of Heat in the Visible Spectrum. 



207 



Considering the nature of the data, and especially the small 

 scale of MM. Fizeau and Foucault's diagram (the portion re- 

 presenting the visible spectrum being only about 4 inches 

 long), the measured and calculated numbers agree fairly well 

 together. 



The same process was applied- to the curves given by La- 

 mansky (-Pogg. Ann. cxlvi. p. 200) for the distribution of heat 

 in the solar spectrum, with flint-glass and rock-salt prisms. 



Flint-glass prism 



Ordinates of the 

 heat-curve § 



03). 



Rock-salt prism. 



22-8 



21-0 



19-0 



17-5 



16-0 



14-5 



12-7 



11-2 



10-3 



9-8 



9-3 



8-5 



7-6 



6-8 



6-0 



5-0 



— D. 



8-4 



9-5 

 11-8 v 



20-9_ 6 



20-7 



15*8 



14-0 



14'7_ F 



14-0 ' 



13-2 



12-1 



10-8 



8-5 



7-5 



7>1 -G. 



Ordinates of the 



heat-curve 



O). 



18*2 



16-4 



14-3 



13-0 



11-5 



10-3 



9-3 



8-2 



7-0 



6-2 







6-0 

 6-0 

 7-9 

 9-2 

 8-2 

 7-3 

 7-1 

 5-8 

 5-0 

 4-4 



— D. 



— E. 



-F. 



— G. 



Assuming the curves to be the dispersion-curves for the 

 prisms, the ordinates were measured and calculated as in the 

 former case. 



D 

 E 



b 

 F 



G 



Flint-glass prism. 

 Measured. Calculated. 



23-6 

 17-8 

 15-4 

 10-2 

 4-0 



16-0 

 14-6 

 10-8 



Rock-salt prism. 

 Measured. Calculated. 



D 

 E 

 F 



G 



18-3 



13-7 



9-8 



6-0 



13-4 

 10-3 



In none of the other measurements that have been made of 

 the heat- spectra, as far as I am aware, are the positions of the 

 solar lines stated ; and therefore part of the data for eliminating 

 the action of the unequal dispersion of the prisms is wanting. 



An attempt was made to deal with Knoblauch's measure- 



