81.8 Mr. H. Smith on the Spectroscopy of the Electric 



In very dilute solutions the line A, = 4368 of the series 

 spectrum appears distinctly in the negative brush. Wherever 

 the elementary line spectrum appears this line cannot be 

 detected on account of the presence of the lines A = 4367 and 

 \=4369 from which it is not separated on account of the 

 low dispersion. Its appearance in the negative brush in very 

 weak solutions where the potential is still high, shows that 

 the absence of the elementary line spectrum is due not to a 

 lack of oxygen but to a change in the electrical conditions. 

 The series spectrum does not appear in the stronger solu- 

 tions. Its behaviour was shown in the previous paper to be 

 similar to that of the series spectrum of hydrogen. 



With the data available it is not possible to settle definitely 

 how the elementary line spectrum arises in the different cases. 

 In air it appears in the spark without self-induction. It is 

 developed in the capillary of a Greissler tube with a condensed 

 discharge when the pressure is sufficiently low. It does not 

 appear in the capillary with the ordinary discharge, but 

 appears together with the negative bands in the neighbour- 

 hood of the cathode. That the spectrum should appear in 

 the positive brush is not surprising considering its spark-like 

 character. To see whether the fluctuations in the current 

 were responsible for the spectrum, a positive brush was 

 placed in series with a negative in a solution of sulphuric 

 acid, so that the same current flowed through each brush. 

 The spectrum appeared readily in the positive brush but it 

 was very weak in the negative. A negative brush in caustic 

 potash was then placed in series with the negative brush in 

 dilute sulphuric acid, the two cells being connected with un- 

 guarded platinum wire. While the negative brush in caustic 

 potash yielded the oxygen spectrum very brightly, the brush 

 in sulphuric acid showed only traces of the spectrum. Con- 

 sequently fluctuations in the current seem to have little 

 connexion with the spectrum. A condensed discharge was 

 now sent through the brush in sulphuric acid, the condenser 

 being in parallel with the cell and a spark-gap in series with 

 it. It was found that when the spark-gap was arcing across, 

 the oxygen spectrum was developed quite strongly; when 

 the spark-gap was kept working correctly with an air draft 

 blown across it, the oxygen spectrum could not be detected. 

 On account of the brightness of the continuous spectrum in 

 the latter case, the exposure had to be kept short so that the 

 decrease in intensity in the oxygen spectrum may only be 

 relative. The experiment shows, however, that a change in 

 the electrical conditions will develop the oxygen spectrum 

 in the negative brush in acids. 



