On the Refraction of Light, 359 



sible coloric is indicated, and this, I have observed, not only in dif- 

 ferently colorific rays, but in different rays of the same color : thus, 

 by transmitting a beam of red light through colored glass, and causing 

 it to be refracted as in the foregoing experiment, it will be found, by 

 a delicate thermometer, that the heat is greater at the base than at 

 the summit of the image. The development of caloric^ is, there- 

 fore, directly as the density of the colorific rays. Now when a beam 

 of white light is incident on a dense diaphanous medium of unequal 

 thickness, such as the prism, its red rays, as they possess the greatest 

 density, and hence penetrate glass with greater difficulty than the 

 rarer rays, are trajected through the thinnest portion of the prism, 

 which they permeate with most facility. Having passed through 

 less of the glass than their concomitant colors, they will consequently 

 be less refracted, and occupy the base of the spectrum. The orange 

 rays, being next in the order of density, pass through the next thin- 

 nest portion of the prism, and having permeated a part of the glass, 

 thicker than the red, and thinner than the yellow rays, they undergo 



* When we introduce a pencil of red rays into a dark chamber through a plate 

 of colored glass, the thermometer being placed in them, indicates a higher tempera- 

 ture than in blue or violet light, admitted through colored glasses ; but it would be 

 preposterous to conclude that, because more heat is found in red than in blue light, 

 this difference is owing to the circumstance of the calorific rays being less refracted 

 by red than by blue glass, or that heat consists of differently refrangible rays. A 

 more gratuitous conclusion cannot easily be imagined ; for what proportion exists 

 between copiousness of heat and refrangibility ? were the rays of caloric differently 

 refrangible, why recur to glass prisms to prove it? or what analogy is there be- 

 tween bodies that transmit light, and those which transmit or conduct caloric ? In- 

 deed these properties generally occur in bodies in inverse proportion with regard to 

 each other. If the rays of caloric were differently refrangible, metallic prisms, 

 particularly silver ones, would answer the purpose of measuring the degrees of 

 their refrangibility much better than glass ones, since they are much more perfect 

 conductors of coloric. t Far from having any proof that the rays of heat are dif- 

 ferently refrangible, we have not a single fact to countenance the opinion that, 

 per se, they undergo refraction at all. 



For when a beam of heat strikes on a piece of silver, in the shape of a double con- 

 vex lens, the thermometer does not indicate any focus or concentration of calorific 

 rays. I may here mention an experiment which cost me much trouble. It was 

 instituted with a view of ascertaining, whether the rays of heat, after passing 

 through a conducting medium of metal, shaped like a burning glass, would converge 

 to a focal point. A circular piece of glass was cut with a diamond, out of the mid- 

 dle of a plane mirror, and in the hole was fixed a double convex lens of silver which 

 exactly fitted it. The mirror was mounted perpendicularly on a supporter, and an- 



t Does the refrangibility depend on the conducting power ? heat is jiardly con- 

 ducted by water at all, solar heat instantly radiates through that fluid. 



