MELLONI ON THE NOCTURNAL COOLING OF BODIES. 463 



ferent sorts of silver follow exactly the inverse ratio of the den- 

 sities. The differences which I have investigated were due solely 

 to the radiating power of the metal in its state of greatest acti- 

 vity, whilst the experiments of MM. de la Provostaye and De- 

 sains determine the emissive power of silver and of other metals 

 with regard to lamp-black ; whence probably has arisen the 

 slight historical error contained in the introduction of their 

 memoir. According to them the ratio hitherto admitted be- 

 tween the emissive power of metals and that of lamp-black would 

 result as well from the experiments of Leslie as from the labours 

 of Petit and Dulong and from my researches. It is very true that 

 Petit and Dulong have obtained results but little differing from 

 those of Leslie*, but no researches on this subject have been 

 published by me. The only questions which have appeared to 

 me sufficiently cleared up by experiment to merit the attention of 

 philosophers, are, first, the influence of inequalities of surface, 

 which we have just mentioned, and the influence of colour, 

 questions which have each received a negative solution. I next 

 examined the influence exercised by the thickness of the radia- 

 ting stratum of the heated body on the phaenomena of radiation, 

 the sole cause, in my opinion, of the enormous differences ob- 

 served between the emissive powers of different substances. As 

 to the numerical determination of the radiation of metals referred 

 to that of lamp-black, any one may easily convince himself 

 that no mention has been made of this subject in the different 

 memoirs which I have published on radiating heat. 



I will add, lastly, that experiments analogous to those of 

 MM. de la Provostaye and Desains had already caused me to 

 suspect the error announced by these skilful experimenters. But 

 the difference between the new and the old value was so great, 

 that I was tempted to attribute it to some faults of construction 

 in the thermo-multiplier which I had employed. At present the 

 agreement of the results obtained by such different methods as 

 the radiation from the cubes of Leslie on the thermo-multiplier 

 and the radiation of thermometers with metallic surfaces towards 

 a serene sky, appears to me to remove ail doubts. Let us hope 

 that these observations will be repeated and completed by ex- 

 perimenters, and that before long the very inaccurate values of 

 the emissive powers of gold and silver, of copper, tin and brass 



• Reckcrchen sur la Mesure des Temps. Paris, 1818, p. 75. 



