[ • 729 ] 



LXXXI. Hydrogen Spectra from Long Vacuum Tubes. By 

 R. W. Wood, Professor of Experimental PJiysics, Johns 

 Hopkins University, Baltimore* . 



[Plates XXI. & XXII.] 



I^HIS communication deals extensively with the changes, 

 chemical perhaps, which take place under the influence 

 of the discharge in vacuum tubes, and the modifications of 

 the spectra which result. 



In the Proc. Hoy. Soc. xcvii., I showed that, by employing 

 a vacuum tube of gre.it length, and utilizing the light of the 

 central portion only, it was possible io photograph the 

 Balmer series of hydrogen down to the 20th line, a consider- 

 able advance when we consider that its intensity is only 

 about Y50 °f that °^ the 12th, the last line usually photo- 

 graphed in the laboratory. 



The gain resulted from the remarkable circumstance 

 that, in a very long tube (2 metres or more) under proper 

 conditions, the secondary spectrum is confined to the ends, 

 while the central part shows the pure Balmer spectrum. 



So many new and curious spectroscopic phenomena were 

 exhibited by this tube, that I shall give a resume of some of 

 the results already published, as it seems piobable that the 

 technique developed in the work with hydrogen will be of 

 value in connexion with the investigation of other gases. 



The work with hydrogen has been continued without 

 interruption during the past year, in the hope of still further 

 extending the series, and though this hope has not yet been 

 realized, much new information about the hydrogen spectrum 

 has been obtained. . 



With dry hydrogen it has been possible to get the till e 

 into such a condition that only the secondary spectrum 

 appears, the Balmer series vanishing entirely. This I 

 referred to as the " white stage"'" in the earlier paper-. 

 With the tube in this condition (which comes only after 

 several days of continuous operation) the spectra of other 

 gases, admitted to the tube after removing the hydrogen. 

 are very different from those shown under ordinary circum- 

 stances, nitrogen, for example, glowing with a golden-yellow 

 colour, almost exactly matching that of h-lium. This was 

 found to be due to the fact that the blue and violet nitrogen 

 bands had nearly vanished, while the red, yellow, and gree i 

 bands were very strong; the red anil green together gave 



* Communicated by the Author, 

 PJdl Mag. S. 6. Vol. 42. No. 251. Nov. 1921, 3 C 



