86 DOCS.
71,
72
APRIL
1915
Hence,
because
the
7uv’s can
be chosen
arbitrarily,
dx'a dx'r
Kr
_
\Zirg
^
dXft
dxv
a/-F
This consideration
evidently
cannot capture
what
you
mean,
because it
is
too
obvious;
but
I
hope
nonetheless
that
by
means
of
the
same
you
will
be able
to
lead
me
toward
understanding
more
easily
what
moves
you
to
object.-
You have
probably
received
my
letter
refuting
your
example.[3]
I
shall
repeat
myself.
My proof
of
the
invariant nature of
8J
fails with such infinitesimal
transfor-
mations in
which
the
guv's
of
the
original system are
constant, because
then the
quantities
Auv
cannot be chosen
freely
but
vanish
altogether.
But this
does
not
apply similarly
to
changing
guv's.
Thus
the
proof
does not fail
generally,
but
only
in
certain
special cases.
Since
the
example
relates to
such
a
special case,
it
proves nothing
about the
validity
of
the
principle
in
general.
I must
even
admit
that,
through
the
in-depth
considerations
to
which
your
interesting
letters have led
me,
I
have become
only
more
firmly
convinced
that
the
proof
of
the
tensor
character
of
uv/-g
is correct
in
principle.
With
cordial
greetings,
yours,
Einstein.
72. To Geertruida de Haas
[Berlin,]
13
Wittelsbacher
St.
[before
10
April
1915][1]
Dear Mrs.
de
Haas,
I have calculated
by
a new
method the
enclosed
curve,
which
your
dear hus-
band
recorded
particularly
meticulously.[2] Although
the
linearity
of
the
damping
term is used
here,
the
legitimacy
of
this
procedure
is
proven
on
the
basis of
the
curve.
It
is
odd
that the
small
displacements
seem so
systematically incorrect;
I
can give
no
explanation
for
it.[3]
The
good agreement
with the
theory is
by
chance,
of
course;
but
it
is
real
inasmuch
as now
any
doubts about
the
theory’s accuracy
are
necessarily
silenced.
I
ask
you
please
to send back the
manuscript
as soon as
possible
so
that
I
can
pass
it
on
to Mr. Scheel.[4] Do
not
forget
also
to
enclose
the observed
curve.
With
cordial
greetings
to
you
and
your
little
children, yours,
A.
Einstein.