On the Mechanics of Luminosity. 377 



from the values found for the energies in absolute measure. 

 According to p. 161, L = E /&, where E denotes the energy 

 emitted in unit time (i is there used instead of E ), whilst 

 the condition of the body remains the same as at the begin- 

 ning. E has been already determined by the measurements 

 already detailed. We must further know the value of b, 

 which determines the velocity with which a body exposed to 

 a cause producing light comes to rest again. 



We evidently obtain the value b if we know the intensities 

 J t and J , to which the radiated energies are proportional at a 

 time and a time t, according to the equation 



J t =J ,-« or 6= log (J„/J ( )/(doge). 



For this calculation, of course, those phenomena are not to 

 be employed in which, although a gradually diminished emis- 

 sion of light occurs, the decrease of brightness does not depend 

 only on the diminution of the luminous motions, but the 

 radiated luminous energy is partly replaced by other processes. 



This takes place in a body glowing in consequence of an 

 increase of temperature, which gradually cools by radiation. 

 For although, by cooling, the molecules do at first lose a 

 portion of the energy of these vibratory motions, if the energy 

 of the luminous motion falls below the amount corresponding 

 to the temperature, a portion of the same is restored by the 

 impacts of the molecules at the expense of the motions of 

 translation. 



If, further, the body absorbs much light, and if the layer 

 used is so thick that all of its molecules cannot emit rays, then 

 a portion of the energy of the inner, hotter, not radiating 

 layers will be conducted to the outer cooler layers, which are 

 cooling by radiation. In these cases, therefore, not only the 

 energy contained in the form of luminous motions is lost, but 

 the total quantity of heat which was communicated to the 

 body by heating from the surrounding temperature to another 

 higher temperature. Measurements of the times of cooling 

 which, for example, gave 1*2 second for a glow-lamp to cool 

 from a white heat to darkness, and about 8 seconds for a pla- 

 tinum wire, 03 millim. thick, cannot be employed at once for 

 the determination of our magnitude b. 



Investigation of Balmain's Luminous Paint. 



35. Processes in which we can assume radiation in con- 

 sequence of the luminous motions present, without such 

 secondary conditions, are phenomena of phosphorescence. 1 

 have therefore investigated, in the first place, how far this 

 assumption in the case of the ordinary phosphorescent sub- 



