208 On the Distribution of Heat in the Visible Spectrum. 



ments of the solar spectrum, and with Tyndall's of that of the 

 electric light, by the same method. 



In both cases the experiments were made with rock-salt 

 prisms ; and these were assumed to have the same dispersive 

 power as the one used by Lamansky, and the curve plotted 

 accordingly. Two sets of measurements are given by Knob- 

 lauch ; and the mean of these was taken for the following cal- 

 culations. 



Knoblauch's measurements. 



Ordinates of 



heat-curve 



08). 



16-1 ... 



12-0 ... 



8-7 ... 



7-5 ... 



7-0 ... 



6-8 ... 



6-6 ... 



6-3 ... 



6-0 ... 



5-7 ... 



23 

 3-6- 



3-9 



4-1 ™ 

 4-6 _ E 



4-8 

 4-7 

 4-5' 

 4-6 

 4-0 



D. 



F. 



Tyndall's i 



measurements 



Ordinates of 



£ 



heat-curve 



(/3). 



« 



28-3 ... 



... 3-0 



21-0 ... 



... 3-0 



15-7 ... 



... 4-7— D. 



12-0 ... 



... 5-4 



8-2 ... 



... 5-7 ™ 

 ... 4-3~ E - 



6-5 ... 



5-0 ... 



... 3-5 



3-5 ... 



... 2-5 ™ 

 ... l-7- F * 



2-0 ... 



1-5 ... 



... 1-1 



•9 ... 



... -6 



The nature of the available data is such that the only defi- 

 nite conclusions which it is possible to draw from these calcu- 

 lations are, that the distribution of heat in the normal spectrum 

 differs greatly from that in the dispersion-spectrum, and that 

 in the dispersion-spectrum the great calorific intensity of the 

 red rays, and therefore in all probability of the invisible rays 

 beyond them, is due to the action of the prism in concentra- 

 ting these rays upon the face of the thermopile. The intensity 

 of the heat in the different portions of the normal spectrum, 

 except in the case of Lamansky's measurements with the flint- 

 glass prism, apparently varies but little through a considerable 

 space; and this affords some support to Dr. Draper's hypo- 

 thesis, that every colour ought to have the same heating 

 effect. 



After I had finished these calculations, I found that Gr. 

 Lundquist had investigated (Pogg. Ann. civ. p. 146), from 

 Lamansky's measurements, the distribution of heat in the 

 normal spectrum, and had shown that it differed greatly from 

 the distribution in the dispersion -spectrum — the maximum 

 intensity in the case of the flint-glass prism being near D, and 

 in the rock-salt prism near E. He also found from Tyndall's 



