Chemical Attraction between Atoms from Physical Data. 93 



We have already obtained Evidence of this before. The 

 writer * has deduced a formula for the radius of the sphere 

 of action of a molecule on the supposition that matter does 

 not consist of molecules but is evenly distributed in space. 

 The radius of the sphere of action obtained on this supposi- 

 tion decreases with rise of temperature. 



The decrease of the attraction with rise of temperature 

 may be partly direct and partly indirect. Indirectly a change 

 could be, and probably is, produced in the following way. 

 We have obtained some evidence f that the deviation from 

 the additive law for the attraction of a molecule is due to an 

 interaction between the atoms which decreases their attrac- 

 tion. Now the atoms in a molecule are probably in rotation 

 round the centre of gravity of the molecule, equilibrium 

 being maintained between the centrifugal forces and the 

 forces of attraction. The speed of rotation very likely 

 increases with an increase of temperature as this produces an 

 increase of the internal energy, and in order to maintain 

 equilibrium between the forces the molecule must contract. 

 This brings the atoms still further under each other's influence, 

 which decreases the force of attraction of the molecule at an 

 external point. 



Jf we assume that 



*ih t) 



is a function of the temperature only so that it may be 



/ T \ . 



written </> 2 f ™-j, the equation for the latent heat becomes 



L = A A g)(pf-^ 3 ), 



and that for the surface-tension 



/T 



where A 1 and A 2 are numerical constants. If the assumption 

 is true, the value of the expression 



Hpi-p-z) 2 



should be independent of the temperature. This is, however, 



* Phil. Mag. June 1910, pp. 840-846. 

 t Phil. Mag-. May 1910, pp. 79S-801. 



