DOCS. 470-472
AUGUST
1913
353
Only
now,
after
this
ugly
dark
spot
seems
to have
been
eliminated,
does the
theory
give me
pleasure.
With cordial
greetings, your very
devoted
A.
Einstein
471. To
Heike
Kamerlingh
Onnes
[Zurich,
18 August
1913][1]
Dear
Colleague,
Many
thanks for
your
postcard.
My
wife[2]
and
I
will look for
you
in
your
hotel[3]
after
lunch,
around
3 o'clock,
the
day
after
tomorrow
(Wednesday).
With
best
regards,
also from
my
wife, to
you
and
your
wife,
yours,
A.
Einstein
472.
To
Erwin Freundlich
[Zurich,
before 26
August
1913][1]
Dear
Colleague,
I
am
very
pleased
that
you
have
managed
to
arouse
such
great
interest
in
our
question regarding
a
bending
of
light
rays.[2]
From
a
theoretical
point
of
view,
the
matter
is
now
settled
more
or
less.
Privately,
I
am
quite
certain that
light rays
do
undergo
bending.
I
am
extremely
interested
in
your
plan
to observe
the
stars
near
the
sun during
daylight.
This should
be
possible
unless
suspended particles
of the order of
magnitude
of
optical
wavelengths,
which deflect the
light only slightly, are
present throughout
the
atmosphere.[3]
I
am
afraid that
your plan
could founder
on
that. But
you
certainly
know
more
about
these
things
than
I do.
Regarding
the
note in
that
astronomical
journal,
it
seems
that
you
have understood
me.[4]
If
one
works with
one
optical
system,
then the entire
sun
together
with
the
part
near
the
edge
in which
we are
interested
must be
gotten
on one
plate.
In that
case,
the
stars
to be
investigated
come
to lie
far
from the
optical
axis,
i.e.,
the middle of
the
plate,
which
is
almost certain
to
cause an
error.
With
that
in
view,
it
is
proposed
that
two
optical
systems
be
rigidly
linked to
each other
and
that
one edge
region
be
photographed
with
one
of the
two
systems,
the
opposite
edge
region
with
the other
one.
What
is
not
immediately
clear
is
how to
relate
the two
pictures
to
each other. Should
one, perhaps,
also make
a
photograph
of
well-known
portions
of
the
sky on
these
same
plates
in
order
to establish the
relationship
between the
plates? I
do
not
know,
but
cannot
think of
any
other
solution. I
am very
curious to
hear
further details
from
you
about the methods
under consideration.