on the terrestrial Rays that occasion Heat. 443 



with respect tp GQ ; and, as the part GQ is already known to 

 follow the law of refraction we have mentioned, it is equally 

 evident, that the additional heat of AG must follow the same 

 law. We do not enter into the dispersive power of different 

 mediums with respect to heat, since that would lead us farther 

 than the present state of our investigation could authorise us to 

 go ; the following experiment however will shew that, as with 

 light so with heat, such dispersive power must be admitted. 



22d Experiment. Correction of the different Refrangibility of 

 Heat, by contrary Refraction in different Mediums. 



I took three prisms ; one of crown glass, having an angle of 

 25 degrees ; another of flint glass, with an angle of 24 ; and a 

 third of crown glass, with an angle of 10 degrees. These being 

 put together, as they are placed when experiments of achromatic 

 refractions are to be made, I found that they gave a spectrum 

 nearly without colour. The composition seemed to he rather a 

 little over adjusted ; there being a very faint tinge of red on 

 the most refracted side, and of violet on the least refracted 

 margin. I examined both extremes by two thermometers ; 

 keeping, No. 3 as a standard, while No. 2 was applied for the 

 discovery of invisible rays ; but I found no heat on either side. 

 After this, I placed No. 2 in the middle of the colourless illu- 

 mination ; and in a little time it rose two degrees, while No. 3 

 still remained unaltered at some small distance from the spec- 

 trum. This quantity was full as much as I could expect, con- 

 sidering the heat that must have been intercepted by three 

 prisms. Thus then it appears, that the different refrangibility 

 of heat, as well as that of light,' admits of prismatic correction. 

 And we may add, that this experiment also tends to the estab- 



