340 RADIANT HEAT. 



must all be reduced by about .05. This will give the corrected series 

 of results as follows : 



Source of heat. Index of refraction 



for rock salt. 



521 



548 

 537 

 543 

 533 

 522 

 538 

 534 

 518 

 527 

 522 



Locatelli lamp 1 



Locatelli lamp, transmitted through alum 1 



Locatelli lamp, transmitted through glass 1 



Locatelli lamp, transmitted through opaque glass 1 



Locatelli lamp, transmitted through opaque mica 1 



Incandescent platina 1 



Incandescent platina transmitted by glass 1 



Incandescent platina transmitted by opaque mica 1 



Brass at 700° 1 



Brass at 700° transmitted by clear mica 1 



Mercury at 450° 1 



Mean luminous rays 1 • 552 



From the experiments described in this section the following gen- 

 eral conclusions are deduced: 



1. The mean quality, or that of the more abundant proportion of 

 the heat from different sources, varies within narrow limits of refran- 

 gibility. 



2. These limits are very narrow, indeed, where the direct heat of 

 any source is employed. 



3. All interposed media, (including those impermeable to light, ) so 

 far as tried, raise the index of refraction. 



4. All the refrangibilities are inferior to that of the mean luminous 

 rays. 



5. The limits of dispersion are open to further inquiry; but the 

 dispersion in the case of sources of low temperature appears to be 

 smaller than in that from luminous sources. 



Reflection of Heat : Melloni. 



A short paper, by M. Melloni, entitled "Note on the Reflection of 

 Heat," was read to the Royal Academy of Sciences, November 2, 

 1835, and published in the Ann. de Chim., ix, 402, of Avhich a trans- 

 lation appears in Taylor's Sci. Memoirs, Part III, p. 383. 



After referring to the experiments of Leslie, to show that the re- 

 flection of heat depends materially on the texture, polish, <fec, of the 

 reflecting surfaces, he proceeds to consider what takes place in dia- 

 thermanous substances, as in rock salt, where, there being no absorp- 

 tion, the difference of the heat transmitted gives the quantity reflected 

 at the first and second surfaces. With other media, as glass, rock 

 crystal, &c, very thin plates exercise no sensible absorption; hence 

 heat, after traversing a thick plate, being intercepted by a very thin 

 plate, the loss which this occasions is due solely to the two reflections. 

 These considerations afford the means of estimating the intensities of 

 reflected heat from different substances; and the author, in conclu- 



