304 



INFRA-RED EMISSION SPECTRA. 



Iron. 



In figure 136 are given the curves ot the arc between iron terminals. 

 Curve b is for the electrodes, while curve a (scale is one-half of b) 

 represents the distribution of the radiation from the vapors. Here the 

 vapors have a more intense radiation than the solid terminals. This 

 is no doubt due to the fact that we do not get the radiation from the 

 hottest part of the electrodes, while in the arc, which is very rich in 

 oxides, there is a better chance for the hottest region to emit radiation. 



In the Zn arc the oxides are formed so rapidly that it is almost 



Fig. 136. 



impossible to work with this metal. The problem then is to separate 

 the black-body radiation of the oxides from that of the vapors, which 

 is very different from the work in the visible spectrum. 



Chlorides of Metals. 



The chlorides of Na, Li, and K were then examined in the carbon arc, 

 using for the purpose hollow carbon electrodes filled with the salt. 

 The carbons used varied from 6 to 9 mm. in diameter, the holes being 

 from 1.4 to 2.5 mm. A direct current of 15 amperes from a 104-volt 

 circuit was used. The radiometer slit was reduced to o.i mm. in width, 

 nevertheless the radiation at 2 /a, which at most gave a deflection of 

 only a few millimeters, was not resolved into individual lines. This 

 is in marked contrast with the strong emission lines at i \x., and as will 

 be noticed later on, the continuous radiation at 2 /a would blot out any 

 weak emission lines, as far as a radiometer or a bolometer is concerned. 

 Here a photographic process would be better, since the effect upon the 



