SOiUC CONTE.XfrOR.tKV .ifH\tXCliS /.V I'llVSICS-lH 279 



Helium follows hytlroKt'n in tlif procession of elements. Its spec- 

 trum includes several line-series. The frequencies of the first four 

 members of one of these series, the principal series of the singlet or* 

 parhelium spectrum, are as follows (all the numbers in the successions 

 P. E. F. G. and // should be multiplied by 10'^): 



(D) 1.4.")7, .").nSl, 7.">()7, <S.:«)0 



Subtracting each from the frequeiUA' of the scriis-liniii , wliicli i;, 

 D.BIW, we obtain the succession of numbers 



(E) (O.GO*)- 8.152), (<1.()()<)- 3.628), (<).(1()'.)- 2.012), ('.l.dO'.t- 1.3()<)) 



which suggests a succession of orbits, ha\ iiig ihc tnllnw ini; c(insecuti\c 

 energy \alues": 



(F) -9.()()<ty/. -8.1.")2;/, -3.628//, -2.012/;, - l.:}0()/;. 



The consecutive frequencies of this series arc emitted when the \aience- 

 clectron falls from the second, third and consecutive orbits of this 

 succession into the first one. One would suppose that the \alence- 

 electron normally abides in this first orbit. But if this were so the 

 energy required to ionize the atom would be O.OOi)/; • lO", equivalent to 

 3.96 volts; and waves of the frequencies gi\en by (D) could displace 

 the electron and be absorbed thereby. But the ionizing-potenlial 

 of the atom is about 25 volts and the frequencies (E) do not appear as 

 dark lines in the absorption-spectrum of helium. Therefore there 

 must be still another orbit much deeper down, with a much higher 

 (negative) energy-value, than any listed under (F). In 1921-22 

 Lyman discovered (with his highly-cur\ed grating and shortened 

 light-path, and pumping arrangement for keejjing the pressure low) 

 a new series of lines of wave-lengths 584. 4A, 537.1. A, 522.3.-\ and 

 515. 7A. Their frequencies are 



(G) 51.34, 55.85, 57.44, .58.18 

 which may be written as the successif)n of numbers 



(H) (.59.49-8.15), (.59.49-3.64), (.59.49-2.05), (.59.49-1.31). 



Comparing these with the succession (E) we recognize the same set of 

 subtrahends," and accordingly identify the common quantity .59.49'10'* 



" It is customary to designate the orbits by their energy-values divided by he, 

 or 19.68- Itr". 



" It would not be neccssarj- to call attention to this if we could calculate the 

 frequency of the series-limit, which would give the energ>-value of the new orbit 

 immediately: but the four discovered lines are hardly sufficient for such an extra- 

 polation (there are fourteen gf the other series to use for calculating its limit). 



