292 



INFRA-RED EMISSION SPECTRA. 



energy supplied is spent in heating, and, for the same current work, the 

 total radiation increases with increasing pressure. 



After noticing the present results it will become quite apparent 

 that Angstrom's predictions are in remarkably close agreement with 

 observed facts. On the other hand, in his review of the recent work 

 on electrical gas spectra, the predictions of Berndf are not so close in 

 agreement with the present work. For example, in speaking of Ang- 

 strom's observations, which show that total radiation increases while 

 the visible radiation decreases with increased pressure of the gas, he 

 remarks that this would show that the spectral distribution of the 

 energy must shift with change in pressure, which has been observed. 

 But, to say that the distribution of intensity changes in such a manner 

 that, with decreasing pressure, the center of gravity of the distribution 

 of the spectral energy shifts toward the shorter wave-lengths, does not 

 seem admissible, since slight traces of CO, may have been present. 



In his efficiency investigation of vacuum-tube radiation Drew' ex- 

 amined the emission spectrum of air, using for the purpose the writer's 

 radiometer, which at that time was not very sensitive. He found 

 several weak lines in the region of i /^, while the greater part of the 

 energy is concentrated in a strong band at 4.75 /x. The other maxima 

 worthy of notice are those at 0.66, 0.74, 0.89, and 1.03 )w. For air the 

 0.89 /A band is more intense than the one at 1.03 fx, while in the present 

 work on nitrogen the reverse has been found. 



The infra-red emission spectrum of the mercury arc, in the form of 

 the Arons lamp, has also been explored.' Three strong lines were found 

 just beyond the red at 0.97, 1.045 and 1.285 /*> beyond which no lines 

 were found, except in the region of 4.2 to 5.8 /x where the radiation from 

 the hot cell interferes with the work. The fact that no line was observed 

 at 4.75 IX is of interest in connection with the present work. 



Visible Spectrum. 

 Several investigations in the visible spectrum are of interest in the 

 present work, and they will now be mentioned. 



Certain prominent lines of hydrogen and nitrogen have been meas- 

 ured photometrically by Ferry.' The tube was excited by means of a 

 battery of accumulators consisting of 1,200 cells, the same as Ang- 

 strom used for his infra-red work. He found that at constant pressure 

 the intensity is proportional to the current, and with constant current 

 the intensity increases with decrease in pressure, in a regular manner, 



^Berndt : Jahrb. der Radioaktivitat u. Elektronik, i, p. 247, 1904. 

 ^Drew: Phys. Rev., 17, p. 321, 1903. 

 ^Coblentz & Geer : Phys. Rev., 16, p. 279, 1903. 

 * Ferry : Phys. Rev., 7, p. i, 1898. 



