SOMf: CiWTF.MfOR.tRy .ll^.tMhS IX rilYSlCS nil 455 



ilu'ii h.ul a (iTt.iin iu-ii.iti\i' \.iliii'. .ill ilu- nilu-r ener^jy-valiirs wvn- 

 iu'xaii\t' .iiul sc.ittfri'<l JK-lwi-fii {Uv>v iwd. In this case of lIu- X-ray 

 siHTtra, the eruTKy-valiR's of tlir siatioiian- states are nu-asured 

 tifnvards from the normal state of the neutral atom, to which the 

 enerj;y-value zero is assijjned. while all the other energies are |>ositivc. 

 In Fii;. 13 this zero-line must he imaijined just luuler the level marked 



The exact (Position of this zero-line for the hi^h enerj;\- stationary 

 states is not \-cry accurateK' known; although the distance between 

 any two levels is determinetl with all the usually \ery great exactitude 

 of X-ray wavelength-measurements, the distance from any level to 

 the zero-line is uncertain within a few tens of volts. This uncer- 

 tainty is not great enough lo be ini|)()rtaiit wlu-ii dealing with the 

 high-frequency X-rays. 



This point being attended to, we are now in position to consider 

 the striking difference between X-ray emission-spectra and X-ray 

 absorption sjiectra — striking indeed when one looks at typical photo- 

 graphs, apparently altogether a different matter from the contrast 

 between optical emi.ssion-spectra and optical absorption-spectra, yet 

 in principle very much the same thing. In dealing with optical 

 spectra, I remarked that while an atom may absorb any frequency 

 which it can emit — while the complete absorption-spectrum of a gas 

 is identical with its complete emission-spectrum, yet the absorption- 

 spectra one ordinarily sees contain only a small selection of the emis- 

 sion-lines. This occurs because when a gas is being examined for its 

 absorption-sfiectrum in the laboratory, by sending light through it, 

 it is generall>- in an untroubled and quiescent ccjndition, each of its 

 atoms being in the normal state; therefore it absorbs only such fre- 

 (juencies as provoke transitions from the normal state to the various 

 excited states, and not such frequencies as would induce transitions 

 from one excited state to another, for few or none of the atoms are in 

 any one of the excited states to start with. Such also is the case 

 with the X-ray spectra. Quiescent atoms absorb only such X-ray 

 frequencies as produce transitions from the normal state into one 

 of the stationarv- states flesignated by K, or L\, or Li, and so forth — 

 they do not absorb such frequencies as would produce the transitions 

 from L\ or Lj to K, for instance, for the atoms are not initialK- in the 

 states Li or Lj. This is quite the same behavior as is observed in the 

 resp<^)nse of atoms to radiations in their optical spectra. It is much 

 more pronounced, however; for, while it is possible to make a gas 

 absorb frequencies which produce transitions from one excited state 

 to another, by maintaining the gas in a state of intense electrical 



