Prof. J. Thomsen on the Mechanical Equivalent of Light. 247 



Hence the entire radiation of the candle, which burnt 8*2 grms. 

 Spermaceti in an hour, wan 5*75c . 36 o, 5 = 210c in each minute. 

 As the corresponding quantity of spermaceti disengages about 

 1400c in a minute, it follows from this observation that only 

 one-seventh of the entire heat disengaged leaves the flame as 

 radiant heat and light, while six-sevenths of the entire quantity 

 are carried away by the heated air. 



Experiments were then made with more powerful flames. A 

 moderator lamp, whose luminous intensity was 8*6, and a gas- 

 flame of 7 '7 luminous intensity, were placed at a distance of 2 

 metres, and the results referred to the spermaceti candle as uuit. 

 The results were as follows : — 



Nature of the flame. 



Luminous 

 intensity. 



Radiation for 



the unit of 



light per 



minute. 



Spermaceti candle 



1 



1-2 



7-7 

 8-6 



210 

 201 

 199 

 199 





Moderator lamp 



It follows from these results that the radiation of a flame is 

 'proportional to the intensity of light, and that for the unit of light, 

 8*2 grms. of spermaceti in an hour, it is about 200c in a minute. - 



Bat this is the action of the entire radiation ; in order to de- 

 termine that part which is due to the luminous rays, the thermal 

 rays had to be eliminated. Probably it is not possible entirely 

 to separate the thermal from the luminous rays ; but it may be 

 done approximately. It seemed best to effect the absorption of 

 the thermal rays by water; and by various experiments I con- 

 vinced myself that the invisible rays cannot traverse a layer of 

 water 0*2 metre in thickness, or at all events traverse it to so small 

 an extent that they exercise no appreciable influence on the result. 

 The flame of a Bunsen's lamp emits, as is well known, a very feeble 

 quantity of light ; its thermal radiation is nevertheless very great. 

 I observed, for example, that a Bunsen's lamp which, without 

 access of air, had a luminous intensity of 1*2 unit of light and 

 radiated 255 thermal units, when air was admitted and the 

 flame was non-luminous exhibited a radiation of 196c in a 

 minute. Of these non-luminous rays not a trace passed through 

 - 2 metre water; but as soon as the access of air ceased and the 

 flame became luminous, the needle was deflected, and indicated 

 that 4*3c were transmitted. Whereas the needle of the mul- 

 tiplier remained unchanged at 0° so long as the radiation of 

 the non-luminous flame took place through water, a deflection of 



