514 DOCS.

492,

493

MARCH

1918

involved.[8]

The

numerical

discrepancy

is connected

to

the

fact

that the

term

with

A,

which makes

the

constancy

of

g44

and

the

vanishing

of

the

pressure possible

in

the

first

place

in

my case,

is

missing

in

the

equations

in

Schwarzschild’s

case.[9]

An

interpretation

of

the

world

as a

whole

that

can

be in

any way

satisfactory

and

compatible

with

the

facts does

not

seem

possible

without the introduction

of

the

A

term in the

field

equations.

I

am

very

much

looking

forward to

your

new

lectures.[10]

Lately,

I

read

cor-

rection

proofs

of

a

book

by

H.

Weyl

on

the

theory of

relativity,[11]

which made

a

great

impression on me.

It

is

admirable

how

well

he masters

the

subject.

He

derives

the

energy

law of matter with

the

same

variation

trick

you

use

in

your

recently published

note.[12]

Finally,

I

heartily

thank

you

and

the

other

colleagues

of

yours

involved in

having an

immense scientific

prize assigned

to

me.

I

received word

today

from

Hamburg.[13]

In utmost

respect,

I

am

yours very truly,

A.

Einstein.

493. To

Gustav Mie

[Berlin,]

24 March

1918

Dear

Colleague,

In

joyous

anticipation of

your promised visit,

I

reply

just

briefly

to

your long

letter

of

the

21st.[1]

According

to

my

interpretation,

the distribution

of

matter

does

not

need

to

be such

that

p

is constant

in

reality.

It

suffices

that

spaces

exist

that

are

large compared

to

the

distance

to

neighboring

fixed

stars

but that

are

small

enough

to

be able

to

look

upon

the

metric deviation from Euclidean

behavior

as

still small. For such

a

region

of

space,

the

field equations

can

be

substituted

by

linear

ones,

where

Tuv

and

the

A

term

are on

the

right-hand

side.

The

field

at

the

boundary of

this

space

then

depends

just

as

little

on

the detailed

distribution

as

in Newton’s

theory.

Based

on

my

theory

of

1915,[2]

it

is

not

legitimate

to

regard an

Earth

enveloped

in clouds

as

rotating

in

reality.

For

the

gravitational

and

inertial

fields

are

viewed

as

an

indivisible unit.

This total

field

determines

the

result

of

Foucault’s

pendu-

lum

experiment

and-with

respect

to

a

coordinate

system

not

rotating

relative

to

the

Earth-satisfies

the

differential

equations.

At

the

time,

I

did not set

up

boundary

conditions

except

to characterize

special

cases.

The

differential

equa-

tions

are

satisfied

everywhere, though, no

matter how far

the

space

in which

they

are

to

be

tested

is

chosen to

extend,

and

no

matter how

the coordinate

system

is