DOC.

605

AUGUST

1918 631

605. To

Felix Ehrenhaft

[Ahrenshoop,

20]

August

[1918][1]

Dear

Colleague,

I

studied both

of

your long papers

with

avid interest.[2] The

energy

with

which

you pursue your goals

is

admirable.

Now

it

can no

longer be

doubted

that

your particles

are

spherical

and

of

the size

you indicate.[3]

Considering

the

great

care you

devote to

the

subject,

it

would be desirable if

you

also

proved

stringently

that the

air in

your

condenser

is not

brought

into motion

by

the

electrical

field.[4]

For

extremely

weak movements

already

suffice

to

destroy

the

rationality.

It

is

certain that the

formula for

the

Brownian motion must be correct

(because

of its

relation

to

mobility)[5]

since,

for this

derivation, only

the

law of

osmotic

pressure is needed,

which

we

know

happens

to

be

definitely

valid for

the

smallest structures

(of

molecular orders

of

magnitude).

Without

having

examined

the

calculations,

I

vaguely suspect

a

source

of

error

from

the

circumstance

that

it

is

not

the

momentary

position

of

the

particle

that

is

being registered,

but

some

temporal

mean

value. The Brownian

motion

appears

“smoothed out” to

the

observer. As

a result,

it

could be

merely an

apparent

shortening

of

the

perceived

paths.

Practical

testing

immediately produces

the

-t-law. The

A in

the

Brownian

paths

must be

proportional

to

the

root

of

the

observation time.

The limit

A/t2

for

a large

t

yields

the correct

value.[6]

Then

one

easily

sees

what

large (numbers)

times[7]

are

needed to circumvent

subjectively

determined

errors.

Your

negative

photophoresis

must be based

on an as yet

unknown

type of

momentum, i.e.,

a

secondary process

different from

radiation,

which

is

likewise

directed

and

is

immediately produced by

the

radiation.[8] For this

phenomenon

as well,

it would be

very

desirable

if

it

were

shown

that the

light ray causes

no

horizontal

motion in

the

gas, although you

have

already

removed

the thorn

from this reservation with

your

magnificent

proof

of

the

pressure independence

of

the

force of

photophoresis

and with

your

demonstration of

the

Schwarzschild-

Debye

maxima.[9]

If

you already

can

prove

from

your previous

measurements

that the

motions of

the

gas

could not have had

a

disturbing

influence

on

this

elementary

quantum

problem, please

let

me

know

the

grounds

for it. In

your

theoretical

survey

of

the

bases

for

an

electron

of

constant

size

you,

just

as

Mr.

Konstantinowsky,

failed to touch

upon

the

strongest

arguments.[10]

Bohr’s

theory

of

spectral

lines

is

conceivable

only

with

an

exactly

constant

e.

The

discovery by

Franck and Hertz

that

electrons at discontinuous

potentials

cause

the excitation

of

mercury vapor proves

the

equivalency

of

the

elementary charges

of

electrons.[11]

The

constancy

of

e[12]

likewise makes

the

constancy

of

both

quantities

at least

plausible. Quite

apart

from all

the

particular theories,

the

type

of lines

from

the