608 On the determination of the rate of chemical change, etc. 
solution. We thus obtain the value 3 62 c.c. per min. for 
nitrogen, which is rather near to the corresponding value 133 c.c. 
per min. for oxygen at 20°, in view of the doubtful accuracy of 
the number assumed for the solubility of nitrogen. 
The most important results of this investigation may be sum- 
marised thus : 
When a chemical reaction takes place in a solution resulting 
in the formation of a substance gaseous at ordinary temperatures, 
the solvent becomes supersaturated with the gas. 
The degree of supersaturation may be so great that the solvent 
contains more than 100 times the normal amount of the gas, and 
from the solution the gas is gradually given off on standing. The 
excess of gas so dissolved may be almost entirely expelled in a few 
seconds by briskly agitating the liquid by rapid stirring. 
Investigation of the rate of decomposition of hydrogen per- 
oxide shows that under such conditions of efficient stirring the 
rate of evolution of a gas furnishes an accurate and reliable 
method of investigating reactions resulting in the formation of 
a gas. This has been proved also by Prof. Bredig. 
The decomposition of diazobenzene chloride is a unimoleeular 
reaction throughout, the velocity constant being much higher 
than the values previously assigned to it. The value of the 
velocity constant for the reaction is found to vary with change 
of temperature, increasing logarithmically with increase of tem- 
perature. 
The reaction between nickel carbonyl and iodine (in alcoholic 
solution) is of a complicated nature, even in cases where either 
reagent is in considerable excess. When the nickel carbonyl is 
present in great excess a remarkable phenomenon is met with, the 
action proceeding in a steady manner until a point is reached at 
which it abruptly terminates almost without warning. 
The decomposition of formic acid by sulphuric acid is a true 
unimoleeular reaction, and not bimolecular, as it has been formerly 
considered. 
The decomposition of urea in the presence of sodium hypo- 
chlorite does not follow a simple law. The rate of reaction increases 
to a maximum and then gradually diminishes. 
The rate of evolution of a gas from its supersaturated solution 
may be determined from experiments performed with and without 
stirring. The rate of evolution does not depend in a simple manner 
on the degree of supersaturation, probably being greatly affected 
by the nature of the vessel, area of free surface, presence or absence 
of effervescence. 
The author wishes to thank Mr H. 0. Jones for his interest in 
the work and for many valuable suggestions throughout the course 
of the experiments. 
