734 Prof. R. W. Wood on Hydrogen Spectra 



background can be suppressed. Fig. 1, PI. XXL shows the 

 series lines 9 to 20 inclusive, the upper spectrum taken with 

 a current of 20 amps, in the transformer, the lower with a 

 current of 1 amp. The hitter shows the series only as far 

 as the 13th line, and nothing would be gained by longer 

 exposure on account of the continuous background : the 

 positions of the strongest lines of the 2nd spectrum are 

 indicated by dots placed below the spectrum. These lines 

 have all been identified with the lines measured by Watson 

 and others. With the heavy current seven more Balmer 

 lines nre brought out (the 14th widened and intensified by 

 the 3704 mercury line which lies very close to it on the long 

 \ side). The secondary spectrum lines have become rela- 

 tively weak, but a host of other lines have developed, which 

 have not been previously observed, so far as I can find. 



The distinction between this new spectrum and the 2nd 

 spectrum is better shown on PI. XXII. Spectra "c" were 

 taken with the tube approaching the " white stage *' (dry 

 hydrogen), which will be referred to more fully presently. 

 In this condition the 2nd spectrum is very strong and the 

 Balmer series weak. Spectra "a" were taken with the tube 

 in the black stage with a heavy current, and spectra " b " 

 with a weak current. The new spectrum is seen in the 

 "a" spectra, in which the secondary spectrum is almost 

 completely absent : the broad group of lines between the 

 9th and 10th hydrogen lines should enable one to identify it 

 at once if it has been found under other circumstances. 

 It is now apparent that it is this new spectrum and not the 

 secondary spectrum that masks the further extension of 

 the Balmer series. Nearly midway between the 4th and 5th 

 hydrogen lines, a little to the right of the Hg line, is a close 

 triplet (often very strong), PI. XXII. upper spectrum "a/' 

 with wave-lengths 4050-48, 4050*78, and 4051-54. Of these 

 the first two are of equal intensity, while the third is much 

 fainter. On one plate only the first line appears. I have 

 been unable to find these lines in any of the tables of wave- 

 lengths given by Kayser, and am very anxious to have them 

 identified, as they may furnish a valuable clue to what is 

 taking place in the tube. They appear only on plates made 

 when the tube show r s the Balmer lines at great intensity, 

 i. e. they seem to belong to, or at least to appear with, what 

 I have called the new spectrum, which develops only with 

 heavy currents. 



The spectrograms were made, for the most part, in the 3rd 

 order spectrum of a large plane grating used in connexion 



