28 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 65 



as five magnetons already (§9), but Protofluorine (7), certainly, 

 should be an even more strongly "negative" element than Fluorine 



(y + 7). 



As far as these atoms are inactive, or combine only with Hydrogen, 

 their absence from the earth is to be expected, for they are light 

 enough to escape from the atmosphere, as even Helium is believed to 

 do slowly; but the absence (or excessive rarity) of Proto-oxygen 

 and Proto-fluorine must be attributed to unknown causes of the same 

 sort as condition the rarity of Neon, Krypton, and Xenon, and the 

 apparent absence of the analogues of Manganese. 



Strong evidence for the existence of the proto-atoms is the occur- 

 rence in the spectrum from the corona of the sun (where gravitation 

 is much stronger than on the earth) of the bright unfamiliar line 

 attributed to an unknown element, Coronium ; on similar grounds 

 an element Nebulium is believed to exist in the nebulae. Such ele- 

 ments as these could apparently find no place in the Periodic Scheme 

 except before Helium. (The proto-elements have been discussed, 

 though from a different point of view, by J. W. Nicholson (Phil. 

 Mag., 22, 864, 1911).) 



A noteworthy feature of this magneton theory is that it leads to 

 numerically identical constitutions for the atoms of the three elements 

 in each of the triplets in the transition group (Fe, "Co, Ni ; Ru, Rh, 

 Pd; Os, Ir, Pt). According to Moseley's calculations from the 

 Rontgen ray spectra of the elements (Phil. Mag., 26, 1024-1034, 

 1913), the constant difference of one unit (presumably one electron) 

 from atom to atom applies to these elements just as to the rest. 

 There may indeed be some such regular difference in the nucleus, 

 but we have seen above that Moseley's results cannot well mean 

 anything for the " outer shells " of the model atoms of the present 

 theory, and the way in which the physical and chemical properties 

 of these elements throw them together in one group suggests strongly 

 that there is in their case some less fundamental difference in the 

 structure of that part of the atom. 



PART III. VALENCE 



§9. Two Kinds of Combining Action and Three Kinds of Bonds 



There is no simple term in general use for the " combining action " 

 of an atom that is broad enough to include the ideas of a numerical 

 factor (valence), an intensity factor (affinity?), and sign (in the 

 conventional chemical sense), all within itself. I shall therefore 

 frequently speak of the action of an atom, to include all this. 





