chemical change hy measurement of the gases evolved. 599 
off in experiment B. The point P in experiment A corresponds 
therefore to the commencement of measurement of evolution of 
gas in experiment B, and since from these two points the curve 
representing the evolution of gas in the two experiments is the 
same, there is good reason for thinking that (apart from the 
presence of the products of reaction in experiment A, which, as 
Fig. 7 . 
we have seen, do not influence the rate of reaction) the composi- 
tion of the reaction mixtures at the time represented by the 
point P in experiment A is the same as that of the reaction mix- 
ture in experiment B at the commencement of measurement of 
evolution of gas. This would seem to indicate either that the 
reaction which results in the evolution of gas is not preceded by 
any change, or that such change is one which takes place very 
rapidly compared with the rate of the reaction which is measured 
by the evolution of gas. 
The nature of the reaction which takes place between nickel 
carbonyl and iodine is not elucidated by experiments in which 
the iodine is present in large excess. Under these circumstances 
the action which takes place follows no simple law. A representa- 
tive set of observations is given in the following table : 
Nickel Carbonyl solution ... 5 c.c. 
Iodine ... ... ... 20 c.c. 
