278 Hutchins and Pearson — Air Radiation. 



On clear, still days, excess temperatures of 50° to 60° of the 

 hot air stream were obtained, and from the deflections produced 

 as compared with those produced by a lampblacked surface at 

 known temperature, we got values of the radiation constant that 

 lay on both sides of the mean result of 1892. Great difficulty 

 was experienced in getting a steady flow of hot air, and the 

 behavior of the radio-micrometer was far from satisfactory. 

 The experiments were discontinued when it was found that 

 nothing new was to be learned by this method. We could at 

 least conclude that the difference between pure air and that 

 contained in an ordinary room with respect to radiating power 

 was inappreciable. 



The winter of 1902-3 was spent in improving the radio- 

 micrometer, and an instrument of remarkable sensitiveness and 

 accuracy was produced.* 



This season we have taken up the problem anew with much 

 improved apparatus and in very much improved surroundings. 

 The investigation was carried out in the constant temperature 

 room of the Searles Physical Laboratory, and the extreme range 

 of temperature during the weeks of experiment has been less 

 than 2°. 



Finding our knowledge of the absorption of air for its own 

 radiation in a very imperfect state, we turned our attention first 

 to that problem. 



Description of Apparatus. 

 The radio-micrometer w r as mounted upon a massive stone 

 table, and screened from external sources of radiation. In line 

 with the opening of the radio-micrometer was placed a trun- 

 cated cone of sheet tin, 45 cm long, having an opening l*5 cm in 

 diameter, corresponding to the opening of the instrument, and 

 enlarging to 5*5 cm at the other end. The cone is extended by a 

 cold-drawn seamless brass tube, polished within, 280 cm long, 

 and 5 cra internal diameter. Over the end of the brass tube is 

 slipped a tin tube 8 cm in diameter, held in place by wooden 

 rings and projecting 70 cm beyond the brass tube. These 70 cm 

 are thickly set with diaphragms, having 5 cm openings, and the 

 tube and diaphragms are carefully blackened. The legitimacy 

 of using reflecting tubes for passing along a radiation from a 

 distant* source has often been called in question. All doubt 

 should, however, be set at rest by recent experiments made 

 upon the reflecting power of metals bathed in air, for long 

 waves. Hagen and Rubens showf that all metals are practi- 

 cally perfect reflectors for radiations of great wave length, and 

 it is certain that any difference between the reflecting power 

 for air radiation, which is known to be of very great wave 

 length, and the radiation from a lampblacked surface at slight 



* This Journal, vol. xv, April, .1903. f Drude's Annalen, vol. ii, p. 873, 1903. 



