136 M. P. Desains on the Emission of ' 



middle of the tube that is to be exhausted, and at right angles 

 to it. 



At first this adapting tube is only pushed far enough down 

 to dip a little way into the quicksilver. The bottle is then in- 

 verted so as to fill with mercury the tube that is to be exhausted, 

 and this, if it be thought proper, is boiled out, the bottle being 

 afterwards slowly set upright again, while the adapting tube is 

 pushed further and further in, at last nearly to the bottom. 

 The nozzle of the cock inserted into the tubulus is now con- 

 nected with the air-pump by means of a caoutchouc tube, in the 

 manner already indicated. Scarcely half a stroke of the piston 

 will be needed to empty the tube, which can now, after shutting 

 the cock and removing the caoutchouc tube, be either melted off 

 or left connected with the bottle. 



'Although only a very low degree of rarefaction of the air in 

 the bottle is needed in order to empty the tube completely, it is 

 nevertheless advisable, especially when the tube is not to be 

 melted off immediately, to give both corks a good coat of seal- 

 ing-wax before pumping out. 



XVII. On the Emission of Luminous Radiation at a Red Heat. 

 By M. P. Desains*. 



MPROVOSTAYE and I, a dozen years ago, proved that 

 • at a red heat different substances may have very differ- 

 ent powers of emission for light. And we have also given some 

 .indications as to the relative values of light emitted at this red 

 heat by the same extent of surface of oxide of copper, platinum, 

 and gold. In our experiments oxide of copper had about ten 

 times the emissive power of gold. Platinum was between the 

 two, but nearer the second. 



In the researches which I now submit to the Academy, I pro- 

 posed to myself to extend to a larger number of bodies the study 

 of the luminous emission at a red heat ; the results I have ob- 

 tained may be summarized in the following manner : — 



At a red heat, and in a perpendicular direction, oxide of cop- 

 per, oxide of cobalt, green oxide of chromium, brown oxide of man- 

 ganese, and red oxide of iron have for light emissive powers which 

 are virtually equal. Under these circumstances a coating formed 

 of sulphate of lead and a little borax also emits luminous rays 

 very easily. It is a little less brilliant than the preceding sub- 

 stances, but more so than platinum. Platinum is more so than 

 gold, and this metal, lastly, more so than a layer of oxide of zinc 

 rendered adherent by borax. 

 * Translated by Dr. E. Atkinson, from the Comptes Rendus, July 3, 1865. 



