsbach lamp" is placed on one 

 side of the large screen (fig. 1) and a candle or petroleum- 

 flame on the other side, the cardboard ring of the screen 

 facing the Welsbach light appears of a light bluish-green ; 

 the side of the ring facing the candle or petroleum-flame 

 appears of a light carmine. 



Matching these hues on the rotator with split disks gave 

 the following results when the two sides of the screen were 

 equally illuminated. The hue of the ring on the side facing 

 the Welsbach lamp was matched by 8 parts of emerald-green, 

 10 of cobalt-blue, 25 of ivory-black, and 57 parts of white 

 Bristol board, as seen illuminated by the Welsbach lamp, 

 The hue on the side of the ring facing the candle or petro- 

 leum-flame was matched by 12 of carmine, 17 of ivory-black, 

 and 71 of white cardboard, as seen illuminated by the petro- 

 leum-flame. 



Experiments with the Rotating Photometer-disk placed 

 betiveen the Welsbach Lamp and the Petroleum-flame. — -On 

 rotating the photometer-disk, without any central disks, 

 between these lights, the same hues as those already de- 

 scribed were seen on its sides, only those hues w T ere much 

 less saturated. 



The white disks placed on the sides of the rotating photo- 

 meter made the hues less saturated. 



With a white disk on the petroleum -flame side and a black 

 disk on the Welsbach-lamp side, the differences between the 

 hues of the sides of the photometer were more pronounced. 



With a disk of cobalt-blue, emerald-green, and white on the 

 Welsbach-light side and a white disk on the petroleum-flame 

 side, the side of photometer facing the Welsbach lamp ap- 

 peared white and the side facing the petroleum-flame was less 

 carmine, the differences in hue being less than in any of the 

 previous experiments. 



By trial I found that w r hen a disk of 33 parts of emerald- 

 green, 42 of cobalt-blue, and 25 of white was placed on the 

 Welsbach-lamp side of the photometer-disk, with a ring of 

 the same hue on the periphery, and a disk of 75 parts of 



