DOC.
60
393
for
the
statistical properties
of
the
radiation
pressure
as
does that of
Planck.
As
far
as
interpretation
is concerned,
the first
thing
to note
is
that
the
expression
for the
mean
square
of fluctuation is
a sum
of
two terms. [29]
Thus,
it
appears
that there exist
two
different,
independent
factors
causing
a
fluctuation
of
the radiation
pressure.
From
the fact that
A2
is
propor-
tional
to
f,
we
conclude that
pressure
fluctuations for adjacent parts
of
the
plate,
whose
linear dimensions
are
large
compared
with the
wavelengths
of the
reflection
frequency,
are
mutually independent events.
The
wave
theory provides
an
explanation
only
for the
second term of
the
expression
found
for
A2. According
to
the
wave
theory,
beams
of
not
very
[30]
different
directions,
not
very
different
frequencies, and
not
very
different
states
of polarization
must
interfere with
each
other,
and
to
the
totality
of
these interferences,
which
occur
in
the
most random
fashion, there
must
correspond
a
fluctuation of the radiation
pressure.
That the
expression
for
this fluctuation
must
have
the
form
of the
second
term
of
our
formula
can
be
seen by a
simple
dimensional
analysis.
One can see
that the
wave
structure
of
radiation indeed
causes
the fluctuations
of
radiation
pressure
to
be
expected
from
it.
But
how
to
explain
the
first
term
of the formula? This
term
is
by
no
means
to
be neglected;
on
the
contrary,
it alone is relevant,
so
to speak,
in
the
domain
of
validity
of the so-called
Wien
radiation
law. For
A =
0.5
u
and
T
=
1700,
for
example,
this
term
is about 6.5
.
107
times
larger than
the
second
one.
If radiation consisted of
very
small-sized
complexes
of
energy
hv,
moving
through space independently
of each
other
and
reflected
independently
of each
other--a
conception
that represents
the
very
roughest
visualization of the
hypothesis
of light quanta-then the
momenta
acting
on
our
plate
due to
fluctuations of the radiation
pressure
would be of
the kind
represented
by
the first
term
alone.
Thus,
in
my
opinion,
the
following must
be
concluded
from
the
above
formula, which
is,
in
turn,
a
consequence
of Planck's radiation formula.
In
addition
to the
nonuniformities
in
the spatial distribution
of
the
momentum of
the
radiation
which arise
from
the
wave
theory,
there also exist other
nonuni-
formities in
the spatial distribution
of the
momentum,
which at
low
energy
density
of
the radiation
have
a
far greater
influence
then the first-mentioned