must leave the question undecided why and how the specific heat 
varies with the temperature and with the character of the substances. 
§ 4. If we now compare equation (7) and (8), if appears that: 
hd OA Ne nee See eee) BN 
and as according to § 2 neither 4 nor /’, can be dependent on 
the time during the reaction, equation (8) is satistied throughout 
the reaction. The same reasoning holds of course with very little 
change for dilute solutions too, and then also leads to equation 
(10). So it appears that in $ 1 we have defined these functions 
not closely enough, when we introduced them as functions of 
the temperature and of constants characteristic of the reacting 
substances, and eventually of the occurring intermediate states. For 
the supposition : 
OBE eh ee oj 
fa CA es | 
would be in accordance with this definition, in which «,, 6,, ¢, are 
characteristic of the system before the reaction, a, 6,, c, for the 
system after the reaction, and mutually independent. Nay, this 
supposition would even be the most obvious one. Equation (10), 
however, shows that it must be rejected. The constants in /’, cannot 
be independent of those in 4; they must be quantities which in 
some way or other are equally in relation with the two systems, 
that before and that after the reaction '). 
The simplest supposition then would be that all these constants 
were = 0, and so that / would be a pure general function of the 
temperature, like the '/, 27’/n7’*) from equation (11) of the preceding 
paper, or that possibly this too would be wanting, and /’ = 0 
might be put. However, on this supposition we come to just such 
an absurdity as made us reject equation (4) in the preceding paper. 
For from experimental determinations of: 
> ay 
I 
vV_£,.—1l=[vy,,+2v fe, df — Tv,|— dT + RT Er 
Peo; Tr p fee ray (oar Al R1 P, 
k. = Ce RT 
and . “Cor 
ss « id We 5 ain BS x - Mr TS is $2 pr ] 11 AIDS 
en ron ra fon rf en 
Jel 
ea Ce 
1) So the above considerations lead to the assumption of two opposed reactions, 
which are, however, not ‘independent’ of each other. For the functions F occurs in 
both velocities, ie. both partial velocities depend om the same “intermediate states“. 
u 
*) If the kT is brought into the exponent. 
