110 PROCEEDINGS OF THE AMERICAN ACADEMY 



I describe my apparatus here in principle, not in detail ; and I omit 

 many preliminary experiments, as well as some minute corrections 

 applied for small instrumental errors, giving my results in general 

 terms. One difficulty attending a simultaneous comparison was to 

 obtain a station looking into the "converter" at the time it was in- 

 clined and pouring, and yet necessarily outside the building in the 

 sun-light. To do this, I stood in a window (whence the sash had been 

 removed) of the west wall, sixty-one feet from the " converter " mouth. 

 A platform was erected here for my apparatus, part of which was 

 clamped to the wall itself ; but though this was the best point of 

 observation, the noise, the shower of sparks driven over the instru- 

 ments from within by the blast at each " pour," and the rain of wet 

 soot without which fell thick at times on apparatus and observer 

 from the combined steam and smoke of adjacent chimneys, made the 

 task of observation another thing from what it is in the quiet of a 

 physical cabinet. 



From this window-station, a porte-lumiere reflected the sun's rays, 

 so that traced through the dusty air the beam was seen to enter the 

 " converter " mouth, or fall on the stream which ran from it. Jn the 

 path of this beam was a cylinder, containing within a double enclosure 

 an Elliott thermopile of forty small elements, similar to that I had used 

 for some years on the sun, and surrounded by all the precautions 

 against air-currents and extraneous influences taught me by experi- 

 ence. The pile exposed both faces at once, one to the furnace, the 

 other to the reflected sunbeam ; and a Thompson reflecting galvanom- 

 eter read by an assistant, and placed at a considerable distance from 

 any moving iron, gave prompt evidence as to which face was hotter. 



The angular area, subtended at the pile by the fluid metal, was 

 always many times that subtended by the sun's disk, and there was no ' 

 lens or medium of any kind (except air) between the " converter " mouth 

 and the pile. Supposing, then, the metal to have only presented a 

 disk equal in angular diameter to that of the sun, if the needle re- 

 mained stationary, it is plain that each was sending an equal amount 

 of heat, and that any square foot of the solar surface was radiating 

 at least as much heat as a square foot of the metal ; for it is obvious 

 that the distances of the two sources have nothing to do with this 

 effect under the given conditions. 



The metal area, however, being many times that of the sun, the 

 latter still over-balanced the metal ; showing that the sun was actually 

 very much the hotter. Accordingly, there was interposed between 

 the porte-lumiere and the pile a telescope which diffused the sun-light 



