34 Larmor, Physical Aspect of the Atomic Theory. 



principle, must, however, be a matter of experimental 

 scrutiny. 



The acceptance of the idea of ionic dissociation by 

 solution has been opposed by scruples of a more funda- 

 mental kind, not altogether unlike the difficulties once 

 attaching to the Berthelot doctrine that the extent to 

 which a reaction proceeds depends on the relative 

 amounts of reacting substance. It is a fundamental 

 postulate that a molecule is a self-existent aggregate, 

 whose intrinsic binding affinities are independent of 

 temperature : as we have seen, one of the main a posteriori 

 reasons for this conclusion is that each (sharp) line of the 

 spectrum is characteristic of the molecule, alterable in 

 position only very slightly, or not at all, by any change 

 of physical conditions. And this view agrees with the 

 kinetic theory which connects temperature with the 

 average translatory (and concomitant rotatory) motions 

 of the molecules in space, and sometimes with partial 

 dissociation, but not with any intrinsic change in 

 structure in the molecules that are present. Thus 

 the bonds of atomic affinity which have to be over- 

 come, say in the ordinary non-ionic dissociation of a gas, 

 are the same at high temperatures as at low. But occasion- 

 ally a collision with another molecule may be well-directed 

 towards breaking these affinities, like the sharp impact of 

 a mason's trowel on a brick or tile, and as a rule it will be 

 the more effective the higher the temperature. The verifi- 

 cation of the theoretical law of equilibrium, in ordinary 

 gaseous dissociation, enables us to assert that strong 

 affinities are in fact occasionally thus shattered ; that high 

 affinity is to be measured not by entire absence of disso- 

 ciation but by its relative rarity, though the products of 

 disruption can be accumulated when the opportunities for 

 recombination are removed. The dissociation into ions 



