Mechanics of Luminosity . 399 



where E denotes the energy emitted by the unit of area 

 (1 square centim.) of platinum. In the region between C 

 and F, a quantity of energy, 0-016 x 4*7 gr. cal. per second, 

 is radiated from the platinum ; so that in this region 



E'=0-13x 0-016x4 7 = 0-0098 gr. cal. per sec. 

 In fluorescent light, such as would be radiated in all direc- 

 tions by a crystal of uranium nitrate whose surface is 1 square 

 centim., we find the energy 



F = 0'015E / =l-47 x 10" 4 gr. cal. per sec. 

 The thickness of the crystal investigated was 0*1 millim., its 

 specific gravity 2*5. The luminous mass, which has 1 square 

 centim. surface, is 



^=2-5x10- g, 



Therefore in 1 second there is radiated from 1 gr. a quantity 

 of energy 



Jo= 2-5 X X 10- 2 =5 ' 9 X 10 ~ 3 gr * calories ' 



We had found above for the quantity b 2 x 10 3 , so that the 

 store of luminous energy per gramme and the conditions 

 becomes 



L= ^ = 5 'l X ]°~ 3 =2-9 x 10- 6 gr. calories, 

 b 2 x 10 d fe 3 



if we take the initial intensity to be ¥=i , as measured in 

 energy. 



This quantity of energy of a purely luminescent motion 

 agrees, as to order of magnitude, with that found ( § 38) for 

 the sodium rendered luminous in a Bunsen flame, When 

 heated from the absolute zero to 0° C, 1 gr. uranium nitrate 

 (if we take its specific heat as 0'5) would receive a quantity 

 of heat equal to 137 gr. calories ; in comparison with this the 

 store of luminous energy disappears altogether. 



The further discussion of these and other numbers shows 

 that the temperature of luminescence of the uranium nitrate 

 examined is very high. The considerations, as well as a more 

 accurate determination of particular values, must be reserved 

 for a later communication. 



I desire here to express my best thanks to Dr. Ebert, 

 who has most kindly assisted me with these investigations, 

 both in the measurements and in the calculations. 



Erlangen, December 1888. 



