112 BELL SYSTEM TECHNICAL JOURNAL 



which seems to bear upon the arrangement of the electrons in the 

 atoms; but some of it leads to conclusions opposite to those which the 

 remainder suggests. 



On the one hand, line-series are discernible in the spectra of ele- 

 ments in the second and the third columns of the Table, and even 

 in those of some others; and from these line-series, systems of Station- 

 ary States are deduced which resemble those ascertained for the alkali- 

 metal atoms; and it is natural to extend the same explanation from 

 that case to these, supposing again that each atom consists of a 

 nucleus and a certain number of electrons, all but one of which are 

 tightly bound into a residue, around which the one remaining electron 

 circulates in one or another of various quantized orbits. 



On the other hand, the chemical behavior of these elements does 

 not confirm this easy classification of the N electrons of an atom into 

 (iV— 1) very-tightly-bound electrons and one which is very loosely 

 bound. Thus, the atoms of elements of the second and third columns 

 of the Periodic Table — "alkaline-earth metals" and "earth metals," 

 as they are called — when floating in water as the fragments of mole- 

 cules of dissolved salts of these elements, are found to be deprived 

 of two and of three electrons, respectively; and the composition 

 of these salts is such as to suggest that the atoms of the other element 

 or elements involved in them have annexed two or three electrons, 

 respectively, from the alkaline-earth atom or from the earth-metal 

 atom. These facts suggest rather that the N electrons of an alkaline- 

 earth atom, or of an earth-metal atom, should be classified into (iV— 2) 

 or (iV — 3) very-tightly-bound electrons and two or three which are 

 loosely-bound, respectively. The very tightly bound electrons will be 

 equal in number to, and presumably arranged like, the electrons of the 

 atom of the next preceding inert gas. Henceforth I will reserve the 

 word "kernel" for such a system, and the word "residue" for what 

 is left behind when one electron is separated in fact or in imagination 

 from the atom. Thus these two words will not mean the same thing 

 except in special cases, such as those of the alkali-metal atoms. 



Specifically, let us consider the four consecutive elements argon 

 (inert gas, 18th element of the Periodic Table), potassium (alkali 

 metal, 19th element), calcium (alkaline-earth metal, 20th element), 

 and scandium (earth-metal, 21st element). 



The evidence from chemistry and from electrolysis impels us to 

 think that the argon atom consists of a nucleus surrounded by (eigh- 

 teen) electrons tightly bound, in a stable and imperturbable arrange- 

 ment; that the potassium atom consists of a kernel much like the 

 argon atom, with one additional electron loosely bound and hence 



