332 RADIANT HEAT. 



an important step towards simplifying the theory. In this instance 

 again the improvement of the thermometer is a primary requisite. 



I may here mention that I have recently had a more delicate ap- 

 paratus made, with which I have repeated my former experiments, 

 still with the same result; it consists of two thermometers mounted 

 together, as before described. They were contrived for me by Mr. 

 Cary, so as to have very large degrees for a small part of the scale a 

 little above ordinary temperature. 1° Fahrenheit occupies about half 

 an inch; but the bulbs are large, which is unfavorable to the rapid 

 communication of the effect. These experiments are of a very tedious 

 nature to repeat with precision, owing to the necessity of waiting 

 between each repetition for the thermometers to cool and become 

 stationary. 



But it should be observed that there is nothing in my results which 

 contradicts the idea that simple heat may have in a very slight degree 

 a power of transmissibility through glass; all I have assumed is, that 

 it is sufficiently distinguishable in this respect from the heating power 

 which accompanies the light, and which undergoes no diminution. 

 Connected with these points, again, is the question, whether if simple 

 heat can radiate through solid transparent media, it cannot also com- 

 mence radiating in them. It is commonly asserted that radiation can 

 only take place, or commence, in elastic media. This, then, is an in- 

 quiry which will lead into a wide field of research, and may be found 

 connected with the intimate nature of radiation. It will also be a 

 question whether, and how far, radiant heat passes through elastic 

 media without heating them, and what support this gives to Leslie's 

 theory of pulsations. The whole subject should be viewed in con- 

 nexion with the admirable remarks of Sir J. Herschel in his Discourse 

 on the Study of Natural Philosophy, p. 205. 



The radiation of heat in vacuo is another point on which further in- 

 quiry is much wanted. The greater capacity of air tor heat, as it is 

 more rarefied, would occasion a more rapid abstraction from the hot 

 body; and thus in an atmosphere of extreme rarity the cooling ought 

 to be extremely rapid, and this must be accurately estimated in 

 measuring the radiation. But it appears from the experiments of 

 Gay-Lussac, (see Edinb. Phil. Joum., vi, 302,) that when air is re- 

 duced to the most extreme degree of rarefaction possible a very con- 

 siderable compression makes so little difference in its actual density 

 that the giving out of heat which ought to take place from diminishing 

 its capacity is absolutely insensible. 



But even in this case it is very questionable whether so complete 

 an approach to a real vacuum is obtained as to warrant inferences 

 respecting the radiation of heat in an actual vacuum. 



In fact, we want a connected series of determinations to show the 

 order and increase of conducting powers, as connected both with the 

 radiation in and through different media, and the interception which 

 they offer to its passage. 



In solids it is -presumed no radiation can commence; it is disputed 

 whether it can continue even partially; but conduction goes on rapidly 



