202 
DR. A. VERNON HARCOURT ON THE VARIATION VTTH 
for this process. In the discussion of the experiments, I.-IX., h and c are con¬ 
sidered to he both zero, and a is the constant value of m deduced from the 
observations. 
In the series of experiments II.—XII. taken at random from a number of experi¬ 
ments by other authors on the effect of temperature upon chemical change it has 
been shovm that in the majority of cases II.-IX. in has a constant value. The 
reason for the variation of in with temperature in X.-XIII. has yet to be discovered. 
The writer intends to discuss at some future time all the experiments upon the 
subject, and hopes with the aid of Harcouet to discover an explanation of the 
variation. 
The experiments recorded in the Bakerian Lecture were made upon a simple 
unimolecular reaction, .and the utmost care was taken to eliminate all possible sources 
of error in the determination of the effect of temperature upon the reaction. On 
p. 860 the assumption is made that the relation is of the form 
= {[c+t)j{c+t^)Y\ 
Tlie value of c, the temperature of no action, is determined from the experiments to 
be —272°'6 C. This agrees closely with the zero of absolute temperature, —273° C. 
In the subsequent experiments of the Lecture, in Harcourt’s recent experiments, 
and in the experiments of other authors, the relation is for this reason assumed to be 
i-/i, = (T/T.)«, 
with the consequent relation 
in = 
In studying the effect of one phenomenon. A, upon another phenomenon, B, which 
mutually influence each other, it often happens that the chain of processes from 
A to B which cause the influence is unknown. It would seem reasonable, in our 
ignorance of these processes, to assume that when the chain is uniform the ratio of 
the increase per cent, in B to the increase per cent, in A is constant, and further to 
assume that if the chain is not uniform, and this ratio, therefore, is no longer constant, 
the ratio will still be a measure of the effect of A upon B. 
In the present case the phenomenon A is the molecular energy of heat, and the 
phenomenon B is the chemical energy of a reaction which may be energy of atoms or 
electrons. The mode of influence of the one kind of energy upon the other is 
unknown, and the measure in of the influence is taken to be 
m = (B-kffi)/(A-kZA). 
If A is of the form and B of the form another form of the measure is 
in — {v~^dv)f{u~^du), 
and in is the ratio of the percentage increase of the average velocity of atoms or 
electrons to the percentage increase of the average velocity of molecules. 
