AND CHLORINE UNDER THE INFLUENCE OF LIGHT. 
113 
second when the action is a simple decomposition, such as occurs when a complex 
sulphur molecule dissociates into simpler molecules. The third when a re-arrangement 
of atoms takes place in a molecule ; and the fourth where the actions are complex and 
consist of firstly a combination and finally a breakdown of a complex molecule into 
simpler molecules. In this class it appears probable that all true substitutions and 
double decompositions should be placed."^ 
If we apply the Law of Mass Action to actions of the first three classes, we find in 
very many cases good agreement between the calculated and observed results for the 
velocity of the actions. But for the fourth class the case is different, as may be seen 
from an example. 
Consider a double decomposition represented by 
AB + CD = {ABCD} = AC + BD. 
In the ordinary theory this is a bi-molecular reaction, and the equation for the rate 
of action is dx/dt = hyz, where x is the number per unit volume of molecules of AC 
or BD and y, z are the numbers per unit volume of molecules of AB and CD; we 
suppose that the action in the reverse way does not take place. For simplicity, let 
y — z and x = n — y, so that AB and CD are originally present in the same concen¬ 
tration, n molecules per unit volume. Integrating the equation, we have 
y n 
But if the intermediate compound ABCD exists, we must take into consideration 
another step in the action. Let w be the number of ABCD molecules present, then 
we should have 
dwjdt = kyz — dxjdt and dxjdt = k'w, 
with the condition that initially w = 0. At first, then, dxjdt = 0. 
If k' is large compound with k, we should liave after a certain time 
dwjdt = 0 and dxjdt = kyz, 
so that after a certain time the action would go on as a bi-molecular action. But up 
to this time the velocity of the action would increase from 0 to a maximum value. 
In other words, we should have a period of induction. The duration of this period 
of induction will evidently depend on the quantities k and k. k were large 
This view receives some support from the phenomena connected with the decomposition of organic 
compounds, alone, and in the presence of nitrogen, by means of the silent discharge. The compounds 
produced frequently contain more carbon atoms in the molecule than in the initial substances, and can be 
explained on the assumption of the breakdown of additive molecules. See Berthelot, Comptes 
Rendus,’ 1898. The decomposition products of saturated hydrocarbons are also explainable in this way, 
e.g., acetylene is produced on heating methane. 
VOL. CCII.—A. Q 
