SECT, xxv.] MELLONl's EXPERIMENTS. $45 



Melloni observed that the difference between the quantities 

 of caloric transmitted by the same plate of glass, exposed 

 successively to several sources of heat, diminished with the 

 thinness of the plate, and vanished altogether at a certain 

 limit ; and that a film of mica transmitted the same quantity 

 of caloric, whether it was exposed to incandescent platina or 

 to a mass of iron heated to 360. 



Coloured glasses transmit rays of light of certain de- 

 grees of refrangibility, and absorb those of other degrees. 

 For example, red glass absorbs the more refrangible rays, 

 and transmits the red, which are the least refrangible. On 

 the contrary, violet glass absorbs the least refrangible, and 

 transmits the violet, which are the most refrangible. Now 

 M. Melloni has found, that although the colouring matter 

 of glass diminishes its power of transmitting heat, yet red, 

 orange, yellow, blue, violet, and white glass transmit calo- 

 rific rays of all degrees of refrangibility. Whereas green 

 glass possesses the peculiar property of transmitting the 

 least refrangible calorific rays, and stopping those that are 

 most refrangible. It has therefore the same elective action 

 for heat that coloured glass has for light, and its action on 

 heat is analogous to that of red glass on light. Alum and 

 sulphate of lime are exactly opposed to green glass in their 

 action on heat, by transmitting the most refrangible rays 

 with the greatest facility. 



The heat which has already passed through green or 

 opaque black glass will not pass through alum, whilst that 

 which has been transmitted through glasses of other colours 

 traverses it readily. 



By reversing the experiment, and exposing different sub- 

 stances to caloric that had already passed through alum, 

 M. Melloni found that the heat emerging from alum is 

 almost totally intercepted by opaque substances, and is 

 abundantly transmitted by all such as are transparent and 

 colourless, and that it suffers no appreciable loss when the 

 thickness of the plate is varied within certain limits. The 



