Mechanism of Molecular Action. 805 



In the case of reversible uni molecular actions (<?. g. 

 N 2 0, = 2N0 2 and SeH 2 = Se + H 2 ), the back action is de- 

 pendent on the collision of molecules, and is of one or other 

 of the above forms. The result of Part L shows, therefore, 

 that the dissociating tendency can also be expressed in one 

 or other of the same forms, i. e. as 



[P,]«"Vtf or [PJ/^. 



where [P 2 ] is the concentration of the undissociated ga«. 



The interpretation of this case, however, demands further 

 investigation. 



The velocity of action is found to be proportional to the 

 first power of the concentration ; this means that each mole- 

 cule acts quite independently of its neighbours, the action is 

 not due to collision, but to the internal state of the individual 

 molecule. 



The fact that many unimolecular actions are slow actions 

 shows that the molecules are not all in the same condition at 

 the same time — a certain proportion only being suitable for 

 dissociation at any particular instant. 



The additional fact that the dissociating tendency changes 

 with the temperature in accordance with a function of the 

 above (exponential) shape, suggests and requires that the 

 dissociation should be caused by the collision of particles 

 internal to the molecule (e. g. of atoms) moving with such 

 relative velocities that the chance that the members of any 

 particular colliding pair of atoms have a relative velocity 

 lying between v and (r + Sr) can be given in accordance 

 with Maxwell's law. 



For either one or other of the following arguments will 

 apply : — 



(1) The complete dissociating tendency is of the form 



a 



&[p,]«~V§, 



Avhich has been deduced in the first place on the assumption 

 that action occurs when the velocity of collision is greater 

 than a certain amount v. Making this same assumption for 

 the atomic as for the molecular collisions, the form 



a 



k{?{\e''\/e 



7 \ 0/ - -5> 



gives ft e /0 > 



the chance * that the relative velocity of a collision lies 

 * Of. Langevin, Le Radium, 1913. 



