SO.\[E CONTEMf'OK.lKY ADVANCFS IX I'llVSlCS III 



277 



tifld as the dislaiue from tlicm increases). The same ar^;iiment 

 apphes to elements of the second, third, and foiirtli cohimns, though 

 with diminishing force, for the series become more difficidt to trac<? 

 and ilepart greatly from the archetype. In the crowded and com- 

 plic.ited spectra of elements sucli as neon, argon, and iron, it is very 



Fig. 3 — Periodic table of the elements showing their atomic numbers and ionizing- 

 potentials. (Cf. footnote 1.^.) 



difficult, though apparently not impossible, to arrange frequencies 

 into series, and this is in accord with the belief (founded on evidence 

 of other kinds) that in these atoms there is no single outer electron 

 far beyond all the others, but rather an outer shell of several similarly- 

 placed electrons. Any one of these might imitate the beha\ior of a 

 valence-electron, however, when removed to an unusually large dis- 

 tance from the nucleus and from the rest. It is to be observed also 

 that when atoms are brought close together in the liquid or solid 

 state, the line series can no longer be excited. 



Wherever, therefore, there are discernible line-series, one infers 

 an electron far enough beyond all the others to have a behavior and 

 deserve a title of its own. Generalizing Bohr's wonderfully successful 

 model of the atoms of hydrogen and ionized helium, we imagine that 

 this electron enjoys a particular set of orbits, in the narrowest and 

 dee{)est-lying of which it normally abides, while in an\ one of the 

 others it can make only a transient halt.'" 



" h may not be sii|xrl1uous to complete the description of Bohr's modtl by saying 

 that when the electron goes from one orbit to another, the difference li' between 

 the values of the energy of the atom in the two states is radiated in a ray of fre- 

 quency ^U/h. 



