THE SPACE RELATIONS OF ATOMS. 



(Continued.) 



^T^HE attempt to advance beyond the distinction between 

 A two- and three-dimensional molecules, and to as- 

 sign a definite form even to the least complex of atomic 

 aggregates has been found full of difficulty. In its simplest 

 form the problem is to find the relative position of the 

 centres of gravity of the atoms in a state of rest, i.e., at the 

 absolute zero. The fact that in the case of a diatomic 

 molecule these points must be in a line, and, in a triatomic 

 molecule, in a plane, does not help us unless we can de- 

 termine the relative lengths of the lines joining different 

 atom-centres, which at present we cannot. Above the tem- 

 perature of absolute zero complications are introduced by 

 the motions of the atoms, and as to the extent and direction 

 of these motions we know practically nothing. And in Van't 

 Hoff's view all inferences drawn from the movements of 

 an atom within a molecule are liable to an objection at the 

 outset. The objection is that the moment an atom begins 

 to move within a molecule it is liable to move out of the 

 molecule altogether. 



Van't H off argues — and the generalisation is as striking 

 as it is simple — that it is only at absolute zero that a mole- 

 cule is internally stable, seeing that all molecules must be 

 supposed to undergo dissociation at a sufficiently high tem- 

 perature, and experiment and theory agree in showing that 

 though dissociation diminishes with the temperature, it 

 would never cease short of the absolute zero. Above this 

 temperature, then, the equilibrium of a molecule always 

 depends on its exchanges with others. 



The inference is that atomic motion should, for the 

 present, be ignored in our attempts to determine atomic 

 positions. Indeed the argument apparently points to 

 atomic chaos as the only state possible above absolute 

 zero ; but the inter-molecular exchanges may be supposed 



