462 MELLONI ON THE NOCTURNAL COOLING OF BODIES. 



thousandths, which we may conclude from the nature of the in- 

 struments employed and from the great regularity of the different 

 series of observations. Hence laminated silver radiates about 

 four times less than experimenters have hitherto supposed. 



This result approaches very near to that obtained lately by 

 MM. de la Provostaye and Desains with regard to the emissive 

 power of this same metal by the aid of the thermo-multiplier 

 and Leslie's cube. In fact, if we continue to denote by 100 the 

 radiating power of lamp-black, silver chemically precipitated on 

 copper will, according to them, have its radiating power ex- 

 pressed by5'37 in its natural state, and by 2*10 when polished; 

 the emissive power of ordinary silver would be 2"94 when it first 

 comes out of the flattening apparatus {laminoir), and 2"38 after 

 having been burnished. 



As early as the year 1 838, I had deduced from some observa- 

 tions, that the difference of emissive power (as determined by 

 the well-known experiment of Leslie) of a rough surface and a 

 polished one of the same metal did not arise, as was then generally 

 supposed, from the greater or less degree of polish of the surface, 

 but in reality from the difference of density produced, in metallic 

 substances, by the scratches on the metal made in order to change 

 the smooth surface into a rough one, which scratches, in the or- 

 dinary cases of laminated metals, laid bare the interior part more 

 tender and radiating than the surface ; so that these changes of 

 density sufficed to explain the phasnomenon hitherto obsei'ved, 

 even in the case where the metal is not oxidable. This propo- 

 sition appeared to me to be incontestably demonstrated by the 

 two following facts : — 



1st. Silver melted and cooled slowly in the moulds, polished 

 •with oil and chai'coal, and afterwards scratched with the point of 

 a diamond so as to compress and condense the bottom of the 

 furrows, diminishes instead of increasing its radiating power in 

 passing from the state of polish to roughness. 



2nd. This same kind of polished silver loses its emissive power 

 as well, when smartly hammered on an anvil or passed through 

 the flattening apparatus*. 



It is hence easy to perceive that the same principle is in- 

 volved in the experiments of the two French philosophers ; for 

 silver chemically precipitated in copper being much less hard 

 and dense than polished silver, the emissive powers of these dif- 



* Bibliolheque Universelle de Genece, ann^e 183S. 



