562 DOC. 27
DISCUSSION
OF
DOC.
26
Discussion
comments
to Einstein
1914 (Doc.
26).
Published
in
Verhandlungen 1914,
pp.
353-
364.
[1]In
Lorentz
1912 (Lorentz
1914),
Lorentz
argues
that
the
energy
distribution
of
black-body
radiation is
incompatible
with
a
description
of
this
physical system
by
Hamilton's
equations,
because,
as
he shows in
this lecture
by applying
methods of statistical
mechanics,
these
equa-
tions lead
to
the classical
equipartition
theorem which
experiment
had shown
not
to hold for
black-body
radiation.
[2]In
Einstein
1914
(Doc. 26),
pp. 347-348,
Einstein
briefly
discussed
the
possibility
of
giving
up energy
conservation.
[3]Beginning
with Einstein
1905i
(Vol.
2,
Doc.
14),
Einstein
had
on a
number of
occasions
used Boltzmann's
principle to
draw conclusions about the statistical
properties
of
black-body
radiation and other
physical systems
from
their
thermodynamic properties. See,
in
particular,
Einstein
1909b
(Vol. 2,
Doc.
56),
pp. 187-188,
and the
introductory
section to Einstein 1910d
(Doc.
9);
for
a
historical
discussion,
see
Vol.
2,
the
editorial
note,
"Einstein
on
the Foundations
of Statistical
Physics," pp.
54.
For Einstein's
controversy
with Planck about
the
notion
of
probability,
see
also Doc.
25,
pp.
506-507.
[4]The
following
argument
was
first
given
in
Einstein
1906b
(Vol.
2,
Doc.
32),
pp.
375-376.
[5]For
Perrin's
original report
on
his experiments,
see
Perrin 1908.
He also
presented
his
findings
at
the
Solvay Congress;
see
Perrin
1912,
pp.
176ff,
or
Perrin
1914a,
pp. 145ff.
[6]The
problem
of
the choice
of
the
parameter
with
respect
to
which the
probability
of
a
state
is
defined became the
subject
of
correspondence
between D.K.C. MacDonald and Einstein
in
1953.
[7]Prior
to
presenting
the
following argument
at
the
Solvay Congress,
Einstein had discussed
it with
Besso;
see
Einstein
to
Michele
Besso,
second half of
August 1911,
and Einstein
to
Michele
Besso,
11
September
1911.
It
is
also alluded
to in
Einstein's lecture
notes
on
kinetic
theory (see
Doc.
4,
[pp. 50-51]).
[8]As
correctly printed
in
the French
version,
Einstein
1912a,
p.
441,
the
S
should
be
a
d.
[9]See
Gibbs 1902.
For
a
discussion of the
relationship
between the work of Gibbs and that
of Einstein
on
statistical
physics, see
Vol.
2,
the editorial
note,
"Einstein
on
the Foundations of
Statistical
Physics," pp.
41-55.
[10]The
following argument was
first
presented
in Einstein
1905i
(Vol. 2,
Doc.
14).
[11]Einstein had earlier
pointed
out
the
following
observation
in
Einstein
to H. A. Lorentz,
23 May 1909,
and
in
Einstein
1910b (Doc.
5).
[12]Planck
1914, p. 102,
and
Planck
1912, p.
84.
[13]Lorentz
is
probably
referring to
the first
term
in the
formula
for
(E/E)2
on p.
342 of
Einstein 1914
(Doc. 26).
[14]Kamerlingh
Onnes referred
to
a
passage
deleted
in
the
published
version
of
Einstein's
lecture;
see
note 47
of Einstein
1914 (Doc.
26).
[15]The
remark
appears
on
p.
301 of
Nernst
et
al. 1912
and
on p.
242 of
Nernst
et
al
1914
where
Kamerlingh
Onnes mentioned
a
calculation of the
specific
heat of
hydrogen according
to
Nernst's
theory.
[16]For
this formula and Einstein's
criticism of
it,
see
Einstein
1914
(Doc.
26), p.
351,
fn.
1.
[17]Nernst and Lindemann
1911b.
[18]The conversation
probably
took
place during
Einstein's
visit to
Leiden
to
deliver
a
lecture
in
early 1911; see
Einstein
to H. A. Lorentz,
15
February
1911,
where Einstein mentions his
conversations with Lorentz "about the
quanta
in relation to
the oscillation of material
systems"
("über
die
Quanten
bei
der Oszillation materieller
Gebilde").
[19]This
is
one
of the earliest
uses
of
adiabatic invariants
in
the
context
of
the
quantum
hypothesis;
for Ehrenfest's earlier
discussion,
see
Ehrenfest
1911b,
in particular,
p.
94;
for
evidence
that
Einstein had read Ehrenfest's
paper
before
attending
the
Solvay Congress, see
Einstein to
Michele
Besso, 21
October
1911.
For
a
discussion of the
relationship
between
this
remark
by
Einstein and Ehrenfest's adiabatic
principle, see
Klein,
M.
1970, p. 269,
fn.
8.