60 Prof. Magnus on Thermal Radiation. 



only be employed at low temperatures, was difficult to fix, in 

 consequence of its being heated. I could find no better me- 

 thod of securing this fixity than to press platinum-black lightly 

 on a platinum plate which had previously been smeared uniformly 

 with a very thin layer of fat. On afterwards shaking the plate, 

 the black adheres to it uniformly. Plates so prepared radiate 

 about 25 per cent, more heat than when they are covered with 

 spongy platinum. 



Lampblack, when distributed in a similar manner over a pla- 

 tinum plate, has a radiating power quite similar to that of pla- 

 tinum-black. Whether the thermal colours of the two are the 

 same or different was not examined. 



The question whether the magnitude of the radiation is or is 

 not determined by the roughness of the surface is an exceedingly 

 important one in every theory which refers the phenomena of 

 heat to motion. Melloni and Knoblauch, in the memoirs already 

 cited, assert that the radiation depends, not on the form (Gestalt) 

 of the surface, but upon its density solely. None of the inge- 

 nious experiments upon which Melloni bases his assertion, is 

 more conclusive than his observation that rough surfaces do not 

 invariably radiate more than smooth ones ; for example, marble, 

 jet, ivory, quartz, gypsum, and some other substances do not, 

 according to this philosopher, radiate more heat when in a rough 

 than when in a smooth condition. I have found, too, that both 

 the white and black varieties of mica (Glimmer von Miask) as 

 well as several non-metallic substances, deport themselves in a 

 similar manner. Alum in a finely powdered condition fuses 

 at 100°, and its fused surface radiates almost the same as does 

 the rough powder. Powdered sugar scarcely radiates more 

 than fused sugar. 



On the other hand, Melloni admits that metals in the state of 

 chemical precipitates (e. g. when distributed in a finely divided 

 state over the surface of a Leslie's cube) possess very great 

 radiating powers. 



Filings (Feilspdhne) deport themselves in the same manner. 

 They, too, increase the radiation very considerably when distri- 

 buted over the rough surface of the same metal. In order to 

 meet every objection against this experiment, I spread the filings 

 upon so thin a platinum plate that their most prominent points 

 were not nearer to the thermo-electric pile than was the ante- 

 rior surface of the thicker rough platinum plate with which 

 the filings were compared. This plate, too, was cut from the 

 same piece of metal which furnished the filings. 



An experiment with aluminium gave like results. This metal 

 in the rough, as well as in the smooth condition, is a better 

 radiator of heat than either platinum or silver. 



