Mechanics of Luminosity. 379 



and the luminous colour appeared equally bright. The time 

 at which this adjustment was completed was read off on a 

 chronoscope. Then adjustments were made at short intervals 

 at the commencement of the experiments, when the bright- 

 ness changed rapidly ; later, when the changes in brightness 

 occurred less rapidly, at longer intervals. In order to be able 

 to determine the brightness of the phosphorescence as rapidly as 

 possible in the first few seconds after screening off the exciting 

 light, the proper position of the Nicol prism of the photometer 

 was determined approximately by preliminary experiments, 

 and in the final measurement only the last adjustments were 

 quickly made. The intensities i are proportional to sin 2 «, in 

 consequence of the choice of zero-point on the photometer 

 (the comparison-stars are dark for a = 0); we put i = 10 s sin 2 a. 

 After the tablet had been illuminated for one second the fol- 

 lowing angular measurements were obtained at the times z on 

 the rotating graduated arc, carrying the Nicol prism of the 

 Zollner's photometer : — 



Series I. 



t 4". 20". 40". 1'3". 1'30". 3'. 4'. 



«° 9-6 50 37 3-2 2-5 1-8 1-8 



i x 278 7-6 4-2 3-1 1-9 1-0 1-0 



Then after three minutes' illumination : — 



Series IT. 



t ... 3". 15". 25". 40". 55". 1' 10". 1' 40". 2'. 3'. 4'. 5'. 

 a ° ... 248 17'1 122 82 73 64 54 47 41 36 3-6 

 i 2 ... 17'5 865 44-7 20-5 16-1 124 8-9 6-7 54 3-9 3-9 



These numbers, as well as numerous others, and a graphical 

 representation of the intensity as a function of the time, show 

 that the falling off from a definite brightness of the tablet 

 is dependent upon the duration of the previous illumination. 

 If this was only very short (e. g. 1 second), the falling off is 

 much more rapid than when the illumination has lasted longer 

 or has been very intense. In Series I., for instance, with short 

 illumination, the brightness sinks from 27*8 to 7*6 in 16 

 seconds; whilst in Series II., with long illumination, the 

 brightness sinks from 20*5 to 8' 9 in 60 seconds, &c. i 



In a substance incapable of undergoing change that is 

 heated to luminosity, e. g. a glowing platinum wire, the 

 change in brightness, observed from any brightness J, must 

 be independent of the maximum brightness, which, for ex- 

 ample, the body had before it, by cooling, reached the 

 brightness J. The phenomena with the luminous paint thus 



