314
DOC.
411
JUNE
1912
I
separate
these
two
conjectures
from
one
another because
it would
surely
be
conceivable
that
in
general,
to
be
sure,
the
equivalence
obtains
only
for
oo
small
fields,
but that
in
the
special
case
of Born's
hyperbolic
motion the
equivalence
also
obtains
in
the
finite.
Permit
me,
furthermore,
to
use
the
following
abbreviated
expressions:
micro-
equivalence
requirement (micro-req.); macro-equivalence req (macro-req.)
I have
been
searching
your
papers
for
those
places
from which
you
draw
the
inference
that the
macro-req. is
not satisfied in
general. Only
in
§4
of
your
3d
paper
(pp.
455)
and
456)[14]
can
I
find such
places.
I
hope
I
understood
correctly
that
§4
of
your
2d
paper[15]
(in
which
you carry on
a
polemic
with
Abraham)
is not
directly
related
to
the
macro-equivalence question
even
though
it
sounds
so
similar!
But
this
§4
rests
on
rather
complicated
dynamical
considerations. Of
course,
there
you
discuss all
the
possible ways
out,
and find all
of
them closed.
But
all this
is
still
very
complicated
and
one
cannot help
but feel that
you
perhaps
overlooked
some
possible
way
out.
This
is,
after
all,
the
reason
why you
content
yourself
with
expressing
merely
the
conjecture
that
only micro-equivalence
can
be
required.
o
The circumstance that
you
operate
with
truncated series
expansions
also
makes
an
overview of
this
question
more
difficult.
Given this state
of
affairs,
it
is
certainly
not
uninteresting
to
try
to
answer
the
following
question:
One
shall
find
the
most
general laboratory
motion
for
which the
macro-equivalence
requirement is satisfied. [The
"macro"
problem]
Of
course,
the
concept "macro-equivalence reqirement" is
in
need of
analysis
here.
1.
We confine ourselves to
equivalence
with
a
stationary
gravitational
field.
2.
We confine
ourselves
to
equivalence
with
respect
to
experiments involving geometrical
optics
and
to
velocity-of-light
measurements.
We do
not
treat
the
requirement
of
equivalence
with
respect
to
dynamic, electrodynamic,
and
thermodynamic experiments.
For
it
is
much
easier
to
introduce the latter after
it has first
been established
which class
of
labor. motions
satisfies
the
optical macro-eqiv.
req.
3. I
will
now
break
up
this
optical
macro-equiv.
req.
into
a
series of
partial
requirements,
and
would leave it to
you
to
decide
which
of these
partial requirements
you
want
to
view
as
not
relinquishable
and which
ones as
possibly relinquishable.
I
list
these
requirements
one
by
one
and shall show
later
to
what
solutions
of the "macro
problem"
one
is
led
depending
on
whether
one
views this
or
that
group
of these
requirements
as
not
relinquishable.
o