STRUCTURE OF CRYSTALS — WYCKOFF. 221 



point to the existence of certain especially stable configurations of 

 electrons in atoms. For instance, sodium with one more and mag- 

 nesium with two more electrons than neon readily lose one and two 

 electrons, respectively, in becoming positive sodium and magnesium 

 ions. Furthermore, as the electronegative elements are approached 

 a strong tendency becomes evident to add on enough electrons be- 

 yond the number required for atomic neutrality to complete the next 

 stable arrangement. If one of the atoms of a compound is highly 

 electropositive and another strongly electronegative, the latter may 

 be able to extract as a result of this tendency electrons from the 

 positive atom; the compound in the solid state thus becomes, as we 

 have supposed to be the case with the alkali halides, an aggregate 

 of an equal number of positive and negative " ions." Such a union 

 has been called a polar bonding. If, however, the combining atoms 

 are both either strongly electronegative or are to be found among 

 the elements of an amphoteric and less pronounced electrical char- 

 acter, then each atom will strive, without complete success, to abstract 

 electrons from the other in order to complete its especially stable 

 group. The electrons thus held in common b}- two atoms form a 

 second sort of bonding — the valency bonding. Since a complete 

 valence unit (in the old chemical sense) consists both of the tendency 

 of one atom to acquire an electron from another and the tendency 

 of some other atom to take one of this same atom's electrons, the 

 number of electrons in combination with any particular atom may 

 be twice as great as its chemical valence. Such valency bondings 

 must be supposed to be operative between the atoms in those com- 

 pounds of the sort discussed which are not of the sodium chloride 

 type. This explanation accords well with the natures of such com- 

 pounds as the diamond and carborundum, and accounts in a satis- 

 factory manner for the unions between the chlorine atoms in chlorine 

 gas or for the bonding between carbon and oxygen in the carbonate 

 group. It is surprising, however, to find some metals, as zinc and 

 copper (in zinc sulphide and cuprite), appear to exhibit the same 

 sort of union in some of their compounds. It should be pointed out 

 that the only metals for which this rule fits, iron, zinc, and the like, 

 are without exception atoms in the center of a long period of the 

 periodic table. About the arrangement of the outside electrons of 

 these atoms nothing is known at present. Of course, it may be 

 argued that we are dealing with only chance agreements with a rule, 

 but it must be remembered that such structures as zinc sulphide de- 

 part so far from being closely packed arrangements of atoms that 

 some sort of directional character to the forces of combination be- 

 tween their atoms seems necessary in order to account for their ex- 

 istence as stable groupings. 



