DOCS.
313,
314 MARCH 1917
305
I
have
only
one
test particle
and
no
other
physical[8] masses (sun, etc.) [From
physics
I
know
only
that
I
must find
guv's
very
close to
those
of
the
old theor.
of rel. for finite
distances],
but
perhaps
I
am going
to
need
supernatural
masses,
which
I
assume
to be
static,
thus
T44
=
p,
all
other
Tij
=
0.
Then
the
equations
become:
(in
coordinate
system III)
From this
follows:
for
the
fourdimensional
system
both
cases:
Gii (^
~b
2
^“P)9n
0
i
=
1,
2,
3
Gu

(A
+
^np)g4i
=
up
.
1
2
x+rp=w’
1
=
3
p
=
0
Therefore,
the
supernat.
masses
are
needed.
Supernat.
masses
do not exist.
I
am
curious
about
whether
you
can
agree
with this
approach
and
whether
you prefer
the
threedimensional
or
fourdimensional system.
I
personally
much
prefer
the
fourdimensional system,
but
even more so
the
original theory,
without the
undeterminable
A,
which
is just
philosophically
and
not
physically desirable,
and with noninvariant
guv’s
at
infinity.
But if
A
is
only
small,
it makes
no
difference,
and
the
choice is
purely a
matter
of taste.
I
hope
that
your
health
has
improved[9]
and
you
will
soon
be back
to
normal
again.
With
cordial
greetings, yours truly,
W. de
Sitter.
314. To Moritz Schlick
[Berlin,]
21
March
1917
Esteemed
Colleague,
Upon rereading your
fine
essay
in
Naturwissenschaften
I
do find
another
small
inaccuracy.
I
am
informing
you
of
it
in
case
your
article
is reprinted elsewhere.[1]
The
derivation
of
the
law
of
the
motion
of
a
point provided
on
page
184
assumes
that,
seen
from
the
local coordinate
system,
the
point moves
in
a
straight
line.
Nothing
can
be derived from
this,
however.
The local coordinate
system
is
generally
of importance
only
at
the
infinitesimal
level,
and at
the
infinitesimal