246 PHENOMENA, ATOMS, AND MOLECULES 



The nitrates and perchlorates of all metals are readily soluble and most 

 of them are deliquescent. From this similarity of properties we may con- 

 clude that the other isosteres of Types ii and 14 of Table I should be 

 interrelated. Thus carbonates and sulfates should show a general re- 

 semblance to each other in their solubilities. 



The smaller solubility of salts of divalent ions is, of course, due to the 

 greater forces holding the divalent ions, making it more difficult for them 

 to separate. The process of solution of salts involves, according to the octet 

 theory, the separation of the ions. Thus sulfates and carbonates of most 

 metals are difficultly soluble compared to the nitrates and perchlorates. 

 Salts like lead and barium carbonates and sulfates in which both ions are 

 divalent, are thus particularly difficultly soluble. 



Boron nitride, which consists of tervalent ions, has still greater stability, 

 and is infusible and insoluble in all solvents. 



The observed differences between the ordinary properties of the differ- 

 ently charged isosteres are thus to be expected. But there is one property, 

 namely crystalline form, which should depend on the arrangement of the 

 electrons in substances rather than upon the magnitude of the forces be- 

 tween their atoms. By a 'comparison of crystal forms it should therefore 

 be possible to obtain direct evidence of the similarity of the arrangement 

 of the electrons in isosteric substances even if the charges on the comole- 

 cules are different. 



For example, since the sodium ion and the fluorine ion are isosteric 

 and cubic in form, we should expect them to pack together in a crystal 

 lattice in exactly the way that Bragg has found they do. The magnesium 

 and oxygen ions, however, are also isosteric with the sodium and fluorine 

 ions, so that we should expect magnesium oxide to have a crystal structure 

 identical in form with that of sodium fluoride. According to Groth both 

 substances are cubic, sometimes crystallizing as octahedra, and both show 

 good cleavage parallel to the (100) plane. To answer the question thus 

 raised, Dr. A. W. Hull has studied the X-ray patterns obtained with 

 magnesium oxide and sodium fluoride and has thus recently found that 

 the crystal structures of both substances are alike, except that the atoms 

 in magnesium oxide are drawn closer together by the greater forces. Thus 

 magnesium oxide and sodium fluoride should be looked upon as iso- 

 morphous. By the octet theory the covalences of all the atoms in these sub- 

 stances are zero, since they do not share electrons with each other. The 

 isomorphism is thus in full accord with Mitscherlich's rule by which 

 isomorphous substances should have similar structures. Since the ions 

 0"~, F", Na" and Mg''* are isosteric we should also expect that MgF2 and 

 NaoO should be isomorphous. Groth states that sodium oxide has not been 

 obtained in the form of well developed crystals. However, by Hull's 



