DOCS.
400,
401
NOVEMBER-DECEMBER
1917 409
you
will
be luckier at
the
search.
I
am
firmly
convinced
that,
in the
end,
all
field
quantities
will
emerge
as
essentially
identical
[wesensgleich].
But
premonition
is
easier
than
discovery.
5)
Whether the
guv's
are
ascribed to
an
ether
or
not is totally unimportant.
At
all
events,
one
should not
think
of
ether
as
anything analogous
to normal matter
insofar
as a
uniform
translatory
motion relative
to
it
seems
to have
absolutely
no
physical reality.
That
is
why
it does
seem
better
to do
away
with this concession
to handed-down conventions.
In
great
respect,
A.
Einstein.
401. From Hans
Thirring
[Vienna,
3
December
1917]
After
an
interlude
of several
months,
which
was
filled
with intensive
practical
work,[1]
I
now
finally
come
back
to
scientific
pursuits
and
am
just
busy writing
up my
little
relativity
paper
of
the
summer
and
publishing
it in
two articles
in
Phys.
Zeitschr.[2]
In
the
process,
some
doubts
arose
which
I want
to hold back
from
the
publication,
for the time
being,
but
which
I
would
like to
discuss
with
you directly.
According
to
the
gen.
theory
of relativity
there
is
a
spatio-temporal
distribu-
tion of
mass,
the
gravitational field of
which
is
equivalent
to
a
“centrifugal
field.”
One could also
say
from observations
of fixed stars
that this
spatio-temporal
dis-
tribution
is
such
that,
on
the
whole,
the
masses
rotate like
a
rigid body
around
an
axis. Let
us
say,
at
angular
velocity
w
around
the Z
axis. If
the
gravitational
field
of these
masses
is
to be
equivalent
to
a
“centrifugal
field,”
then the
following
field
components[3]
must
exist at
a
point
of
origin
on
the Z-X
plane
(x4
=
t):
T144
=
Cw2x +
2T24 =
C'w
where
C
and
C'
are
constants.[4]
(The
calculation of the field
for
a
rotating
hollow
sphere
actually
does
yield
such
T’s
as well,
of
course.) Now,
naturally
one
will
demand
that
no
force
phenomena
occur on a
mass-point
that
is
likewise
orbiting
around the Z
axis at
angular velocity
w,
since from its
point
of view
everything
is
at rest.
The
x
component
of
the
motion
equation
now is[5]
d2x
ds2
{rj4 +
2(r{"i
+
rV
+
IV)}
+
[x,ÿ,i\
2-
2