Results of Crystal Analysis. 427 



We have previously found that the constitution o£ the 

 solid state o£ zircon, rutile, and kassiterite should be 

 Zr0 2 Si0 2 , (Ti0 2 ) 2 , an d (Sn0 2 ) 2 ; but, from what we have 

 said, we should not from these formulae be able to conclude 

 that they were adequate expressions for the chemical 

 properties of the substances. 



If we assume xenotime to be a phosphate, we might 

 equally well suppose that these substances were to be 

 considered as silicate, titanate, and stannate from the point 

 of view of chemical properties. 



Another way of putting it would be to say thai the con- 

 stitution formula varies with the state of the substance. 



The ordinary chemical constitution formula is intimately 

 related to the idea of a molecule; but in the crystalline 

 state the idea of a molecule as an individual system has lost 

 its significance. In certain cases it may be convenient for 

 the description of the structure in quite a formal way to 

 regard molecular elements, but they must not be supposed 

 to be real individual molecules. The forces connecting 

 neighbouring atoms of different molecular elements may 

 be equally strong and of essentially the same nature as 

 those existing between the atoms of the same molecular 

 element *. 



And when the idea of molecules loses its meaning, it 

 is really no wonder that the chemical constitution formula 

 ceases to express the structure of solids. 



In the liquid and gaseous state, however, we have to 

 do with individual molecules, or molecules dissociated into 

 ions ; and it may be possible that in these states the atoms 

 would arrange themselves more in agreement with the 

 chemical constitution formula. This may also be very 

 likely to occur, on account of the fact that the chemical 

 reactions which determine the chemical constitution mostly 

 take place in aqueous or gaseous systems. 



As the result of our reasoning, we might say that the sub- 

 stances belonging to the zircon group — xenotime included — 



* If we suppose that in a solid the atoms are not connected up into 

 molecules, the whole lattice structure will be kept together by the 

 same kind of forces as those which are engaged in the chemical binding 

 of the atoms of a molecule ; in other words, the elastic forces of a solid 

 body should be intimately related to the affinity forces which constitute 

 the attraction between atoms. 



Now the forces engaged in the binding of the atoms of a molecule are 

 much greater than the forces by which the molecules of a liquid are kept 

 together, and thus we see that the structure of the crystals gives a natural 

 explanation of the great increase in the elastic forces which accompany the 

 passage, of a substance from the liquid to the solid 



