FOURIER S RESEARCHES ON RADIANT HEAT. 413 



the fact, that in all the points of a space terminated by 

 any envelop maintained at a constant temperature, we 

 ought also to experience a constant temperature, and pre 

 cisely that of the envelop. Now Fourier has estab 

 lished, that if the calorific rays emitted were equally 

 intense in all directions, if the intensity did not vary pro 

 portionally to the sine of the angle of emission, the tem 

 perature of a body situated in the enclosure would depend 

 on the place which it would occupy there : that the tem 

 perature of boiling water or of melting iron, for example, 

 would exist in certain points of a hollow envelop of 

 glass ! In all the vast domain of the physical sciences, 

 we should be unable to find a more striking application 

 of the celebrated method of the reductio ad absurdum of 

 which the ancient mathematicians made use, in order to 

 demonstrate the abstract truths of geometry. 



I shall not quit this first part of the labours of Fourier 

 without adding, that he has not contented himself with 

 demonstrating with so much felicity the remarkable law 

 which connects the comparative intensities of the calorific 

 rays, emanating under all angles from heated bodies ; he 

 has sought, moreover, the physical cause of this law, and 

 he has found it in a circumstance which his predecessors 

 had entirely neglected. Let us suppose, says he, that 

 bodies emit heat not only from the molecules of their sur 

 faces, but also from the particles in the interior. Let us 

 suppose, moreover, that the heat of these latter particles 

 cannot arrive at the surface by traversing a certain thick 

 ness of matter without undergoing some degree of absorp 

 tion. Fourier has reduced these two hypotheses to cal 

 culation, and he has hence deduced mathematically the 

 experimental law of the sines. After having resisted so 

 radical a test, the two hypotjieses were found to be com- 



