342 Dr. H. Hudson on the Theory of Exchanges. 



vibration ; and hence the idea that the temperature of the earth's 

 surface can fall (by radiation) below that of the atmosphere im- 

 mediately resting on it (as is assumed in the received theory of 

 dew-formation) I hold to be utterly untenable. Again, if we 

 imagine such a body, at any temperature, to exist alone in bound- 

 less space, it appears to follow (on Prevost's theory), whatever be 

 the temperature of the medium, that the vibrations of the mole- 

 cules of the body must ultimately cease altogether in consequence 

 of its receiving no return for its own wasted radiation. 



Experimental evidence. 



Let A B represent a metallic mirror, in whose focus (C) one 

 of the balls of a differential thermometer is placed, E a cubic 

 canister (two of its adjacent sides being bright metal, the other 

 two varnished), F representing a large tin screen. Suppose E to 

 be filled with ice-water, and its metallic surface facing the mirror. 

 The focal ball will he chilled, by an apparent "radiation of cold" 

 from the canister, which is reflected by the mirror to its focus. 

 The admirers of Prevost say that E (being colder than A B) can 

 no longer radiate so much heat to it as it received (and reflected 

 to C) whenE was of its own temperature, and therefore the tem- 

 perature of C is reduced. Well — let the varnished surface of E 

 be turned towards the mirror, and we find as a result that C is 

 more chilled than it had been by the metallic surface ; thus a 

 more powerful " radiation of heat," which ought at least to di- 

 minish the previous chill, causes the reflection of more cold to 

 the focal ball. I have always considered the above simple expe- 

 riment of Leslie's to be fatal to Prevost's theory of exchanges ; 

 for the radiation (from either A B, or C) is assumed to be wholly 

 independent of the radiation of E, and consequently we have 

 only to consider the effect of the latter as a radiator. 



There are many variations of Leslie's experiment, all leading 

 to the same conclusion. Thus, having gilt one of the balls of 

 his differential thermometer, he denominated it a " pyroscope." 

 Now, in repeating the above experiment with this instrument, the 

 results correspond with those already related, whether the gilt 

 or the glass ball be placed in the focus ; in every case the var- 



