Page:Elementary Principles in Statistical Mechanics (1902).djvu/130

This page has been proofread, but needs to be validated.

106

THE FUNCTION $\phi$ AND

e^{{\frac {\psi -\epsilon _{0}}{\Theta }}+\phi _{0}}\left({\frac {-2\pi }{{\Big (}{\frac {d^{2}\phi }{d\epsilon ^{2}}}{\Big )}_{0}}}\right)^{\frac {1}{2}}=1,

(342)

{\overline {\epsilon }}=\epsilon _{0},\qquad {\overline {(\epsilon -\epsilon _{0})^{2}}}=-{\bigg (}{\frac {d^{2}\phi }{d\epsilon ^{2}}}{\bigg )}^{-},

(343)

whence

{\frac {\psi -\epsilon _{0}}{\Theta }}+\phi _{0}={\tfrac {1}{2}}\log {\frac {{\Big (}{\frac {d^{2}\phi }{d\epsilon ^{2}}}{\Big )}_{0}}{-2\pi }}=-{\tfrac {1}{2}}\log(2\pi {\overline {(\epsilon -{\overline {\epsilon }})^{2}}}).

(344)

Now it has been proved in Chapter VII that

{\overline {(\epsilon -{\overline {\epsilon }})^{2}}}={\frac {2}{n}}{\frac {d{\overline {\epsilon }}}{d{\overline {\epsilon }}_{p}}}{\overline {\epsilon }}_{p}{}^{2}.

We have therefore

{\overline {\eta }}+\phi _{0}={\frac {\psi -\epsilon _{0}}{\Theta }}+\phi _{0}=-{\tfrac {1}{2}}\log(2\pi {\overline {(\epsilon -{\overline {\epsilon }})^{2}}})=-{\tfrac {1}{2}}\log {\bigg (}{\frac {4\pi }{n}}{\frac {d{\overline {\epsilon }}}{d{\overline {\epsilon }}_{p}}}{\overline {\epsilon }}_{p}{}^{2}{\bigg )}

(345)

approximately. The order of magnitude of

{\overline {\eta }}-\phi _{0}

is therefore that of

\log n

. This magnitude is mainly constant. The order of magnitude of

{\overline {\eta }}+\phi _{0}-{\tfrac {1}{2}}\log n

is that of unity. The order of magnitude of

\phi _{0}

, and therefore of

-{\overline {\eta }}

, is that of

n

.^[1]

Equation (338) gives for the first approximation

{\overline {\phi }}-\phi _{0}={\bigg (}{\frac {d^{2}\phi }{d\epsilon ^{2}}}{\bigg )}_{0}{\frac {\overline {(\epsilon -\epsilon _{0})^{2}}}{2}}=-{\tfrac {1}{2}},

(346)

{\overline {(\phi -\phi _{0})(\epsilon -\epsilon _{0})}}={\frac {\overline {(\epsilon -\epsilon _{0})^{2}}}{\Theta }}={\frac {d{\overline {\epsilon }}}{d{\overline {\epsilon }}_{p}}}{\overline {\epsilon }}_{p},

(347)

{\overline {(\phi -\phi _{0})^{2}}}={\frac {\overline {(\epsilon -\epsilon _{0})^{2}}}{\Theta ^{2}}}={\frac {n}{2}}{\frac {d{\overline {\epsilon }}}{d{\overline {\epsilon }}_{p}}}

(348)

The members of the last equation have the order of magnitude of

n

. Equation (338) gives also for the first approximation

{\frac {d\phi }{d\epsilon }}-{\frac {1}{\Theta }}={\bigg (}{\frac {d^{2}\phi }{d\epsilon ^{2}}}{\bigg )}_{0}(\epsilon -\epsilon _{0}),