DOC.
25
SOLVAY
DISCUSSION REMARKS
397
from
theory.
A
more
careful
investigation
must
show
whether
this
conception
will hold
The
following
discussion remark
is
transcribed
from
Nernst
et al.
1914, p.
241.
See also Nernst
et al.
1912, p.
300.
A
manuscript
version does
not exist.
If the
forces
that
cause
the
oscillations
are
proportional
to
the
distance from
the
equilibrium
position,
then
it follows from
the
symmetry
of the
cubic
system
that
a
material
point
cannot
possess
two
frequencies,
at
least
not
as long as one
adheres
to
the
laws
of
mechanics.
In
a
comment following
Einstein's
previous remark,
Poincare brought
the
subject
of
the behavior
of
gases at
low
temperatures
into
the
discussion. In
the
course
of
the
ensuing exchange among
Nernst, Poincare,
Rutherford,
Kamerlingh Onnes,
Einstein and
Langevin,
Nernst
related this behavior
to
the
rotational motion
of
the molecules and mentioned
the
difficulties of
applying
the
"quantum
theory"
to
this motion. In his
Solvay
lecture,
Einstein criticized Nernst's theoretical
treatment
of
the
rotational motion
of
molecules,
and
made
a
remark similar
to
the
comment
printed below; see
Einstein 1914
(Doc. 26), pp.
350-351.
No.
181
(Nernst
et
al. 1914,
p. 242;
Nernst
et al. 1912,
p. 301)
12)
The
optical
and
energetical
investigation
of the
optical
properties
of
gases
with
a
diatomic
molecule with
an
electric
moment
is
in fact
of the
greatest importance,
because
from
the relation between the
coefficient
of
emission and
the
frequency (or
the
temperature,
if
the
frequency
is
given)
one can
obtain
directly (using electrodynamics,
to
be
sure)
the
statistical
law
of rotational
motion.
In
§6
of
his
lecture,
Nernst claimed
that
his
heat theorem
(the
third
law
of
thermodynamics) can
be derived
from
the
quantum theory
of
specific
heats. This claim
gave
rise to
an
extended
controversy
between
Einstein
and Nernst
on
the
status of
the heat
theorem, starting
with
the
discussion remark
printed
below. The conflict
resurfaced
during
the
second
Solvay
conference where it led
to
a lengthy
discussion
following
Grüneisen
1921
(see
Grüneisen
et al. 1921,
pp. 290-301).
No.
186
(Nernst
et
al
1914,
p.
243;
Nernst
et al. 1912,
p. 302)
13)
I would like
to
remark here
that,
as
far
as
I
can
see,
Nernst's heat theorem
cannot be
inferred
from
the
vanishing
of
the
specific
heat
in
the
vicinity
of
absolute
zero,
even
though
its
validity
is
made
much
more plausible by
this.
For
the
question is
whether,
in
a
sufficiently
close
proximity
to abs.
zero,
a
system
can
be
brought
reversibly
&
isothermally
from
a
state
A
to
a
state
B without the addition
of
heat. This could
not
be
inferred from the
weakness
of the molecular
agitation
if
the transition from
A to
B
could
only
be
produced
by using
this minimal
residue of thermal
agitation;
in
that
case
it would be
absolutely impossible
to
transfer the
system
from state A to state B at
absolute
zero.
Nernst's theorem
amounts to
the
assumption
(that is
quite plausible,
to