282 BELL SYSTEM TECHNICAL JOURNAL 



The "valence-electron" rays emitted by ionized atoms should lie 

 at lesser \va\e-lengths (roughly }/i as great) than the \alence-electron 

 rays of neutral atoms, and therefore should be particularly at home 

 in the region newly opened to exploration. The highest frecjuencies 

 emitted by the ionized-helium atom are perfectly calculable from 

 Bohr's theory; they are the frequencies of the Lyman series, quad- 

 rupled, and the wave-lengths therefore lie between 304A and 230A. 

 They have not been reported, but Lyman in 1919 obser\ed two lines 

 in the spectrum of violently-excited helium, near the positions 1214.9 

 and 1640. lA calculated for the first and third lines of the next helium 

 series (having the frequencies of the Balnier series, quadrupled). The 

 place of the second member of the series was obscured by an alien line. 



Fig. 4- Spictriini of a vacuum spark ln'twc'ci 

 Joiirnat. ) 



arbon electrodes. {Aslrophysical 



The once-ionized lithium atom, judging from the example of neutral 

 helium, should display higher frec|uencies than any other once-ionized 

 atom, and they should be arranged in recognizable series, somewhere 

 near the extreme limit of the explorable range as it stands at this 

 moment. Tlu'\' liaxc nni, howe\er, been reported; Millikan says 

 that his plates >lio\\ no liiliium lines of an\' sort from 17()()A down 

 at least, to 370A, if not fartiier. 



As an example of a speclruiu ixlcndiiii; far inio tlic ni\vl\-con- 

 quered field, a plate representing the spectrum of a vacuum spark 

 between carbon electrodes is reproduced from one of Millikan's 

 articles as Fig. 4. The actual spectrum is in the middle; it is drawn 

 out for better intelligibility, at the side. Most of the marked lines, 

 including the extreme line at 3()0..5A and the strongest line at 1.33oA, 

 are attributed to carbon; some to other elements, particularly the 



