THE IDEAS
OF
EINSTEIN'S
THEORY
THE EINSTEIN'S
THE THEORY SIMPLE
IDEAS
OF THEORY
OF
RELATIVITY
IN
LANGUAGE
BY
J. H.
PROFESSOR
THIRRING,
OF
Ph.D.
PHYSICS AT THE VIENNA
THEORETICAL
OF
UNIVERSITY
TRANSLATED
BY
RHODA
A.
B.
RUSSELL
WITH
DIAGRAMS
AND
AN
ILLUSTRATIVE
CHART
SBCOND
EDITION,
REVISED
METHUEN 36
"
CO.
LTD.
ESSEX
STREET
W.G.
LONDON
First
Second
Published
,
,
November
, ,
3rd
iQ2i jg23
Edition
....
December
PRINTED
IN
GREAT
BRITAIN
PREFACE
THE
us
beginning
a
of the twentieth
century presented
quickly became
:
with
scientifictheory which
all
over
celebrated
Theory
evolution of
the
world
the
Einstein
Relativity.
Whoever
is interested in the
more
of mental
progress
though
mailcoat
non
every
mathematician
apparent
barrier to
further investigation.
Such
matter
truly
great
idea, however,
which
contains
at large, must
be capable
consisting
of being
intelligible,without
mathematical Theory
can
solely of
This certainly
maze
of
formulae.
applies
to
the
of
Relativity ;
clear
be made
without
to
those who
have
training, and,
in point
of
number
their way
purpose of
mere
expositions have
already
literature. book
The
account
present
is not
to
to
give
an
details appertaining
the
theory,
vi
but
THE
IDEAS
a
OF
EINSTEIN'S
and
THEORY
exposition
all mathematical only be able
rather to give
whole,
at
complete
same
coherent
of the
the
time
avoiding
not
we
accessories.
to imderstand
The
what
is meant
maintain
masses
that
suffers
was
the
space
"
gravitational
to
see
curvature
be made
a
how
Einstdn
we
bound
to arrive at such
conclusion. of the
Hence whole
must
theory,
Special
Principle
of Relativity leading up
simple and
primitive form,
on
and
the
farreaching
speculations
the
finiteness of
remain
on
based
be passed
A
over,
it suffices to
us,
say
the suppositions
and
B
to
lead
with
deductions,
on.
to D,
and
to
so
By
arguing other
thus, and
inducing
conclusions
follow
a
each
we
chain,
shall
of
more
lucid view
matter,
by
going
deeply
of what
operations
losing count
written primarily
are
for lajmien,
find
supplement
their knowledge
of the
subject.
must
One
thing
more
be
considered.
serious
PREFACE
exposition of the theory
on
vii
to lay stress
very revolutionary
Einstein's theory
theory,
is from
must
of principle and
but
it
indicate
a
how
very
nonrevolutionary
appears
from
The
to those
we
have to do in
4aily and
in technical
slightly from
can
theories.
fuU
fication justi
for aU
practical
The
astronomer,
to calculate
continue
man
according
will go
on
to the Newtonian
theory, the
of science
little will be
using Maxwdl's
But
equations, and
foimdations been
the mental
have
of the complete
This
as on
of physics elucidated
entirely changed.
examples,
so
by
numerical
to
dispel wrong
and
fandful
the
J. H.
August 1921
T.
The
Translator
offers her
sincere thanks
to Dr.
R.
W.
Lawson,
of Sheffield University,
revision.
never
Without
have
been issued.
PREFACE
TO
SECOND
EDITION
No
Notes
alterations
have
been
made
in the
general from
to, and
book.
I
In particular, in connection
have
with
Chapter
has
XVII,
often
dealt
with
an
which objection
been
raised but
never
fully
of velocities greater of
fixed stars, I
am
indebted
interest in this
littlevolume criticism.
J. H.
November
1922
T.
viil
CONTENTS
PAGE
Introduction
xi
PART
THE
CHAP.
I
OF RELATIVITY
SPECIAL
THEORY
I. Preliminary
OF
Formulation Relativity
:
of
the
Principle
for
.
.
Its
Validity
.
Mechanical II. On
the
Processes
of
Nature
Light
of
,7
for
III. Is
the
Principle
Relativity
.
valid
.
Phenomena?
the
.20
of
Constancy
.
.
of
the
.
Velocity
.
Light
Conflict
31 V.
The
between
the
.
.
Two
mental Funda.
.36
Concept
of
of
Simultaneity
40
Special Total
OF
Theory
THE
Relativity:
from
.
A
the
.
SumTwo
Deductions Principles
Fundamental
45
VIII. The
Apparent
Absurdity
.
of
...
these
clusions Con
55
THE
CHAP
IDEAS
OF
EINSTEIN'S
THEORY
PACE
IX.
The
Union
of
Space
and
.
Time
.
; The
.
Min.
KOWSKIWORLD X. XI.
Numerical Further
6o
Considerations
Conclusions
and
.70
their

mental Experi.
Verification
79
PART
THE
II
OF
RELATIVITY
GENERAL
THEORY
On
The
Inertia
and
Gravitation.
93
EquivalenceHypothesis
of
Curvature
tational
Rays Field
of
Light
in
Gravi 104
XV.
The The
Relativity
of
Rotatory
Motion
109
and of
XVI.
Notion
of
SpaceCurvature
.
New
Theory
from
of
Gravitation
General Finiteness
Deductions
the
Theory
The
Hypothesis
of
....
the
of
Universe
Concluding Supplementary Illustrative
157
.
Remarks Notes
163.
i6s
168
Chart
INTRODUCTION
THE
led to
on
Theory
of
Relativity taking
is
its
branch
of
theoretical
in
physics,
origin
purely
of and
a
physical
universal
experiments. philosophical
conclusions
and
nature
"
space
time
Just as
to
it is
matter
many
more
than
and
scene
of human
extensive
plain,
so
comparatively
small
besides
our
that
usual
conceptions
and
time
are,
in the
main,
erroneous,
although
The the
they
approximate chapters
are
very closdy
intended
to
to the reality.
following
show
how,
on
basis of
of Relativity
was
developed
in two
stages.
first of these It
was
Special Theory
year
lativit of Reby
in the
1905
the
had
German
physicist Albert
Einstein,
after the
way
xii
been
THE
IDEAS
OF
EINSTEIN'S
by
the
THEORY
phjreidst H.
mathematician
prepared
notably
Dutch
A.
theory
and
necessity from
have
so
physical
experience,
magnificently
withstood
of the any
most
that hardly
doubt
In
be
entertained
as
to
the
years 19071915
Einstein
time
the
one.
old
Newtonian
new
only does
an
Einstein's
not
deficienc
of
philosophical
and
such
as
were
apparent
in Newton's
practical applications
astronomy almost
in the
realms
of physics and
are,
in general,
identical with
restdting from be
so,
the
old
theory.
are
Of
to
course
this must
because
found
two
be in accord
with experience.
in which
only
astronomical
phenomena theory in
Einstein's lead
theory
to
and
the Newtonian
of gravitation
cases
both
observation
decides
Einstein.
Nevertheless, in the
Theory of Relativity
as
claim the
same
the Special
Theory
of Relativity.
deficient,it will
remain
INTRODUCTION
a
xiii
only
the its
masterstroke
to
arouse
of genius.
serve
within
not
us
was
built up
with
of
mind
which
"
in proposing
necessary
new
paradoxical physical
ideas,
^itis simply
result of
experience,
followed
genesis of the
as
theory of relativity
the last few
to Kght
can
described
follows.
Within
decades
two
progressive
facts
with
almost
the principle of
of the
relativity
and
principle of the
Now these two
"
constancy
velocity of light.
to
principles appeared
was
be
mutually
^if one
versa.
right the
other must
all physical
anon
be wrong,
In spite of this,
experiments
experience
so
led
was
ever
and
to
principles,
as
that
one
apt to
It he
regard
was
little short
came
of miraculous.
the
rescue,
then
"
that
Einstein
cannot
as
to
when
stated
We
doubt
as
we can
in question, in
our
far
nor
trust
the evidence
found
of
senses
at
all ;
any
fault be
with
that proves
the antagonism
principles.
that proof
But there
in the considerations
are
concerning
the absoluteness
and
of
xiv
our
THE
notions
roEAS
of time up
OF
and
EINSTEIN'S
space, which present
THEORY
appear nobody
to
us
so
to the A
more
has
ever
suppositions,
to
shows
that
that they
they
are
only
not
appear
absolute
be
conceptual
Furthermore,
fication of these
two
aforementioned
discovery
to
This Einstein
sense
proved
decisive step
and
induced
reverse
pursue
in the
and
consequently
simultaneous The
derive
conclusions
arising
from
the
prindples.
sumtotal
is
of Relativity, and
this will be
THE
IDEAS
OF
EINSTEIN'S
THEORY
THE
IDEAS
OF
EINSTEIN'S
THEORY
PART
THE
I
OF
SPECIAL
THEORY
RELATIVITY
CHAPTER
PRELIMINARY
OF MECHANICAL
FORMULATION
RELATIVITY PROCESSES
:
OF ITS
THE
CIPLE PRINFOR
VALIDITY
THE
Let
us
of Relativity bears this name because it deals with the relativity of a special kind of motion, i.e. uniform rectilinearmotion. illustrate this idea clearly and without ambiguity Special Theory
If a ship is sailing smoothly by the following example. before the wind in still water, with direct course and rolling and pitching, we velocity, and without say that it carries out a uniform rectilinear motion. in the As this kind of motion is of great importance and for all, special theory of relativity, we shall say once
constant
rectilinear motion Where is referred to. any other kind of motion is being considered, for instance curvilinear motion, it will be
I
for the sake of brevity, ^that when in the first part of this book, uniform
"
"
we
talk of motion
SPECIAL
stated.
THEORY
We
OF
shall put
"
RELATIVITY
forward the
expressly
statement
following
"
concerning this particular uniform rectilinear special principle of relativity motion, and call it the It is evident that we can only speak in its simplest form. bodies, we cannot attach the mutual relative motion
of
of
"
any meaning to absolute motion because it cannot be verified. Given any number of observations or measurements to without reference the made within a closed system (i.e.
we surroundings) are
or
not the
mean
Let
as
us
suppose
an
ocean
to suffer no
deviation
so
f joltingrom
the engines,
that
is really uniform its motion and rectilinear. Are we then in a position to assert that it is in motion, unless look out of the portholes and watch the passing we to negative ? Our experience teaches us this waves question, for in the ship's interior all phenomena would One could play take place justas if it lay in harbour.
a
game
as
same
j of billiards in a ship that does not jolt,ustthe land. Even the most on sensitive mechanical
"
experiments
would
turn
pendulum
same as we
"
university
laboratory.
when
of the
of the ship by
the
with more exactitude and caution, we perceive that there is a rdative motion between the ship and the ocean. To our make meaning clearer, we shall vary our
example
a
to speak
Uttle.
One
another
above, way in
See Supplementary
pp. 165166.
PRINCIPLE
OF
RELATIVITY
we
stand
we
on
the deck
of one ship and look at the other, intervening distance between the two
and that they are moving with respect to each other. We cannot discern more than this by looking at one ship from the other, nor determine by mutual can we servati obour ship is at rest and the other whether
is moving and the vice versa ^whether ours and say : other at rest. Perhaps the reader will object It is surely reasonable to state that the ship at anchor is at rest, whilst the one is in motion, steam under moving,
or
"
speaking, the reader's But we must not forget that when surmise we say the ship lying at anchor is at rest, we are using and more exact stateonly an abbreviation for another ment to : The ship at anchor does not move relatively the and
not
Practically
earth is not absolutely at it rotates its axis once on rest, because every day, and describes an orbit of about 150,000,000 km. radius From in the course our more of a year. cautious viewpoint thus see that the ship at anchor is not actually we earth. We
that the
at rest, and
our
know
that
we
observations
more
the relative motion of both to the other ; but we cannot experiments (" Move surroundings.
or
not
by
"
sense
already form
mentioned.)
of expression
or we
made without reference to the is used here in the restricted For this reason, our breviat abis subsequently a solidly founded know
justified.
building
We
on
it is not motion.
whether
SPECIAL
THEORY
OF
RELATIVITY
the system is absolutely at rest or not, because its motion has no influence on the course of phenomena and experiments taking place within it.
review more critically this Eissertion, as it We constitutes the pith of the problem of rdativity. are talking only of the relamust again recall that we tivity As before and rectilinear motion. of uniform
us
Let
of the earth consists of the daily rotatory motion and the yearly revolution round the We can look upon the latter movement sun. approximately as to the uniform rectilinear motion, owing
mentioned, large radius of curvature of the orbit, but this does not apply to the daily rotatory motion ; hence we cannot that the existence of the motion of the earth maintain
the motion
is not felt in a terrestrial laboratory. On the contrary, daily rotation does influence the progress of physical bestknown The processes. experiment of this kind
is Foucault'g pendulum the daily motion experiment, of the earth
which
without
the {i.e. sun and reference to the surroundings From our past experience, however, an analogous stars). determination of the yearly revolution is impossible.
been cavedwellers who had never the light of day, but that they had attained in seen degree of culture equivalent a caverns subterranean
men
Supposing
had
to that
With the help of modem of the present day. have been able to discover physical apparatus they would the earth's daily rotation, and its angular velocity
well ; but the yearly revolution have entirely escaped sun would
they
would have been mightily to the earth's surface at some that motion
discovered
by
means
of
PRINCIPLE
astronomical
OF
RELATIVITY
We can only maintain our observations. does not influence assertion that the earth's motion in a restricted sense, laboratory experiments admitting refer only to the rectilinear yearly component Furthermore, we must conscientiously of the motion. true that all laboratory experiinquire if it is indeed ments
that
we
What
to
our
by this particular motion. iminfluenced can we state in the first place is this : According do not feel the trivial personal experience we
are
existence
at
of that
motion in any
on
our
own
on no
bodies,
the
means
nor
is it of
proof
for
of course, fundamentally
a
results is by
so
phenomena
a
sufficient
"
important
matter
^for
delicate physical apparatus can we facts that entirely evade our personal ascertain many his cigar The passenger who is smoking observation. is no more in the smokeroom ocean steamer of an
with
the help of
conscious
of the
waves
actually uniform motion. ship's motion, provided it were Yet the telegraphist on the upper deck can verify the existence of these waves without difiiculty,and receive
the message
might could
with
perhaps be constructed
and
the help of his receiver. The reader that some suppose sensitive apparatus
to
show
whether
ship
moves
uniformly,
at what
rate, without
reference
to the
without coming into contact with them), surroundings (or the log, which is used in practice as for instance with
for measuring the speed of a ship. that we Such questions imply
doubt
the
general
The problem validity of the principle of relativity. is therefore as follows : Is the principle of relativity in truth a general principle of nature, having strict validity
SPECIAL
THEORY
OF
RELATIVITY
for all physical processes, or is it merely a rule gained from experience, which tells us that within the limits of our restricted senseperception the detection of uniform
rectilinear motion ? to the surroundings
is impossible
without
reference
the basis of question can only be decided on detailed theoretical and experimental physical research. We for certain reasons shall divide the question
"
This
"
into two,
and
ask
first: Does
hold
vaUd
and all natural physical phenomena, the firstpart of the question we thus have to deal with
we
decide by
operations take
pendulum
experiments,
weighings, measurements
place within a closed system, whether or not that system is moving ? Both theory answer and experiment this question in the negative ; the principle of relativity is thus vahd
for these
are same
laws of mechanics
that they
moving
are
the
The fundamental phenomena. essentially built up in such a way for processes within a uniformly
Thus no effect system and in a system at rest. can take place theoretically arising from such motion have taken place during our can and never experience
in this domain, Hence
we
may
centuries.
assurance
for mechanical
CHAPTER
II
ON
THE
NATURE
OF
LIGHT
WE
The other
found
theory
processes ;
telling
us
that the principle of relativity is valid and that doubtful about it. there is nothing contradictory nor
matter
is very
different, however,
physical phenomena,
accessible Here sensitive and exact measurements. find the point of conflict that gave rise to the origin
of the theory of relativity. Theoretical considerations seemed to indicate that the principle of relativity can have no validity for optical phenomena, ment whereas experi
teaches
us
that it has.
In order to show
how
it is
that theoretical optics leads to such an assertion, we must on the nature of Ught, which insert a short preliminary be omitted by those readers who consider they need can
no
further teaching
on
the nature
of dectromagnetic
oscillations. We aU learnt at school that rays of Hght are waves know more about it will add electric and those who length {i.e.bout half a waves a of very short wave
"
"
thousandth
of
miUimetre).
but
are
That
8 been
SPECIAL
used
to
THEORY
OF
RELATIVITY
waves) are
from
life {i.e. in everyday water call waves length, very different, apart from their wave
waves on
sound
we see
and
luminous
waves.
Let
us
describe
we see
observing water waves. at a small portion of the surface of surface regularly rising and falling. what
When
a
look the
lake,
we
find repeated oscillatory motions at regular intervals for a particular is periodic with time. place, and we say the phenomenon
We
again for a moment, eyes, open them shut our thereby observing the whole surface of the water, and do not perceive that the we more, then dose them once If
we
surface is moving,
but
crests
we
see
i.e.
there
are
wave
each
the eyes
other
at
phenomena
and gaze
We equal say, therefore, is spatially periodic. If we open our freely, we see the at the whole scene
spaceperiodicity, which gives their own peculiar character : advancing,
whereas
we are
succeeding
of
waves
these
fully
waves aware
are
apparently
every
that
single water
a vertical oscillations about Up to the present we have spoken of the surface of we water, because what clearly recognise to be wave
is the motion But we must motion of the surface. not forget that in reality the water particles under the surface and the air particles above it participate in the
movement.
When habit
in the
speak of imagining
we
of
a
wave
process
we
are
visible
of water waves phenomenon advancing along a surface, whereas, in order to apply these ideas to phjreics, we by imagining a train our must mind process which We shall not find occupies three dimensions in space.
it difficult to do
so
let
us
imagine
large number
ON
of
THE
NATURE
balls fredy
OF
LIGHT
the under suspended surface of the water, of tiny and likewise a number balloons floating in the air above it,so that both balloons
tiay
luminous
balls participate in the motion of their respective What then see, on following the oscillawe media. tions
and
visually, is motion,
the
true
threedimensional
wave
similar to that which plays so great a part in The oscillations of the water physics. particles, and in it, take place vertically, those of the balls suspended the propagation of these oscillations, i.e. whereas
the
transmission
direction
of the
waves,
proceeds
horizontally.
itself,and
of
The
the
of the
oscillatory motion
to of oscillations, are perpendicular propagation Oscillations of this kind are called transversal. each other. But besides these, there are other oscillations distinguished from : these are called longitudinal ones
waves
transversal
by
motion
can
the
fact that
the
directions of
and
hung at of small weights or lead balls are number equal intervals from a long indiarubber tube, fastened
end downwards
at
one
to the ceiling. Pull the lowest ball a little and then let it go ; it will oscillate vertically,
propagated upwards along the tube in a vertical direction, and setting aU the other weights into vertical oscillations. The direction of the the direction of their and oscillations themselves, propagation, to deal with parallel to each other ; hence we have It is a wellknown longitudinal waves. fact that sound waves are nothing else than longitudinal
are
the motion
being
waves
transmitted
The
in air,
wave
solid medium.
one
wavecrest
to the next.
The
wave
length of sound
10
SPECIAL
THEORY
OF
RELATIVITY
varies according to the pitch of the note ; for example^ it amounts to about 130 cm. for the musical tone C, and is smaller for higher and greater for lower notes. Soimd waves have one waves and water characteristic in common, viz. they consist of the real motion of a tangible and ponderable (air, water, substance This cannot
are
etc.).
be the
propagated
with Hght, because rays of light interstdlar space from through (free
case
ponderable
created
matter) as
vacuum.
experimental
as
an artificially well as through Nevertheless, about a century ago, research led to the conclusion that light,
sound, is intimately connected tions. with oscillaFurther data were soon obtained on the nature discovered that they of luminous oscillations.It was transversal oscillations, their velocity of promust pagation be being about one milhon times greater than that We shall of sound in air, i.e. 300,000 km. per second. well
as
all by the letter c. On length is very small, and wave is connected with the colour of the Ught in a similar length of sound is connected with the way as the wave It is greatest for red light, i.e.about pitch of a note.
denote
and
for
sequence : red, orange, yellow, green, blue, violet length, i.e." these last rays having the shortest wave The spectrum ^t^about of visible rays of xu.hns light corresponds exactly to an octave of tones. All this
"
Tir.ffnr
mni'.
a^d
it then
decreases
in
the
following
was
soon
known
excepted : What of Hght ? As we said before, it cannot be supposed to be a ponderable hypothetical unknown substance, and hence an introduced, was was termed something which Ughtaether or worldaether. All that was known with
" "
one
item
ON
THE
NATURE
OF
LIGHT
11
that it was reference to this aether was neither tangible nor that it caused no friction, but that it ponderable, be capable of performing very rapid were oscillations. These aether osdUations to be rays of light.
must
transversal
supposed
was ago an important advance made in the theory of light by Maxwell, his founded who later date, a theory electromagnetic of light. At definitely confirmed by the wellknown this was ments experi
Half
century
Hertz, and, on further investigation, of Hdnrich ing it led to the invention of wireless telegraphy. Accordto Maxwdl, rajrs of light belong to an extensive oscillations, which includes species of electromagnetic
the the
waves
of wireless telegraphy
and
active ultraviolet
are
only
distinguished from
wave
forms the
length
characteristicof them.
a
To make
useful and frequently adopted how Fig. I, showing the different kinds
distributed are over the waveradiations lengths. Here we note two gaps entire scale of wave in the extensive spectrum of electromagnetic oscillations,
which
Thus
far
we
belong to varieties of rays not yet detected. have out the coordination pointed of lengths with various kinds of rays. What
is that ideas of electromagnetic fuUy explained. is it What
our
yet ? Not that oscillates ; is it electricity or magnetism do not find electridty itself osdllating in exactly ; we find electric force and ^we wave an electromagnetic
not
"
1,000,000
cm.
100,000
cm.
Waves
of wireless
telegrajdiy
cm.
10,000
"1000
cm.
Hertzian
waves
produced
in laboratory
.
Heatrays
(ultrared)
o'oooi
cm.
Lightrays
....
Ultraviolet
oooooi
cm.
rays
o'oooooi
cm.
oooooooi
cm.
Xrays
"o'oooooooi cm.
o'ooooooooi
cm.
7rays
"o'oooooooooi cm.
Fig.
I.
ON
THE
NATURE
OF
LIGHT
13
magnetic force. Let us explain what we are to understand by this. Bodies charged with electricity attract or repel each other, according to the unlike or hke sign
of their charges. our minds, let us negative
of small
To
in the
pithbaUs,
metal
which
sphere
are
also charged
attract
with
electricity. Jhe
will
the
particles charged
repel those negatively charged ; an electric force the surroundings acts on of the sphere, and we say : the sphere charged with electricity produces an electric
and
field around
it.
The
force acting
on
small
particle
carrying a unit of electric charge, at a particular point of the field,is called the electric fieldintensity or electric force at that particular poiat of the field. The tude magniand
of
course,
direction of the electric field intensity varies, in various parts of the field. The direction
of the field intensity or electric force in the field of a negatively charged ball is always directed towards the diminishing with the centre of the ball, its magnitude ball which produces field is at rest, and if its charge suffers no the electric and direction of electric force change, the magnitude at one particular point of the field remain the same ;
distance from
the
ball.
If the
but
as
we
of the field. We
and
charged
electric field variable both with Let us suppose that the negatively
to
and
to
be
gradually recharged with positive electricity. Later is the charge of positive electricity to be again diminished
14
SPECIAL
reduced
to
THEORY
zero,
"
OF
^whereupon
RELATIVITY
the ball is again forth. What
and
with negative electricity,and so then with the ball's electric field ? When the charge of the ball is reduced to zero, the electric force will of course be zero, the charge and when
charged happens
of
the
ball becomes
the
direction
of
the
were
Thus, the field intensity in vice versa. distinct part of the field will gradually vary its magnitude and direction, and in this case we have to deal with
field which varies with position and time. Another here : the matter of importance should be mentioned ately effect of a change of charge is not transmitted immedia
contrary, it takes some little time to act at a distance. Hence, the when charge of the ball oscillates between positive and negative at a distinct moment values, and is reduced to zero
to
a on
distance ;
the
of time, will of
the
field intensity
at
very
great
distance
also be reduced to zero, though not until a if the changes in the charge short time later. Now of ball take place very quickly, the say, many miUion
course
"
that at a certain ^itwiU happen distance from the ball an electric field intensity corresponding to positive charge will be observed, whilst the ball itself is again charged negatively. At double the
times
a
in
second,
"
distance,
the
arising from
one
field intensity will of course be the one the negative charge the ball possessed
At three times the distance, the period earher. fieldintensity wiQ correspond to the last but one positive A moment charge possessed by the ball, and so on. later, when the ball has again changed its charge, the field intensities will again be reversed. Thus we see
ON
that
THE
NATURE
bears
OF
the
LIGHT
character
15
the
phenomenon
makes
motion, it as an
which
we
used electric field intensity, and electric valid for the analogous notions of field intensity, and magnetic magnetic
magnetic
waves.
also fidd,
can
We
repeat them word for word, only substituting magnetic for electric, and the words northmagnetism and southfor positive and magnetism negative electricity respectiv
of every such velocity of propagation for electric and magnetic action is also exactly the same fields, being 300,000 km. in the second ; i.e. when the
The
ball
its charge, the direction of the fieldintensity will be reversed at a distance of i m. after the threehtmdredmiUionth Now this is part of a second.
reverses
exactly
and
the
velocity
light
electromagnetic be said to be a
waves,
of light so that
and
are a
none
ray
of
temporally
spatially
fidd similar to the one variable electric and magnetic described This supposition has devdoped into above. It would of time. absolute certainty in the course take
us
too
far to enumerate
we
detailed For
we
the have
of the
ray
permits us to analyse exactly the magnetic field of a ray of light. We require which
tiny test
partides,
one
charged
with
electridty.
16
SPECIAL
THEORY
OF
and
a
RELATIVITY
help of which
part of
us,
a
we
can a
measure
contrivance with the the force acting on these less than provide
a
a
particles, within
space
of time
next
billionth
source
light,and
of
so
second. set it at
We
a
must
of
in front
that its rays strike us horizontally. We then place the tiny ball charged with electricity, and the tiny magnetic pole straight in front of us and alongside of each other, and we happens. We what observe shall
"
with
observe
our
tiny apparatus
that
forces
are
exerted
on
both
particles,
each other, and the ray of the ray of light. Thus, supposing horizontally, then the electric us of Ught to approach the particle will be pulled vertically upwards, and
particle horizontally from right to left. The action of the forces in these directions is only maintained for an inconceivably short time ; after the magnetic
to
billionth part of a second the action is wards, downreversed, the electric particle is then puUed and the magnetic particle from left to right.
thousand
all is again reversed, and so forth at the immense rate of about five hundred billions of oscillations In this way we describe the per second. may
The
next
moment
for a given point. To of this phenomenon we ascertain the space dependency, must imagine other particles to be available besides the first two, these
timeprocess
being placed at different points of the ray of Hght. We should then find that all such particles, which, together with the firstpair, are placed at an equal distance from the
source
i.e. aU
of light, will oscillate in the same phase, those charged with electricity will be pulled
"
simultaneously
upwards,
with
mag
ON
netism
we
THE
NATURE
to
OF
LIGHT
so
on.
17 But if
simultaneously
a
to the source pair of particles slightly nearer a the of light (at distance which we shall call l/znearer they wiU oscillate in the opposite phase to the source), the electric particle in our original pair, i.e. when
place
the electric particle pair is pulled upwards, standard pair is being pulled downwards, and so on. of the nearer to the source A third pair of particles, nearer of light by
oscillate like the first pair, etc. We / the wave length of the light. This call the magnitude varies for different colours ; the Umits for red and at the beginning of this chapter. violet were mentioned
an
amount
I, wUl
of the electric and magadd that the magnitude netic field intensities decreases with the distance from the source shall then have fully indicated of light, we
we on space and time, and qualitatively their dependency we shall have suf"dently described the inner mechanism be briefly recapituThe whole can lated of a ray of hght. transversal thus : rays of light are oscillations
If
of electric and magnetic force. it is quite out of the question for any Of course analysis of a ray of light in the way here indicated to be
actually carried out ; but there are a sufiicient number of indirect proofs, which point to the correctness of the For the physicist, these ideas ideas developed above.
thus have instance, transmitted degree of certainty as, almost the same that infectious diseases the hj^othesis
for
are
by bacteria has for the medical man. The preceding description of the mechanism of a ray of light teaches us that in the case of Hght, quite apart from the far greater velocity of propagation and the have to deal with an entirely length, we smaller wave different kind of oscillatory process from that of water
2
18
or
SPECIAL
soundwaves.
of
a
THEORY
Soundwaves body
OF
RELATIVITY
consist
of
the
movement
material
(air,ater, w
rock,
which etc.)
undergoes
the
case
In periodic changes. spatial and temporal and the of Ught, however, it is the magnetic force which periodically. Nothing
change
nor
concrete
abstract, a As we are
vacuo
"
oscillates in a ray of Ught : it is force varying periodically in space quite able to imagine the existence
of force in
e.g.the gravitational action of the sun planetary space to the remotest reaches through empty orbit and beyond, and the earth attracts a body in an evacuated vessel
"
of the readily conceive We thus see existence of a variable force in vacuo. duction are that we able entirely to dispense with the introso
we
can
of the of oscillations of light. aether, as the transmitter Formerly thus : Rays ran the arguments of Hght be oscillations, consequently to been have proved
of
a
hypothetical
substance,
i.e. that
must exist to carry out these oscillations, something to oscillate. That for we cannot expect nothing call something which oscillates in hght, we aether."
" "
"
One
thing,
however,
had
was
been
unknown
concrete
something
substance. assume that
out
not
if
we
oscillations
(periodic variations
electric
and field intensity. Hence we need not talk of magnetic aether, the idea of an electromagnetic field takes its
"
intensity),iz. the v
of the
place. In spite of this, the word has been retained aether in the terminology of modern phs^ics ; it designates field the very essence of the electric and magnetic
" "
ON
magnitudes. brevity and
THE
We
NATURE
propose,
OF
LIGHT
sake
19
of
use simplicity, to make of the terms aether and aether oscillations in all that follows ; the reader will know what is meant after what has already been said.
It should be emphasised that the discussion on the nature of light in this chapter has nothing to do with the fundamental principles of the theory of relativity. It merely the purpose developments more
serves
of making readily
the subsequent
however,
idea
as
fundamental is
so
that
important,
that anyone
waves something
of the theory
some
of relativity that
have
comparativel
that
has
been
allotted
our
to
not subject
immediately
connected
with
theme.
CHAPTER
III
IS THE
PRINCIPLE
OPTICAL
OF
RELATIVITY PHENOMENA
?
VALID
FOR
EQUIPPED
following problem
:
with the definite ideas on the nature of light set forth in the preceding chapter, we are now the solution of the able to approach
Is it not
possible to detect the ments annuctl revolution of the earth, by laboratory experiwith the phenomena connected of the propagation of light ? On
the classical theory, according to which the aether is to be regarded as a real substance, we should expect
be successful. This is must experiment imagine an analogous easily understood, if we ment expericarried out in the sphere of acoustics.^ Let us board an ocean imagine ourselves once more on steamer that it travels without rolling so perfectly constructed
that such
an
and pitching, and with straight course uniform there is a whinjsical rich shall suppose velocity. We I'U bet you 10,000 old gentleman on board, who says : dollars that none of you can satisfactorily demonstrate
or
"
described in the follovdng could not actually experiment by human be performed observers, because of the infinitesimal involved. It might perhaps be realised with the time difference help of an automatic like those in soundmeasuring apparatus, by the war the English during But this has use artillery.
'
The
nothing
to do with
OPTICAL
that this ship is moving, Thereupon
made
to win
PHENOMENA
without
all
sorts
21
in ^but all experiments made the ship's saloons turn out exactly as they would on dry land. Then somebody than gumption with more
the wager,
"
the
has
an
idea, and
"
exclaims
We'll
do
it
"
stem,
two.
and The
passengers are chosen and taken on is stationed at the bow, a second at the between the other the third precisely midway with a pistol, The other two
Three
rate, going at the same provided with stopwatches to diately and have strict injunctions stop the watches immeThe man in the middle hearing the report. on fires his pistol, the other two stop their watches, and a
of the two
was
watches
very
shows
that
stopped
The
at the stem.
atmospheric air that transmits the sound does not take part in the ship's motion, hence a waves current of air from bow to stem is present on board the to travel at a quicker the sound waves ship. This causes
rate
towards
the
stem
than
towards noted.
the The
bow,
man
"
and
the
difference in time
now
responsib
noted between between them. fired midway The difference is a proof of the existence of motion, and as found without considering the surroundings, I the whimsical old gentleman Not so ! That is not adds : laid. I admit you haven't
"
If the explains : difference in time would the two watches, because the
The
the
way
my
wager
was
22
SPECIAL
THEORY
the
OF
RELATIVITY
mine the ship and the surroimding air. All you could deterthe relative motion between the two, and your was have furnished the same result had would experiment
at
anchor,
and
wind
proves Let us
had
blown
from
Your
experiment
I shall keep my money." settle their dispute between attempting prove the
If
we
them,
and
an
experiment
ourselves,
earth's annual
a
revolution.
through
space,
ocean
as
worldship travelling fiUing aether is the medium to the air surrounding the
vast terms
by
replace ship by earth's surface, air by aether, luminous We waves. then may
sent
out
and
from
the
earth's
forwards at a slower rate will be transmitted in the direction of the earth's motion than in the opposite
surface
direction.
Just
as
current
of air sweeps
the
an
deck
of the moving ship from bow to stem, so must drift be present at the earth's surface in a
aetherdirection
opposed
through
to the
as
the sphere gUdes discuss the possibility the experiment, it will be of the
success or
failure
Let
us
assume
i.e. that there exists a slower rate of light in the direction of the earth's
the
a
opposite
new
motion,
direction.
That
OPTICAL
plished by
means
PHENOMENA
a
23
of
laboratory
and stars. astronomical considerations Has, then, the principle of relativity, as it was discussed in the first chapter, been violated ? We stated there : It is impossible to perceive the existence of motion If we take up without reference to the surroundings."
"
and
out with
the
point
of view
we can
his wager,
experiment
fellow
case
laying
"
This
tween only to the interaction bethe apparatus experiment, and the used for our finally arrived at aether ; the result we surrounding between was merely the existence of a relative motion
reference
had
The more. aether, and nothing principle We see that of relativity has thus not been violated." to a dispute of words, in the end the question amounts the meaning of the restriction "without concerning earth
and reference to the surrormdings." confusion
shall now of words, we form, further no containing relativity into another in which experiment ambiguities, and one wiU clearly For this purpose we shall decide one way or the other.
introduce
new
useful later
"
on.
a
as
itself notion, which will also prove in Chapter I that such stateWe saw ments " " in motion body a body or at rest
"
need
as
an
addition to complete
them,
namely,
or what this state of motion To describe the position or motion body is always required (orat least
to
which can be referred system of lines in the universe), to in stating distances or velocities. In general it is
the material
structure
on
which
executing be run on
our
measurements.
a
we
are
board
ship, and
we
24
SPECIAL
a
THEORY
lo
m.
OF
per
RELATIVITY
second,
we
attained
as
a
velocity of
understand
that this velocity is relative of course to the ship's deck. (The velocity of the runner with be lo m. reference to the earth would per second,
matter
as minus the velocity of the ship, according runner travels with or contrary to the direction The body to which the motion of the or
ship.)
body
of reference
or
system
the earth experiments, itself is our body of reference ; for the sound ment experithe body mentioned above, the moving ship was Thus we may state the principle of reference, and so on.
of relativity in the following form
:
laboratory
In
different systems
rectilinearly with to take place in respect each other, all natural processes exactly the same way. principle of relativity, as here stated, would certainly be violated if it could be proved that rays of light on the earth are propagated more quickly in one direction than in another. partaking Measured from
a
moving of reference
uniformly and
The
system
of reference not
result would certainly not processes in that system of reference would take place in a different way from that occurring on the earth, a have just we result contradictory to the statement made.
"
furthermore, that the question of the vaUdity or nonvalidity of the principle of relativity, in its modified
see,
We
form, includes the problem of the existence of an aether. For. if we say the aether is a real substance, we must have the possibility of perceiving it directly or indirectly,
even
though
it need
not
be
tangible
or
ponderable. of per
Now
way
capable
OPTICAL
ception,
we
PHENOMENA
25
expect that a system of reference be distinguished relatively to which it is at rest must from others relatively to which it is in motion. That, however, be a contradiction would of the principle
must
of relativity.
If this is
so,
then
substantial aether is of no use to us. II that the aetherhypothesis can be entirely dispensed of lightprocesses ; indeed, with for the comprehension
if the
principle of relativity has general validity, the becomes not aether conception only superfluous, but directly as a hindrance. We then use the acts may
aether solely in its abstract meaning, mentioned word II. On the other hand, the at the end of Chapter of the existence of an aether as a real subassumption stance led us to expect that an experiment of the kind give a positive result {i.e. permit us would to discover an influence of the earth's motion on the if propagation of light),we could only succeed in accomplishing Before
at, it
described
the experiment with sufficientlyexact apparatus. final experimental decision had been arrived a
was
only
possible to set up
conjecture concerning
the general validity of the principle of relativity in its led the have latter form. These conjecturesmay
majority
of
physicists
to
assume
that
the
principle
mena, phenocertainly valid for mechanical for optical electro magnetic and

After these preliminary considerations, we now must turn to the question of the actual performance of the of light with the propagation A simple calculation shows that at the earth's surface. described Hteral analogue a of the sound experiment
experiment,
concerned
above
cannot
produce
any
26
SPECIAL
THEORY
OF
a
RELATIVITY
million times greater
case
than
that of sound.^
must,
ground of experimenting the greatest possible length. Let us station the middle is to give the lightsignal, on a experimentalist B, who
of
course,
We
chose
an
is placed in from the direction of the earth's motion 30 km. away the lull, and the other observer C is situated in the Both observers opposite direction, also 30 km. from B. hill visible from all sides.
The
observer
have
accurate
which them
place
rate, going at the same chronometers they stop the very instant the lightsignal reaches
from
B.
To
ensure
accuracy
we
might,
of
a
in the
of observers,
imagine
contrivances
highly
sensitive nature
register the exact arrival of the ray of light automatically, and with Ughtning rapidity. Will a difference be noticed, then, between the two stopwhich clocks ? The
effect to be
the
sun
course,
can expected easily be calculated. The velocity of the earth's revolution around to 30 km. is, of amounts per second, which time the velocity of the hypothetical at the same
aetherdrift. We
sound
fluence should expect this aetherdrift to inrays of Ught in the same way as wind influences be propagated ; hence Ught would with a velocity
km.
per second
of 300,030
J5 to C
relatively to the earth's surface, and with a velocity of 299,970 km. per second in the direction from B to A.^
'
The
prove
more
consider of
a
be
so,
to
every because
here,
OPTICAL
The
distances BA require and BC
PHENOMENA
are
27
so
each 30 km.,
from
that light
to
should
o'oooioooi
seconds B to C.
The
A,
we
chronometers
a
our
estimate is so discouraging the possibility of that the reader may well doubt an actual experiment of this kind, and will perhaps be satisfied to take it for granted that our presentour
numerical
day
technical
means
are
not
in point. And yet problem Michelson the experiment, a and gained performed Later, this definite result more than thirty years ago. became in physics, one of the most famous experiments imon portant account of its constituting the most perhaps
empirical evidence in favour of the principle of relativity. The idea is as follows : A ray of light falls on a glassplate at an angle of 45", and is separated into two
of is propagated to the the glass and perpendicularly original direction, whereas the other part traverses the Both parts traverse glass plate and travels straight on.
parts ;
one
certain distances from their point of origin, and are then to their reflected in mirrors set up perpendicularly respective directions, so that they travel back along When the same path and meet again at the glass plate.
a
and
the
are observed parts reunite, certain optical phenomena of the parts lags called interference fringes, and if one behind by an infinitesimaUy small fraction of a second.
28
SPECIAL
THEORY
OF
RELATIVITY
this gives rise to shifts of these fringes, which are readily In this way, Michelson compared the time perceptible. taken by one ray of light in travelling to and fro parallel
to the direction of the
other earth's motion, with that of anray which travelled to and fro in the perpendicular direction. was (The experiment carried out in the
floated in was way : The entire apparatus free from vibration, and capable mercury, so that it was A thorough investigation was then made, of rotation.
following
to decide
any shifts of the interference fringes brought into various occurred, when the apparatus was positions relative to the direction of the earth's motion.) The The were results obtained absolutely negative.
whether
attempts
exact
were
and
in subsequent years with sensitive apparatus, and finally, Morley repeated contrivances
even so
more
and
accurately
to
a
adjusted,
part
that
an
effect amounting
hundredth
have been detectable. must value of the computed But there was not the slightest trace of a nonuniformity in the propagation the Michelsonof light. Hence
decided with certainty in favour of experiment the validity of the principle of relativity for optical ones. processes, as well as for mechanical
It would concerning depended
not, of
course, a
Morley
such
on one
decision problem
experiment
conscientious its execution. incidental accessory circumstance, of that some body which nohad thought, had paralysed the effect of the earth's experiment, and that the existence motion in this one of this motion might stillbe proved in some other way by laboratory experiments. To begin with, it might for instance have been possible to suppose
the moving
earth
OPTICAL
PHENOMENA
29
to carry sether along with it, justas a body moving in liquid carries along liquid particles on a its surface there would be only owing to friction. In this case
very small relative velocity between sether and earth at the earth's surface, so that the negative result of the Michdson experiment could readily be understood,
a
without regarding it as decisive for the validity of so important The possiand farreaching a law of nature. bility of such an explanation constituted the theme of
close investigation, but the results obtained conilict with other facts of experience, so
come
into
that
the
that the earth cairries sether along with it assumption A number be discarded. must of other experiments quite different from that of Michelson and Morley were devised and carried into
execution,
but
light, but
were
with
other electromagnetic
of the most of portant im
took their origin in chemistry tion The science of the elements, which forms the foundafrom the unsuccessful attempts of chemistry, arose lead and other common of the alchemists to convert metals
into gold, and the law of the conservation of fruitless efforts to effect perpetual energy originated in In the same way, Einstein, as a consequence of motion. the negative results of the aforementioned experiments,
is not conclusion : The dilemma nor can the due to want of sldllon the part of physicists, be insufficient development of our technical knowledge
came
to the following
at
fault.
The
fundamental
cause
30
SPECIAL
THEORY
of
OF
RELATIVITY
by laboratory
absolute
impossibility
determining
the influence of the rectilinear component of the earth's motion in any physical phenomena soever, whatbecause the principle of relativity is valid for all
experiments
natural processes, and not only for those connected formerly supposed. mechanics, as was
with
CHAPTER
IV
THE
LAW
OF
THE
CONSTANCY
OF
OF
THE
VELOCITY
discovery
LIGHT
THE
that the special principle of rdati\'ity is valid for the entire range of physics is in itself far more satisfactory than our former
according
to which
it is vaJid only for one part of ph3reics, namely, mechanics, and not for any We are thus led to ask : Why, then, were we other. of be valid for optical processes ? opinion that it cannot assumption,
aether hypothesis was responsible knew for this. We with certainty that rays of light deduced hence are the oscillatory processes, and Without
doubt
the
erroneous
conclusion and
"
that
there
must
be
something
substantial to carry out these oscillations, being the aether. As long as we this something accept the notion of an aether, the analogy with the board on experiment ship would necessarily sound
concrete
"
the validity of If we the principle of relativity for optical processes. accept the principle of relativity (and that we must, if
appeal
us,
to
and
thus lead
us
to doubt
we
are
to understand
mentioned
the negative result of all the experiments in the last we must give up
chapter),
the hj^othesis
rajre of Ught
of to be
and
magnetic
substantial aether, and consider no than oscillations of electric more II contains a field intensity. (Chapter
32
SPECIAL
THEORY
OF
RELATIVITY
detailed description of what this means.) To complete the description of this process, we must state the and this again of the waves, velocity of propagation indicate the system has no meaning, unless we of the hght is propagated reference relative to which have we should with that velocity. Formerly said :
"
an
analogous
way,
as
sound
330 m. velocity of approximately the aether, in relativdy to the air. Now sense of the word, has lost its significance,
with
longer refer to it when giving a precise devise statement must of the velocity of light. We a more our of reference, for which suitable system
hence
we
can
no
statement
The
of the velocity of light shall be valid. taught and Morley experiment of Michelson
us
that rays of Ught from a terrestrial source, which takes in all directions part in the earth's motion, are propagated with equal velocity. We might, therefore, be inclined to say : Waves of light are "transmitted with a from the source definite velocity as measured of light. the present statement The difference between and the
light has a distinct according to which be made clear by velocity relative to the aether, may both alternatives once more with our example comparing of the propagation of sound along the deck of a moving
former
one,
ship.
Sound
waves
have
distinct velocity
;
of propagation
are
relatively to
the atmosphere
they
not,
therefore, transmitted uniformly in all directions along deck on board ship, but travel more the moving quickly than towards the bow the stem towards of the ship. is not the same in velocity of propagation every direction relative to that sjretem of reference (the deck), of sound. which is at rest relatively to the source
Thus
the
LAW
We find
an
OF
analogue
VELOCITY
OF
LIGHT
88
the correctness of which would have required a If, on the positive result for the Michelson experiment. board ship had been on the experiment other hand,
carried out and had
not
in such a way that the velocity of the bullet from the revolver the velocity of the sound
the experimentalist, who measured, whereby is stationed in the middle, firesboth forwards and backwards,
then
been
between
the
have
been
noticed
(Neglecting
move
with
of reference at This kind of prorest with respect to the marksman. pagation to the last proposed hypothesis corresponds light, with the difference, of course, that concerning process, and are not, in rays of fight represent a wave
use material fike a bullet ; the simile we characterises only the kind of propagation, but not the The hypothesis that light is nature of the process.
themselves,
transmitted system
source
distinct velocity with respect to a of reference, which is at rest relatively to the by the Swiss physicist of fight, was advanced
with
advantage of his theory Hes in the fact that harmony it is in complete with the principle of relativity. According to this hypothesis, rays of fight from
Ritz.
The
velocity
always travel with a definite wouM from it, independently the of whether It would then be
the existence of the earth's motion by experimen on the propagation a of lightrays from be for observers on board as it would terrestrial source,
not.
justas
impossible
the existence of the ship's motion ship to determine the velocity of bullets fired from the by measuring
3
84
SPECIAL
THEORY
The
OF
RELATIVITY
deck
negative result of Michelson's experiment would thus be easily imderstood. The Ritz theory, however, leads us to another conclusion, one and hence which conflicts with experience, of the ship.
"
"
For to give up this hjrpothesis also. obUged to our more purposes of illustration, let us revert once it is shall suppose ship. We simile of the moving the coast, and parallel to it. Bullets fired sailing near
we are
the middle of the ship will travel fore and aft with But let us suppose equal speed for observers on the ship. of observers posted on land, and provided with means from
some
kind
fired from
the
projectiles projectiles
direction will travel more quickly If we than those fired in the opposite direction. call the velocity of the bullet relative to the gun q, and the will velocity of the ship v, the bullets fired forwards have the velocity q\v, and those fired in the opposite Let us apply direction qv, for the observer on land.
According to the Ritz optical phenomena. theory, the velocity of rays of light from a star which is from as the earth must, the measured approaching earth, be greater than that from a star which is moving
this to
This deduction has been tested, both from us. away for rays of light from radially moving for stars, and terrestrial sources those emitted by moving of light.
of the velocity of light on the state of of hght could be detected, and of the source motion be maintained. On therefore the Ritz theory cannot No dependency fact based on the other hand, a new and important experience has been obtained : the velocity of light in has always the value c vacuo 300,000 km. per second,
=
and
is quite
independentof
the state
of
motion
of
the
LAW
source
"
OF
VELOCITY
Einstein
OF
designates
LIGHT
this law
as
35
of
light.^
the
* Principle of the Constancy of the Velocity of Light," and regards it as a fundamental principle of nature, to the Principle of Relativity. equal in importance
Together,
tion principles constitute the foundapillars of the Special Theory of Relativity. It is important to notice that these two fundamental
these two
principles rest
to exact
sensitive
trust
the safest and surest ground known by the most science ; they are supported the most exact and optical experiments
on
astronomical
our own
measurements.
If
we
we
can
in
any
way
experience,
must
confidence in the validity of these two point deserves special emphasis, because
have
shall
soon
have
*
occasion to doubt
to
our own
their correctness.
experience,
this
is valid when the is the earth, but owing to the validity of system of reference the principle of relativity it holds good also for all systems of in uniform to the earth. relatively rectilinear motion reference to complete On we : the description p. 32 said of this process, According
"
of propagation of the waves, indicate the system this again has no we unless meaning, and the light is propagated relative to which of reference with that We are now able to state that light is propagated velocity."
we
must
state
the
velocity
with
the
velocity
relatively
move
uniformly
and
the
rectilinearly with
CHAPTER
THE
CONFLICT
BETWEEN
THE
TWO
MENTAL FUNDA
PRINCIPLES
has been present everj^hing plain deduction from sailing. The of conclusions laws from our and of general experiments,
to
UP
the
in natural quite common something in greater practical science (moreover, branches of much importance then the present), and the of outsiders
experience,
is
majority
rarely trouble their minds in the least about such The fact which distinguishes the theory of matters. relativity and liftsit above the level of everyday experience
when we consider the matter find that it is quite impossible for both
:
is this
more
we
the two is fundamentally contradiction between the Ritz theory and the facts as that between the same
of experience.
Let
us
again
develop
it.
By
way
of
change choose
shall train,
which track.
straight
the
embankment.
36
the
TWO
FUNDAMENTAL
PRINCIPLES
87
principle of relativity, phjreical processes must take place in the moving in a train precisely as they would
train ; hence stationary measured rays of light, as by observers situated in the train, must be propagated both forwards and backwards with the same velocity,
just as
however,
they
According
train
were
stationary.
principles, from the
fundamental
as
the
because
motion
of the train and in the contrary direction, it ought to be independent of the state of
source
demands two of light. These contradict each other, for if the speed of the train be in the direction of the V and an effect be propagated in train's motion with the velocity c (measured the train),
of the
then the
we
must
find
(asmentioned
from
the in the
us
chapter)
to
velocity
01 cv common
c+v,
Sound and
we
measured a propagated
sense
be
teaches
it)can
velocities with which Let us imagine the roofs of railway in such a way to form as together Then
a
of the addition of velocities (as for all be tested and confirmed have to deal in everyday life. we carriages joined a bicycle track.
cyclist, according to the direction in which he traversed the entire length of the train, would naturally by an observer possess quite different velocities as judged
We can see no reason the embankment. why this should not hold for light also, and if we had not made up our minds at the end of the last chapter to trust the
on
two
fundamental
"
principles,
are so
we
should
be incHned
one
to
say : them
As
they be
obviously
contradictory,
of
must
wrong."
(From
the
point
of view
of
38
SPECIAL
THEORY
OF
both
our
RELATIVITY
be
logic, they
other hand,
must
wrong.)
Oii the
we
two
principles, then
and admit so that the apparently obvious analogy of the cyclist have on the roofs of the moving railway carriages can no possible validity for light. suppress
wonted
mode
of thought,
In order analyse
more
to
shall the closely the process of measuring lightsignal from the train and from the
see
how
this
can
be possible,
we
To
must
measure
the
station an in the middle who sends out the train, besides the man hghtsignal, and all three would have to be provided way, observers on the embankment stationed at certain intervals would have to be provided All these clocks with exact clocks. would have to be timed absolutely alike. At a given
with exact
clocks.
In the
same
moment,
one
say,
justas
the middle
of the observers stationed on lightsignal must be sent out, and all observers must stop their clocks the very instant' they see the flash. We could then
tion, calculate the velocity of Ught in each direcboth from the train and from the as measured by means of the differences in the times embankment,
distances. and the measured in Chapter III that We saw
exact
our
clocks
are
not
nearly
enough for a direct measurement of the velocity of light, and for the present experiment, conditions are On the one hand, the base unfavourable. stillmore of our observations is much smaller than in the experiment
the other hand, the velocity of times smaller than that of the the differences of time to be measured
TWO
are
FUNDAMENTAL
smaller
PRINCIPLES
were
89
much
than
have
they
in that
a
case,
so
that
more
our
clocks would
to be about
are,
billion times
us
exact
to enable
to carry out
the necessary
our are
this does not satisfy conscience in relation to the logical difftculties which involved. We cannot srffer two laws of nature to
the discrepancies
are
But
by
modem
technical
therefore, at the outset, take account of the possibility of our having clocks and measuringthe velocity of light with rods exact enough to measure
must,
We
experiment, and that is,that the observers at the ends of the train, as well the embankment, as those at different stations on shall be provided with clocks which not only go precisely,
our
but
which
are
and
this is
an
essential. Einstein
that
contradiction
between
the two
fundamental
principles.
CHAPTER
VI
ANALYSIS
OF
THE
CONCEPT
OF
SIMULTANEITY
FIRST
foimd, up
when
m
of all let it be stated that a contradiction between the principle of relativity and the law of constancy of the velocity of light can only be that exactly timed different places. Two clocks are
we assume
if
clocks exactly
are
set
timed
in clock are simultaneously position as the hands of the other precisely the same If we place both clocks before us on the table, clock. the two can we events readily discern whether ^the of the
one
"
the hands
stroke of and the corresponding positions of the hands The ^take place simultaneously. other clock
"
of
one
dock
at the
simultaneity
in close spatial proximity events of two further definition ; if I see them at the
they
take
when
by
the following
Let
us
illustrate
are
we justified
in asking this question. On the 21st of February 1901, a new star became visible in the constellation of Perseus, Persei by the astronomers. This and was called Nova
star, which
mass
cause
certainly been in existence as a dark previously, had been set aglow by some unknown and had thus become visible. The flamingup of had
ANALYSIS
the star undoubtedly
appearance delay bemg
was
OF
took
SIMULTANEITY
place some by human time
41
before its
observation, the equal to the timeinterval needed by light to travel from the star to the earth. The question arises
place ; what date on the earth coincided with the kindling of the star ? Let us suppose that it were possible to determine the distance
as
discovered
to when
this event
took
of the star accurately, and that the result be expressed in kilometres. We can then calculate, for instance, that the hght
exactly thirty years to reach the earth, and the actual date of the birth of the star would be the 2ist of February 1871. The birth of the star and the date, the aist of February to be 1871, are supposed
took
maintained constancy
with of the
unresult must doubtedly velocity of hght is valid, then our be correct ; for, according to this principle, the time required by a ray of light to travel from a point
be equal to the point B will always l^^i light c, qmte AB, divided by the constant velocity of independently or not the two of whether points are
^
to
a
of this principle, or did not believe in it, ^how then ? Let us take
we
"
up the point of view of the old aether theory, and assume that our earth, together with the entire visiblesystem of fixed stars, and including the new star, are carrying
out
a common
the
earth
rectiUnear motion in the direction from We are towards the star. then advancing
the rays of light coming from the star, and hence Thus the birth of these will need less time to reach us. the star did not take place on the 21st of February 1871, towards but it may, for instance, have taken
place
on
the aist of
42
SPECIAL
Let
THEORY
us
OF
on
RELATIVITY
th6 other hand, that the common motion of earth and star is in the opposite direction. We are then retreating from the rajre of suppose, hght
new
on
July 1871.
need longer to reach us, and the event of the have happened sooner, star must say, for instance, Now Michelson's experithe 15th of October 1870. ment,
; they
and taught
of
us
earth
from it, principle of relativity deduced that we could not detect the combined motion Thus, without the aid of and fixed stars.
the
the law of the constancy of the velocity of light, we are in a position to decide on principle, what date on never the earth was simultaneous with the birth of this new
star.
that principle, it is thus quite meaningless to speak of the simultaneity of two events spatially far apart The philosopher will perhaps take another .^^ Without
mind not being If I hit this table with my able to prove simultaneity. instant a prominence hand and at the same bursts forth point of view. may say
:
He
"
Never
the star Sirius,then these are simultaneous events, to know, all my if I never life,whether the even come latter event happened is the philever at all." Now osopher
from
in the
to prove
right?
He
might
be, if the
inability
the simultaneous occurrence of both events was only due to the imperfection of our presentday technical But this case is different. It would be fundamentally means. quite impossible to determine the simultaneity of spatially distant events, without availing ourselves of
that there is no assuming other effect of than light to bring us tidings of distant events. higher velocity to human is no In point of fact, according there experience, In all this
we are
'
such than
effect.
c,
If
one,
should
ever
velocity
greater
of relativity would
ANALYSIS
OF
SIMULTANEITY
48
of the velocity of light ; and what cannot essentially be observed, cannot be said to exist. We might perhaps concede to the philosopher the idea of absolute simultaneity of spatially distant
events
as
be proved.
pure But
thought fiction,though
even
then
we
must
absolute simultaneity if it leads to contradictions between the facts of experience, as is actually found in practice. The matter is different, however, if we retain the If principle of the constancy of the velocity of Ught. this be valid, then the lapse of time between the birth
on earth is perforce and its perception equal to the distance earthstar, divided by the velocity of whether or not both of light c, quite independently
of
the star
We thus see motion. Hence that it is this principle that defines simultaneity. the notion of the simultaneity of events spatially
are
bodies
executing
common
is not given a priori, but is defined by the principle of the constancy of the velocity of light. It be defined most may simply thus : Events A aiid B,
separated different places, are simultaneous observeis stationed at equal distances from A and B the occurrence of both events simultaneously.
which
occur
at
i"
see
the reader perceive the farreaching importance of this principle of the constancy of the velocity of light, far more than a mere assertion conand that it means cerning
more much itdefines than acquaint us with a mere property of Ught the fundamentally connection between spaceand time. We see, then, that the experiment of measuring the velocity a
Does
physical
phenomenon
It does
"
cannot of Hght from the train and from the embankment lead to any contradiction with the law of the constancy of the velocity of light, for the observers' clocks can be
44
SPECIAL
THEORY
OF
RELATIVITY
are
exactly timed only with the hdp of this law, i.e.they by definition never timed exactly aUke, unless measurements performed with them confirm this law. The discussion in this chapter
contains
us
the
essence quint
of the problem of relativity. Let the matter briefly. Formerly, our minds
to
the
persuasion
that
the
conception
of the
given a priori, of spatially distant events was that it had an absolute meaning, and thus required no previous definition. In the application of this idea
however, we a encounter simultaneity, the principle of relativity and contradiction between the principle of the constancy of the velocity of light, be wrong, so that at least one of these principles must
of absolute the absolute idea of simultaneity is to be abandoned. The importance of Einstein's work lies in the fact that, of these alternatives, he gave preference to the two
or
fundamental than
to
the
on
our
conception follows as
His idea was of absolute simultaneity. The principles of relativity and of the : of the velocity of Ught are correct, for they proved
our
having have
to
a
conceptions of space and modify our way that the velocity of light in two
in such
systems
uniformly moving with respect to each other always hgs the same value c, irrespective of direction. How these modifications are to be carried out, is the
of the special theory inferences that can
subject
of relativity ; it contains all those be logically deduced from the fundamental principles.
simultaneous
existence of both
CHAPTER
VII
THE
SPECIAL SUMTOTAL
THE
THEORY
OF THE
OF
RELATIVITY:
FROM
DEDUCTIONS PRINCIPLES
TWO
left
FUNDAMENTAL
WE
our
the observers in the train and on in the lurch, after we had been embankment fuUy persuaded that they can in no way upset
our
of the validity of the two fundaniental principles, provided they perform the exact adjustment We must now revert to them of their clocks correctly.
conviction
perform
measurements
to
strate demon
resiilting from the coexistence In the firstplace, it is easy to show of both principles. that Einstein's rigid definition of the idea of the simultaneity
is not an absolute of spatially distant events but that it is only relative. Thus, when I say that one, the an event at a given place A (say earth)and another happen simultaneously, this statement at a place B (Sirius)
is valid only
at rest
are
who
for myself and for those observers Other observers, however, relatively to me. in motion will take another relatively to me,
point of view,
events
were
demonstrate
be quite right in saying : The two Let us to not simultaneous. proceed how it is that this results from Einstein's
and
definition ; but it should be pointed out that the differences of timeintervals with
forthwith which
we
46
SPECIAL
to deal here
THEORY
are
so
OF
RELATIVITY
the results far from are We
have
presentday apparatus.
shall suppose only one observer in the train, and him to be stationed of the train. exactly in the middle Electric lamps are set up on the embankment at two places A and B, their distance apart They the length of the train. are contactlevers, a flash of light the moment passes it, and the lamp B
convenient
so
being
equal
to
that
with
the beginning
the moment the end of the train passes it. An observer is st ationed on the embankment between A and B. The train travels
midway
Train
+
C
Fig.
2.
A'
of past, the lamps send out their flashes of light, waves both light are with the velocity c from propagated A and B, and reach the observer on the embankment
simultaneously. time, and same Thus
he
sees
both
events
at
the
if he has A
to
ascertained
by
measurement
that
the
places
asserting that the flashes took place But in the train has the observer simultaneously. A, hence travelled a short distance towards meanwhile
from A reach him sooner than those rays of light coming from B, and he maintains quite correctly that coming did not take place simultaneously. the two events be raised : "Is following The this might
objection
correct
statement
also
from
the point
of view
of the
SPECIAL
THEORY
According
if
same an
OF
to
RELATIVITY
definition, two
47
Einstein theory ?
are
events
sees
observer standing in the middle That is not the Ccise here, time.
for when the rays of light reach him, he is no longer in the middle between A and B." This latter argument, is not correct. however, The observer was stationed' midway
between
the lamps
out
their flashes of light ; it is quite immaterial what position the lamps take up with respect to him afterwards. To meet we the abovementioned objection, attached to the beginning and end of the train ; the first of these sends out a Ughtsignal time as the lamp the at A on exactly at the same
can
suppose
lamps
embankment, and justas it passes A, and the second lamp at exactly the same time as the lamp at B on the as it passes this lamp. (Inthis statement embankment, the idea of simultaneity
we are
no
difi"culty,because
only
concerned
The observer observers. doubt, situated midway between both now, without lamps ; if he observes that they emit their flashes of light at different times, he is quite right from he says : The flashes were point of view when
"
by
both
his
not
difference
between
assertions
with
regard to simultaneity which are correct according to definition, and those which are not, the following our be added : A third observer D is supposed to be may stationed
nearer
on
to
A
A
and
to
too, wiU
light from
than
that from
B, but
he cannot
48
SPECIAL
that,
THEORY
as
a
OF
RELATIVITY
maintain
are the events not consequence, for he is not situated midway between of light. On the contrary, he must
AD difference of path between and BD account, and by doing so he will discover that for him too. events took place simultaneously
into
both
thus arrive at the conclusion : Two events which take place simultaneously for an observer at rest, do not for an take place simultaneously observer in motion. We
According
in
at
in justified
on
considering
himself
in motion ; the embankment hence, of course, the inverse assertion holds good : Two for an observer events which take place simultaneously the' observer in motion, do
not
take
not
mines deterevents
simultaneously
can
for him.
We
contact
generalise our result slightly,by supposing the of the lamp B to have a contrivance for delaying
an
it, so
instant
(saythe
second)after
for the observer on the embanksimultaneously ment (the either, but after a certain timeinterval biUionth part of a ; for the observer in motion,
occur
second)
however,
the timeinterval
former generalisation of our We said then : If the timeinterval between statement. is equal to zero for an two spatially distant events observer at rest, it must
zero
for
an
SPECIAL
THEORY
OF
RELATIVITY
we
49 have
In the more observer in motion. general case, When an the value observer at rest measures between two timeinterval distant events, an
in
t for the
observer
a wiU measure slightly different value t' for the timeinterval between This the same events.
motion
is caUed
*
conmathematical analysis of the foregoing sideratio into detail concerning leads them ^without going K and K' to be two systems to the following : Let us suppose of in uniform to each rectilinear motion reference respect with two very long platforms gliding past each other along other, e.g.
more
"
"
line of separation. In both systems intervals along the Une of contact, set up at certain In addition, the clocks of the system going correctly. K' amongst themselves, as well as those of the system their straight
are
clocks
are
(Clocksare then going correctly exactly alike. when, with their help and with that of a standard measuringhght the velocity the measurement of of rod, results in the the clocks of one one system amongst value c.) Furthermore, then timedi exactly the following conare alike when another dition
to be
timed
holds the
moment
good when
A the
lightsignal
clock
is sent
out
at
there shows stationed It must then arrive at a point B at the moment when the time t+r, t being equal to there shows stationed distance AB
c
is going cora clock which possesses who rectly if is now to to compare the supposed with reference his clock with the motion of those K'd.ock.% that he of motion traveller compares as a (In the same successively passes. way An
observer
at
he the theory with the stationclocks passes.) What follows : He will find that the is then as of relativity teaches his own by those clocks and differ in times registered watch his watch such
an a
way
observer
wiU
he
Similarly, slowly. be able to state that the if clocks which so that reference to his own with watch,
are
going
more
going
can
Clocks
This slowly than K'c\ocks. up in the theory of relativity thus : go more slowly than clocks at rest."
more
50
SPECIAL
We
THEORY
quite
OF
RELATIVITY
also in the relativity
find something
analogous
deduce
of
a
be
easily considerations, that the length from the train itselfmust measured length
as we
of
lengp.
We
can
the
embankment.
This
short
formulation
on
it is true, but
For, fully other C according
to
the with
memory,
be applied
i justifiedn looking
as
C would have to gain or then a and system lose with reference C, according as to whether to the one system in motion looked the other were or as upon and this was what the* antagonists as of the theory of relativity naturally regarded
"
of relativity, every his own as at rest and system upon If, therefore, C be a clock in the system K',
caution. is observer
the
K,
that
between of the motion position of their hands only at one for instance, just at the moment,
comparison
particular moment
when
other;
certain
the
of
both
both
clocks
ar"
a close together, and not and comparison be carried out in this second can of their readings moment of in two different ways the dock C can time viz. either be compared
spatially
"
that
clock given
of the
moment
system
K'
which
it happens
to be
at that
(timed
K'clocks) or
to be
like
"
(timed
of the
exactly theory of
Clocks
in moving
systems
of C to C, results, in the
above
discussed.
it should be possible the fact that simultaneity K'. K'clocks, therefore, for ifobservers,
to arrive at different results is due to in K is different from in simultaneity are timed but alike for K'observers,
not
vice
same
versa.
The
as
kind
that
to be
SPECIAL
THEORY
"
OF
RELATIVITY
"
51
from by is meant the train and measured from the embankment." If an measured observer in the train takes a measuringrod, and begins laying it
what
"
buffer of the last car along the entire length of the train to the front buffer of the engine, then the figure representing the number of down
repeatedly
times
from
the back
laid down was measuringrod wiU be the To from the train." length of the train, as measured the length from the embankment, measure we must the
"
determine
two
B,
so
on
the
of the B, are
the
ment. embank
can poiats A and B, we determine their distance apart in the usual way, by the The result of a measuringrod. repeated laying down is the length of the train as meaof this measurement sured
know
the
"
from
the lamps
"
our
previous
exainple
by the beginning
and
an
(i.e.
between
"
embankment).
these
simultaneous Hence
for
we
observer
measure more
must
lamps
(to speak
the edges of their contact contrivances), be, say, a hundred metres. We which we will suppose to then state : The lemgth of the train as measured may
to a hundred metres. from the embankment On amoimts the other hand, the flashes of light were not simultaneous events for an observer situated in the train. For this
distance between
than that of B. observer, the flash of A took place sooner he must Hence conclude that as the beginning of the than the end travelled train travelled past A sooner past B, the length of the train must be greater than the
52
SPliCIAL
For
THEORY
if the
OF
RELATIVITY
been
line AB.
length had
the
same,
the
; if, on simultaneously the other hand, shorter, the end of it than the beginning would travel past the point B sooner As this is not the case, the train past the point A.
lamps
would
have
been
must
one ensure
be longer than
metre
hundred
on
flashing of the lamps for the the simultaneous observer in the train, the lamp at A would have to be in the direction of the moved slightly farther forward moving train, i.e. farther away from S to a point A'.^ For the observer in motion the distance A'B wiU then be equal to the length of the train. Since AB is less
than length A'B,
we
have
the
following
results
(i)The
of the
train is smaller for an observer on the {i.e. equal to AB) than for an observer in
finds it equal to A'B). (2)For the is observer in the train, the length of the track AB the length of the train, whereas the smaller than as regards ^B equal to observer on the embankment
(who
Hence
at
in objects
rest, and
motion
appear
observer
to
an
observer in necessarily arises from the other, because, according to the principle of relativity, both observers are equally shortened
in justified saying
:
am
other is moving.)
This contraction of length occurs only in the dimensions lying in the direction of motion ; hence only of objects train is shortened, and not its the length of a moving The reason for this Ue5 in the fact height and breadth. that measurements of height and breadth do not involve The the detour over the determination of simultaneity.
'
CJ. Fig.
2.
SPECIAL
THEORY
OF
RELATIVITY
53
observer in the train can determine the wheels in exactly the same way the
embankment
can
determine
the
If the the rails,viz. by the use of his measuringrod. wheels exactly fit the rails, both observers will agree that
the
gauge
between
the
between
simply
use
measuringro
we the embankment, ^ people in the train could fix might proceed as follows : from the top and a sharp point to projectsideways bottom The observer on the embankment of the cars.
measure
it from
slab, the surface of which is placed vertically and parallel to the train's motion, and so dose from the cars to the rails that the points
erects
a
large marble
projecting
scratch two
sharp Unes in the surface of the marble slab, as The distance between these Unes the train travels past.
as
at rest (the embankment) system in motion a system (thetrain) could be This process is quite definite, carried into execution. reversible, and can be repeated ad libitum, so that there
can
be
no
rest
and
an
at
of
measuringrods
to the
dimensions
objects
State
We
must
remind
the reader
once
more
that
are are we speaking only conceptual of which to our the comprehension experience concerning even expect to hear an relativity we may
"
The
theory
of relativity is
nonsense,
as
those
described
above
cannot
be carried out."
54
SPECIAL
THEORY
OF
RELATIVITY
of length and of timeintervals have no absolute We cannot reasonably maintain, for instance, meaning. that a pole has such and such a length ; we must always
ments state the motion
of the
can an we
objectmeasured
say
:
event
A
exact
So many in London
we
of time event B
an
To
be
must
is the earth." situated on because between the timeinterval the necessary, events A and B for an observer on a passing comet would have a different value. We do not, in this case, mean observer the events, i.e.the timeinterval between the arrival of rays of light or electric waves, which bring the observer tidings of both events.
the apparent that both observers carry the contrary, we presume out their measurements correctly, and take into account the time needed by light to reach them from the places
interval between
On
where
the events
happened.
CHAPTER
VIII
THE
APPARENT
ABSURDITY
CONCLUSIONS
OF
THESE
THE
on
conclusions set forth in the last chapter present the quintessence of the special theory of relativity, which brought Einstein great celebrity
one
the
hand,
eyes
and of
a
many
attacks
with
the
philosopher,
of two points of view revolutionary, that only one it is an be accepted : Either it is aU nonsense or forward step in our knowledge.
can
portant im
On
the
objection
some
in itself free from illogical, and not contradictions. is not true, and only shows That, however, that the It has been said, for has been misunderstood. matter instance
"
One
of the
to
the
conclusion
take
Einstein
both they cannot contradictory statements, by academic This argument (advanced even critics) the difference the mistake involves of overlooking
If I say, absolute and relative statements. " My hand has five fingers," and somebody for instance,
make be right."
between
S5
56
SPECIAL
"
THEORY
OF
RELATIVITY
No, your hand has only four fingers," one be wrong, for my statement or the other must of course hand is an the number of fingers on my concerning else tells me,
But if a man at Cape Town says, absolute statement. is situated on Madagascar the righthand side of Madagascar Africa," and another in Cairo says, lies
"
"
the lefthand side of Africa," they but both upholding contrary statements,
on
are are
point of view, because the idea of right or left to Einstein, the idea of simulis relative. According taneity is also a relative one, and has lost its absolute their
own
This relativity, however, does not meaning. the position taken up by the observer, as is the right and left, but to his state of motion.
refer to
case
all practical purposes, we may safely continue to consider the ideas of time and space as absolute, for, as will in Chapter X, the difference between be shown ments stateof time and
one
length for
in motion Similarly,
are,
immeasurably
small. it cannot
be
regarded
as
contradictory
to
measuringrod which is longer for an observer the motion of the train train, than the metre measuringrod at rest
logic, that
the
metre
shares on the
on
the
whereas
for
an
to with regard other contradictions hardly realise them, as can be shown by the that we following trivial example : A calf and are the old cow distance from them, another grazing in a field. At some The firstcow are grazing. calf and another cow naturally
appears
to its
own
cow,
ABSURDITY
owing
to
OF
c6nCLUSI0NS
57
" Hence My the first calf says ; proximity. is larger than yours," and the other calf retorts : mammy is larger than yours." The calf's "No, my mammy involves only the angle subidea of magnitude tended seen, at its eye by the and if we accept the
" "
object
"
"
word
magnitude
in this
sense,
then, of
course,
get used
as
a
to accepting
two
events
angular
way
as
of the theory
point
out
again
see an
depends object
the position of the observer, whereas the space and depends on his state two timeinterval between events have not previously noted That we of motion. of this relativity, is due to the circumstance thing anythat
means are a million carried out by human all motions times too small to permit differences of length or time to be observed.
On
as
a
the
matter
other
history of philosophy) logically correct, and yet without use or having been derived logical deductions suppositions of knowledge. our
from
a
a
of course conceive (and in the of fact it is of frequent occurrence of a succession of ideas being
we can
hand,
nature
We
for entirely void of importance are that, conready to admit sidered of view,
are some
deductions
similar theory
of the
raise
a
liable to
impression.
of
the
relativity has
part of antagonists, the been banteringly termed a Talmud, appear and very there is
no
the
newcomer,
who
approaches
58
SPECIAL
the good
THEORY
OF
RELATIVITY
with
common
proof
(calledound old traditions of thought s sense). By way of example, let us take the is given in the last chapter, that the path AB
smaller for the observer in the train than the length of the train itself. That proof is based on the following : The flash of light sent out from the lamp A takes place
sooner
from
(not only
the time
account
from
rays of Ught to reach him the beginning ; and the end of the train) he concludes from this, that the beginning of the train arrives
appears needed by
than the end at B, and proceeds to conclude that the length of the train is greater than the path AB. Sound common sense, naturally taking the point of view
at A
sooner
of the simpler absolute theory, will raise the following : objection In reaUty, the flashes of light at A and B The observer in the train advances were simultaneous. towards the rays of light coming from A, and therefore
"
; from
movement, of his own and that is where the of this entire way of thinking scholastic hj^ocrisy in." Such thoughts comes must arise in every reader follows the matter attentively, so long as the who
absolute
enough
sufficiently penetrated theory of relativity will defend observer in the train somewhat
and space are rooted deeply the other hand,. whoever has into the progress of ideas in the
the
as
"
hj^ocritical
:
"
follows
That
he
ignores
as we
the have
is quite
repeatedly emphasised, according to the that the embankprinciple of relativity,the statements ment in motion, or that the train is at rest and the train
ABSURDITY
is at
rest
OF
CONCLUSIONS
59
equally
in motion, are quite embankment for justified, the one thing that matters is the and
the
If he, too, were an relative motion. absolutist, the The flash of light observer in the train might say : from A than from B ; the observer was sent out sooner however, on was the embankment, towards moving
"
from B, hence the emission of the rays of light coming both signals appeared to him to occur simultaneously. That is only because he pretends to know nothing of
his
can a own
motion."
We
thus
see
that
both
must
were
in truth that
"
simulwere
taneoiis
not
or
simultaneous,
but
from
they
simultaneous
otherwise
as
seen
their respective
between
the theory
of
are not scholasticism is this, that we dealing with subtleties designedly thought of, but with facts of logical results drawn from two experimental its own Logic never was with Einstein, nature.
object
but
only
an
the
instrument
with
which
he freed physics
from
embarrassing
situation.
CHAPTER
IX
THE
UNION
OF
SPACE
AND
TIME
; THE
MINKOWSKIWORLD
this chapter we propose to consider the results of the special theory of relativity from a new point it all the more makes plausible of view, which
to
IN
We
as
faculty for readers gifted with an imaginative Those who are not so equipped will perhaps geometry. find it difficultto keep up with the following discussion. shall
commence
space. which
We
take
classical theory of absolute shall call events like the flashing of a lamp, the
"
place at a certain point of space, and at a To determine pointevents." certain instant of time, the place and time of a pointevent without ambiguity,
we
must
of the
For pointevent. figure is required, e.g. the number one between have midnight elapsed
which
(Greenwich time)
of the. of the close of last century, and the occurrence But to state the place of the event, three numevent. bers because space has three dimensions. are required,
We
know
that to determine
any place on the earth, its latitude must be given. longitude and however, the point is not yet fully deter6a
UNION
mined,
OF
SPACE
AND
TIME
61
all points that lie vertically above each geographical longitude and latitude. other have the same Hence the height above sealevel must also be stated, and
case
because
then
the
In the point is definitely determined. earth is the system of reference for our
this
coordinat
To
course,
determine
the
positions
of
stars,
On of reference are other systems used. other hand, in order to fix a particular point in a closed it will be best not to use space, for instance in a room, the geographical longitude and latitude, together with the height above sealevd, but to state the distances from each of two perpendicular walls, together with the
height above therefore
of the
the floor. To
always
and one 2'5 m. from the front electric lamp hanging in a room wall, 3 m. from the lefthand side wall, and a m. above the floor, is to flash out at the time 12 seconds after (chosenas the commencement of counting midnight
pointevent, we must three spacecogive four numbers, ordinates For instance : an timecoordinate.
fix
time).
2'3
on m.,
The
3
m.,
coordinates
2
m.,
of the
12
pointevent
are
us
then
and
seconds.
Let
suppose
another
a
room
point event : a second electric lamp standing of the same writingtable in the righthand comer is to flash out at the time 8 seconds, its space
5 m., and 15 m. coordinates being i m., form the differences of the corresponding
of the two events : difference between

Now
let
us
the time elapsing between the flashesof both lamps, the difference of the third spacecoordinates 2 1"5 gives the difference in height of the lamps, and the difference of

the other two pairs of coordinates indicates how much farther, forward one of the lamps is situated than the
62
SPECIAL
THEORY
OF
RELATIVITY
to the right. According more much other, and how to the classical theory of absolute space and time, the timeinterval between the two events is always equal to
seconds, quite independently of which system of The the reference is used. spatial distance between two lamps (the length of a thread stretched rectilinearly
between
pendently certain value, quite indeof the choice of the system of coordinates. (Itis obtainable by a simple mathematical operation from the differences of the spatial coordinates, as
them)
also
has
win
to many
m.)
enclosed within the inclined to the walls of the firstone. We can, of course, also give the positions of both lamps by means of the from the floor and coordinates (distances relative walls) Now to the second room. the second system when of it reference is actually inclined to the first, happens that the coordinates of both lamps different not only are from those of the first system of reference, but also the differences of their coordinates. But if we calculate
The result in our readers. Now let us suppose another room first room, and that its walls are
the
distance
between
the
two
we
lamps
from
the
new
differences of coordinates, before {i.e.*55 as 2 value these results as also the and
are
follows
m.).
obtain We
exactly the
same
may
; they vary accordrelative magnitudes ing to the choice of the system The of reference. distance between two points, however, and the interval two of time between they are independent
"
; magnitudes of the choice of the system of {N.B. ^We are stillspeaking from the point reference. of view of the old theory of absolute space and time.)
events
are
absolute
UNION
To
OF
SPACE
AND
TIME
63
be quite certain of being clearly understood, we shall of the relativity of coordinategive another example of simpHcity, we shall Fig. 3 shows the profile dimensions. right and left by two hillsA and B. ofa plateau bounded We shall suppose the plateau to be slightly inclined towards
sake
for the
the horizontal plane, and a place 0 to be situated To determine the position of the at its lowest point. two hilltops with reference to this place, we can proceed
to
0 and their give their horizontal distances from draw For that purpose zontal hori0. we a height above 0, and drop perpendistraight line hh through
Fig.
3.
heights
with
reference
to
0.
0, and Hence
AC OC
and and
BT) their AC
are
the coordinates
the of the point A, and OD and BD coordinates of the point B with reference to the system Now imagine circumwe can of coordinates chosen. stances
which
of the place hh through
it convenient for the inhabitants 0 not to draw the horizontal straight hue 0, but to draw a straight line ee running
make
parallel to the inclined plane itself,and to use this as the basis for the system of reference. They then define distances of the respective hilltops the perpendicular
64
SPECIAL
that line
THEORY
ee
OF
their
"
RELATIVITY
height."
In
as
this
case
are the coordinates and AC hills have The those of B. different horizontal distance
"
from
0,
as
viewed
what
this
inclined
system
of
reference.
But
always
are
remain indepraident of the system of reference the distances of the hilltops in the serialline from 0, and BO, and the distance AB of the hilltops from tance other. Thus, what is really invariable is the disbetween two points ; the differences of height and
are
i.e.AO each
horizontal distances
on a more or
only
of projections
this distance
less arbitrarily chosen framework of coordinat They take the part of shadows, and vary
in magnitude and shape according to the position of the In other words, the concepts of plane they fall upon. difference of length, difference of breadth, and difference
of height
are
not
absolute
They
are
spatial separation. is all this connected The reader may now ask : How should now with the theory of relativity ? The reason be readily understood. ence Just as, for instance, the differof height between two points according to our classic view has no absolute meaning, but depends on the choice tween of the system of reference, so the spatial distance bepoints, measured two timeinterval between meaning according
two
single concept
aerial Une, and the lose their absolute events, to the theory of relativity. These
in the
ing magnitudes, too, are liable to variation in value accordis the result ? What to the system of reference. In the same way as we said before : The three spatial coordinates are only three single dimensions, the components of the notion
of distance
in space,
so
now
we
UNION
may say
:
OF
SPACE
AND
a
TIME
pointevent they
of
are
65
All
four coordinates
nor
of
are
absolute ;
components
just
imited
the four
includes space and time simultaneously. idea, which From this day forth, Space taken by itself,and Time
"
by itself are
of union
were
to
become
is to
mere
shadows,
of both
retain
kind
These
the words with which the great German matician matheMinkowski Hermann introduced his lecture
of Natural Scientists at Cologne first introduced this viewpoint of
"
where
he
of relativity. According to his proposal, " World the union between space and time was called Minkowski by physicists. Furthermore, that showed
theory
World." and with it is possible, by using this notion of device, to give the help of an ingenious mathematical treatment the mathematical of the theory of relativity
a
"
form
of such
achieved
complete by previously
harmony, any
as
had
never
been The
phjrsical
theory.
it appears although of treatment, relativistic mode to the lajnman, and, even at first sight absurd when he has got used to it, at the least very complicated, lucid for turfls out to be much more simple and more This in itself is a reason treatment. mathematical in favour of the theory of relativity, which must have mental weight with the theoretical physicist. For the expericircumstance be that there is no other way of uniting the two must both of fundamental principles repeatedly mentioned, which have been proved by experience.
physicist,however,
the
deciding
Thus
the
"
"
world
; but whereas
dimensions, the fourth space has three equally justified {i.e. time) is found to play a part available dimension
66
SPECIAL
THEORY
us
OF
RELATIVITY
try to explain clearly the difference dimensions justified and particularised Let us suppose two lamps fixed in a room
each
m.
to
a
be
sofa
them from this side position, close by, and contemplate they appear to me to be situated alongside of each other, from above, so that my head is and if I observe them situated they in the
to
"
appear
convert
above by
both lamps, straight line connecting I can behind the other. Thus be one " behind into beside or ad
" "
"
"
Ubitum,
suitable
position
quite
what
distance, the
points are situated. contemplated In the example of the two hilltops in Fig. 3, a suitable choice of the plane of reference can always result in the " ence difference of height of the two hilltops (thediffer"
two
of their perpendicular distance from ee) being in the theory does this work, now, How equal to zero. Just as the difference in height and the of relativity ?
horizontal distance in the example of Fig. 3 can acquire different values, according to the choice of the system of distances between reference, so the spatial and temporal
two
different values, when conpointevents can assume template from systems of reference in different states of But, whereas in the case of the two hilltops motion.
a
"
line of reference can always be chosen " defined above difference of height
such
appears, disunder
in
the
theory
a
of
relativity
we
may,
UNION
OF
SPACE
AND
we
TIME
cannot
67
pointevents is reduced to zexp, but do so. (We shall find more about As shown are above, we
chapter.)
able to exchange
a
spatial coordinates
by
a
"
suitable choice of
" "
system
of
into above alongside of," reference, and convert but we do this analogously cannot with all worlddimensions we {i.e. cannot completely exchange spatial distance with temporal That the time
succession).
coordinate follows as
particularised role in the world," for our a matter most primitive of course, teaches us that time and space are different plays
a
"
two
of each
to
of relativity
teaches
us
The next chapter that this is not the case. it is that they could be maintained how us
as
ideas, and that for allpractical continue to be so. purposes they stilljustifiably Let us contemplate the classical Newtonian conception
independent
more,
before
we
leave it.
In the
work.
"
Philosophise
^rightly considered
the fundamental
pillar of physics and of exact natural ^Newton says : science generally I. Absolute, true, and mathematical self, time, of it"
"
and without
"
from regard
its own
to
nature,
flows equably
any
thing
always
According uniformly
to
a
Newton,
stream,
like
68
SPECIAL
THEORY
or
OF
not
RELATIVITY
space exists like a it would large empty vessel, and, according to Newton, Long if it contained nothing at aU. still exist, even before Einstein and Minkowski, physicists and philosophers
events
take place in it
; and
perhaps with greatest clearness), here makes had pointed that Newton out assertions the description of the actual facts of which go beyond
(Ernst Mach
anything like time exist if all matter in the universe lay dormant, and executed no movement ? happening at all were and if nothing whatsoever,
nature.
Would
Could
space
exist
if it contained
to
us
nothing
as we
These
may perhaps appear it is necessary subtleties ; but from the conception of time as questions
that
a
stream,
uniformly into eternity, and of the conception of space described above, before we can appreciate the ideas of Einstein and Minkowski, who regard both as single For, accorddimensions of a greater whole ing ^the "World."
"
to the theory of relativity,timeintervals and spatial distances vary for observers in different states of motion.
have followed with sufficient readers who on the the considerations of this chapter attention World, will perhaps, on thinking over Minkowski all
Those
before, put the following question : gone According to the classical conception of space and time, the height differences and horizontal distances of two
that has
the system of reference, and their spatial distance, are absolute ; but nothing in the aerial line, has an absolute value measured In the independent of coordinates. of the system
and
on
points A
B depend
distances of relativity, the spatial and temporal pointevents take a similar part to those taken of two by heightdifference and horizontal distance formerly. theory
UNION
Is there an which
OF
SPACE
magnitude
AND
TIME
69
absolute
in Minkowski's
World,
is geometrically from capable of construction distance, and which spatial and temporal plays the in the same part as formerly the distance measured of the system aerial line, i.e. which is independent of be anreference ? This welljustified swered question must in the afiirmative ; there is an absolute magnitude Worlddistance of this kind, and it is called the
" "
of the two
'
pointevents.^
familiar
the elementary mathematics, with clear in a subjectmatterset forth in this chapter can be made few lines. If the coordinate differences of two points in space readers
law of Pjrthay, z, we find, with the help of the wellknown goras, that the value of the spatial distance of these points is
are
x,
For
If any other coordinate be used to give the position of system differences will in general the points, the value of the coordinate In calculating the spatial distance be different, say x', y', z'. from these coordinate so that results ;
new
differences, the
same
value
as
before
again
Thus
the
'
case
is in motion
relatively to the first coordinate system, and differences are given by *', y', z', and f, the coordinate which find that the equation
in
we
is
two
no
longer
strictly
fulfilled ;
hence
the
(as mentioned its magnitude depends ; meaning The real absolute coordinates. is fully independent of which
pointevents
"
"
repeatedly)
on
spatial has
distance
of
the
choice
worlddistance
of
the two
magnitude mentioned above, the coordinate system (the is now given by the pointevents)
expression
sJx*+y^+z*cH*
where
c
slx'^+y'^+z'^cV*
vacuo.
represents
CHAPTER
NUMERICAL
CONSIDERATIONS
we
IN
from for
an
Chapter
the
of
VII
two
two
drew
the following
from
fundamental
lamps A
flashing each
and
distance
other, although
on as
taking
observer
(2)
If the
the two
lamps
measured the
the train, is smaller than its length. These conclusions, having been drawn
are
without
of a purely qualitative nature ; have not so far indicated the magnitude we of the time and lengthmeasurements discrepancies between
aid of mathematics,
observer in the train and for an observer on the From the numbers quoted in Chapter III embankment. be very the reader will readily imagine that they must is shown in the as small, and that is indeed the case,
for
an
duction denumerical formulae of the mathematical The reader wiU have to accept on trust the theory. in this, and in the next chapter, whereas the statements the qualitative conclusions contained in the preceding following
calcidations, resulting from
which
contain
NUMERICAL
chapters
were
CONSIDERATIONS
a
71
purely logical nature, and could be readily controlled by the thoughtful reader. In the firstplace, it is selfevident that the statements of of the
two
observers wiU be identical, if the relative between is equal to zero. Secondly, them
are
practically equal
to
zero
be given to material bodies disposal. For example : at our with the technical means We shall suppose the train in Chapters VI and VII to travel with a velocity of io8 km. per hour, i.e. 30 m. a
can
second,
and
as
If the lamps itselfis 150 m. emit flashes of for an light exactly simultaneously the observer on he were the observer in the train (if embankment, able
to
measured A and B
from
the train
exactitude ^which in reaUty !) he never can would state that a timeinterval the of 000000000000005 seconds had elapsed between Furthermore, two the length of the train as events.
measurements
"
carry out
of such
measured
from but
the embankment
would
not
be exactly
150
m.
an
be reduced still more. difference in the measurements no of the distancetwo as seen and timeinterval between pointevents from
If the velocity or the smaller, these differences would It follows that we should perceive
atom.
the train and from the embankment, instruments to be were precision of our
even
if the
a
increased
statements
that
we
may
An
as
72
SPECIAL
THEORY
metres,"
"
OF
though
observer
we
RELATIVITY
ought
to
length of ten
quite correct,
motion,"
add, to be such
a
for
an
In e.g. who is at rest relatively to the earth. point of fact, the length of the rod is different for an relatively to it, than for an observer observer in motion
at rest
milliard
a and powers of measurement, million times less, for instance, than the changes which its temperature the length of the rod undergoes when is raised by a tiny fraction of a degree by the near
times
approach
Whilst
of
human
being. by
the theory tivity of rela
that
assume
we
remain have
infinitesimally small for those velocities to deal with in practical Ufe, they would magnitude succeed in to the velocity of Ught.
we
considerable
if
could
km.
per
would permit us to run round the were and if we equator about six times in a second), of length during able to carry out an exact measurement this furious journey (both possibilities are, of course,
(thiselocity v
the observer on the embankment of the question), find the length of the train to be only half as would long as would the observer in the train. But if the
out
train
were
dimensions
with the velocity of Hght, its from the embankment would contemplated to zero Of course the distance altogether. travelling
A and B ^yould, on the other hand, also for the observer in the train, since all
ones,
be equal to
zero
as
with remind
us
velocity greater than that of light ? Let him of what declared in we emphatically
a
NUMERICAL
Chapter based
on
CONSIDERATIONS
the special that theory
no
78
VI,
the
that
supposition
a
with
velocities with to the theory of relativity, According greater than c. the velocity of light pla}^ the part of a Umiting velocity, be exceeded, and which is never cannot able attainwhich We shall have material bodies. this point in the next chapter. by
more
hence
no
than
move
to say
on
along make
with
velocity approacliing
curious observations
other
in connection with the relativity of the notion of time. Let us suppose the that by the year five thousand development
as
of human
not
technics
an
to
admit
only of
but also, that we solar system, planets of our should be able to visit the planets of distant fixed stars, Besides this, we and had established colonies there.
other
the introduction of an interstellar time, shall imagine that the inhabitants of distant fixed stars could set so by to agree their watches with terrestrial watches
know that the idea of wireless signals. As we shaJl define interstellar time as of time is rdative, we the correct time for an observer at rest relatively to our
means
solar
system
of service between
Union
in the
on
the
constructed as to be after starting from the earth, and more accelerated more so that their speed nearly attains to the velocity of light ; and they do not slacken speed until they get
the
stars
are
so
"
brakes
"
are
appUed
We
on
in the year
5500,
traveller goes
74
SPECIAL
THEORY
OF
RELATIVITY
worldships to visit a colony on a planet belonging to a fixed star, at a distance of about hundred lightone from the earth. The ship starts with full years away
force, and
increases
to
(equal nearly
c)
until the
utmost
velocity
reached.
We
that this process of acceleration, according of the traveller and the ship's chronometers, From the instant the maximum months.
is attained, only a few seconds statement of the ship's clocks)ntil they are sufficiently u to the planetary S37stem of the distant fixed star near
to require to slacken
speed
again, and
thus
reduce
the
velocity of the worldship to rest, a process which again That occupies six months. part of the journey performed with the maximum velocity, and during which by
far the
appears duration
the
a
distance
moment.
journey involves
only
the
acceleration and the period of retardation, or a year altogether. But when he leaves the ship in the planetary stellar colony, he will find himself in the year 3600 intertime, and if, after a stay of a few weeks, he then returns to the earth, he will not reach it until the year have disappeared, generations of men 5700. Meanwhile dead are but he his greatgrandchildren and gone, himself is little more than two see years older. We that the dream of H. G. Wells' TimeMachine might be could succeed in some realised, if we way in imparting of locomotion velocities approaching the velocity of light. Let us return to everyday reality. The sober reader will be surprised that exact science places before him
to
our
means
such
fantastic
pictures.
The
"
No
NUMERICAL
such
CONSIDERATIONS
; why
75
sense non
me
should
are
those
who
the
theory
by
For aU reasonable velocities attainable in practical life,the effects are so small that
we
we
that
can't
measure
them fanciful
that such
the
as
most
are
and
absolutely
so
unreaUsable
we are
with
our
modern
technical
means,
that
we
unable to put
same
them
as
to
the test."
end of
To
this
give
:
the
reply
at
the
Chapter
VIII
These
unyielding
logic from
the two
(theprinciple
as our of the experience does not disprove the validity of these two laws, we are to believe deductions made from them, and compelled do so all the more we must when we take account of the
of the constancy
fact that sma;ll bodies do exist in nature, the velocity Exact the velocity of light. approaches of which
on made drawn the consequences
measurements
them from
Of these
Before
we
point
According to the special theory of importance. of happen events may relativity, it is possible that two at different points of the earth's surface and at different times place terrestrial observer, but for an simultaneously observer
a
for
that
they
take
relatively
in motion
to
understand this as meaning that a S3rstem of reference of this kind happening event could be discovered, from which an happening in London in today, and event another
to
the
earth.
But
we
are
not
76
SPECIAL
York
THEORY
OF
RELATIVITY
New
tomorrow,
simultaneously.
events
(thedistance Londonof such spatial proximity New York is very small compared with astronomical can distances) only be seen simultaneously by observers
on
system
as
them,
state
in motion, if the timeinterval between Let us observed from the earth, is very small.
this numerically : Two events at which occur for an observer in different places can be simultaneous rapid motion, only if the interval of time between their detection for
terrestrial observer is smaller, or at most, equal to the time required by light to travel from one Thus, if the rectilinear distance place to the other.
a
between
1000
the
km.,
two
events
happen
for
is
a
the events
or
terrestrial observer
to,
be less than,
an
at most
equal
3^th
we the timemay accept interval of the moving observer would g^th second, have to travel with the velocity of light, if the events
observer
in motion
were
to be simultaneous
for him.
If the timeinterval
by a terrestrial observer were measured ^^th second, be simultaneous for an observer moving the events would they with onehalf of the velocity of light, whereas
order for an observer moving What has been said here with the velocity of Ught. illustrative of the statements as serve near may made the end of Chapter IX, where the singular part taken
would
occur
in the
reverse
by
time
under
amongst discussion.
is
more
one
the
four
"
World
"
dimensions
was
There
much
other far less obvious, but fundamentally important difference between the interof the spatial
changeability
dimensions
(cf.p. 66),
NUMERICAL
and the dependence
CONSmERATIONS
77
of the spatial and temporal distances of two pointevents on the state of motion of the system To return to the example tops, of the two hillof reference. we can arrange, by turning the system of reference, that both the difference
" "
heightdifference
altered.
"
and
we
the
"
horizontal the
will be
If
investigate
different possible positions of the system of reference, discover that if the system of reference be turned in we to make as the heightdifference smaller, such a way
then
vice
the
larger, and horizontaldifference will become is different, however, The in the case versa.
sjTstems of reference moving If the timeinterval between relatively to each other. the two events in the second system is greater than in
and
two
time.
Let
us
contemplate
two
the first,then
the
spatial distance
than in the first, (Comparethe numerical and vice versa. the at the beginning of this chapter, where example length of the train as well as the interval of time between
lamps, as measured flashing of the two from the from the embanktrain, is greater than as measured ment.) As we before, this fact is not so obvious said the
to the
nonmathematician,
the profound creates and difference between space and time.^ two That terrestrial events take place simulcan taneously for a moving they occur observer only when within
*
small
fraction of
formulated,
second
for
an
observer
at
this circumstance is due to the t', enters into the fact that the square of the timedifference, " " for the (cf. footnote, end of worlddistance expression Chapter IX) with the opposite sign to that of x", y\ and i".
Mathematically
78
SPECIAL
on
THEORY
OF
RELATIVITY
that the the earth, is due to the circumstance This magnitude c velocity of light is so immense. in physics, and it is magnitude represents a fundamental
rest
convenient, therefore, to choose the imits of length and to make as time in such a way the velocity of Ught If we retain the second as the unit of equal to unity. In these time, the unit of length will be 300,000 km.
as we units may call them ^the duration life is a very long one, for it amounts on the of human The spatial scene miUions of seconds. average to many the other hand, is very Umited, existence, on of our
" "
natural
"
"
of our globe measures only about 0*04 It wiU perhaps appear to the reader units of length. is without import, because the that this statement
is quite arbitrary, always arrange matters in such a way that and we can in these units is equal the earth's diameter measured ad libitum. either to a large or to a small number,
choice of
our
time
the units of when we choose the relation between length and time so that the velocity of light is equal to But after what ^which is physically wellfounded, unity, measure has been said above, (however ^the numerical we of the duration of our Ufe choose the single units) milMons of times greater than win, in any case, be many
" "
that
of our activity.* We of the spatial extent return to this point in the second part of this book.
implies that all motions statement beings are very slow associated with human
1
will
This
carried
as
out
by
or
compared
with
the velocity
of propagation
of light.
CHAPTER
XI
FURTHER
CONCLUSIONS
AND VERIFICATION
THEIR
EXPERIMENTAL
the
IN
Hence
our
idea
of
was
VI),stress
which involves the law of the constancy of the velocity of light, is without purport unless there is no other effect whatsoever that is propagated with a velocity greater than c. it is not surprising that this theory subsequently gives rise to results, according to which material bodies be given velocities greater than c on the one never can
that
the Einstein
definition of simultaneity,
on
to
that
any
ensue
material)
with a velocity greater than effect could be propagated If an effect of this kind existed, one could that of Ught. in which devise experiments the effect preceded the
cause. so no
we
is quite contradictory to experience, and again have to conclude that there are shall once
This
effects which are propagated with a tioned velocity greater than that of light. Besides, as menin the last chapter, for an observer at rest the length of a body moving with the velocity of light would such
as
things
be reduced
to
zero,
and
furthermore,
that such an observer would obtain an for the length of a body moving with
velocity greater
80
SPECIAL
THEORY
OF
RELATIVITY
than that of Ught, a result which, from a physical point Thus we are of view, would be absolutely unreasonable.
with the velocity of hght in the theory of relativity, as the upper hmit of all velocities. This is in complete accord with our experience, as we know of At one no quickly than Ught. effect which travels more
again
confronted
time
was gravitation propagated with a greater velocity than light ; but this has proved It is interesting to foUow up the matter to be erroneous.
it
was
assumed
that
in regard to the velocities attainable by material bodies. traflBc, shooting, All velocities connected with human and so forth, are so ridiculously small compared with the velocity of light that they the much
do
not
count at
all.
Even
appear in astronomy, the solar system, "the velocities of planets and comets of the latter of which those of fixed stars and of meteors,
greater velocities which
occasionally
^are,
as
a c.
than
material
bodies in nature
the value c. velocities approaching the atoms are of electricity, the socalled electrons, space of a Rontgen such as pass through the evacuated bulb with enormous over, Morevelocity when it is working.
with of the rays emitted by radioactive substances by single atoms consist of electrons, emitted ((8rays)
some
which These
substances with an incredible velocity. The atoms of the elements themselves, when passing through rarefied gases during electric discharge, attain smaller
of these stiU velocities (though balls or cannon radium, compared with those of so stars), do also the socalled arays of know now with certainty, are which, as we
enormous
of the rare nothing else than electrically charged atoms The velocity of all these particles has gas helium.
FURTHER
CONCLUSIONS
81
been found to be measurable, and it was discovered that it varies according Some to experimental conditions. relatively very small velocities of only a few km. less hundred {i.e. than that of many per second but others have the velocities approaching have
comets),
highest
300,000 km. per possible velocity of about If we to put all the velocities appertaining second. find an unbroken we material bodies in nature in a row, From the movement of glaciers, sequence of velocity.
to a small fraction of a millimetre which only amounts per hour, up to the incredibly large velocities of the electrons in the form of /8rays, every possible velocity in nature ; but slightly below is represented somewhere
From the point of the velocity of Ught the scale ends. view of the old classical physics, this might be regarded as merely accidental ; according to that view, it might be thought possible that a j8ray might alsa possess a From the point of velocity of 310,000 km. per second. the upper view of the theory of relativity, however, limit of velocity is not accidental, but is a natural law. It is impossible to have velocities greater than that of light. The theory of relativity is not content in merely la5dng sible down the law that velocities greater than c are impos;
on
the
contrary,
formulae, it actually gives a reason why greater velocities exist. To make this clear, we must proceed still cannot in the first chapter that, It was further. explained the of mechanics, classical theory is strictly valid for mechanical principle of relativity the old ideas of space and time. processes based on EinsteinMinkowsMIf we replace these by the new
according
to
the
world
ideas, 6
we
82
SPECIAL
THEORY
OF
RELATIVITY
longer be valid for mechanical can no classical mechanics is supposed As this principle" however, to processes.
be
no
general law
of nature,
the theory
of relativity has
other
choice than
to state
alteration in the fundamental of equations them satisfy the enabled Einstein to make mechanics views principle of relativity, on the basis also of the new These changes are of such a kind of space and time. sUght
classical for velocities met with in human which occur mechanics in astronomy. For or technics velocities, however, the velocity of light, the deviations from approaching
that
we
can
neglect
the
deviations
from
They are the laws of the older mechanics considerable. consist in the following : In order to set a body into to we as is well known, must, motion {i.e. accelerate it),
apply
to
force to
overcome
ing Accord
Newtonian fundamental law of the elementary mechanics, this force is equal to the product resulting from the inertial mass of the body multiplied by the If, for instance, of the acceleration. magnitude of I kg. is to be accelerated from the condition
so as
a
mass
velocity of lo m. per second force must be applied. In end of I second, a certain before us, this force wotild be about equal the example to the force exerted by gravity on the kilogram weight.
attain
a
to
of rest at the
To
10
cause m.
the per
one
same
body
to
20
to
m.
second
would,
second, have to
increase
to classical mechanics, according force, and so on. Thus to the same expend kilogram the velocity of the same weight
within
another
second
from
10,000,000
m.
per
FURTHER
second, exactly other for instance,
same
CONCLUSIONS
to
10,000,010
m.
83
the
words,
force would
have
second, In to be applied.
per
force and quotient resulting from the acceleration {i.e. inertia! mass)is a perfectly definite for a particular body, and is quite independent number of the velocity. This no longer holds good in relativistic the mechanics
whether
to to
10
m.
according to this, it is not quite immaterial from the velocity zero body is accelerated per second, or from the velocity 10,000,000
:
10,000,010
case
the latter
the
mass
The force required in per second. In other words, would be rather greater. body is not constant, but slightly increases
m.
to the statement
above).
This
detectable
inertia!
mass
on
railway train of 200 tons weight is, thousandth part of when at rest, only about a himdred to travel at the rate of less than if it were a gramme But the changes in the mass km. per hour. 100 of a body, the velocity of which is approaching of hght, are very considerable, so that the
the velocity
mass
of every
body
enormous
one
fingers,we
all the
forces stored in our solar system would not suffice to accelerate this selfsame particle it had attained the velocity of light. of dust, if once
Thus, from the point of view of the theory of relativity, we are why the manifold velocities able to understand are one and all of material bodies existing in nature
limited
justwithin
the boundary
This
84
SPECIAL
a
THEORY
OF
RELATIVITY
tells in undoubtedly striking circumstance, which favour of the theory of relativity ; but, on the other hand, be taken as a direct proof of its validity. It it cannot to examine Einstein's found was possible, however,
is
deduction
of the dependency
of
mass
on
velocity for
that the the /3rays of radium, and the result proved creasin inertial mass really does increase with inof a body demanded by the velocity, and by the amount
formulae of the theory. a The theory of relativity has achieved triumph in recent years, since the German
surprising physicist
succeeded in explaining mathematically the socalled fine structure of spectral lines of A hydrogen and helium with the help of this theory. few remarks on this will be useful. If a photograph
Sommerfeld
in Munich
subject
be taken
of the spectrum
of
luminous
electric
in rarefied gases (Geisslertube), sharply very defined lines appear on the plate, these belonging to light of of a distinct colour {i.e. a certain frequency of oscilla
discharge
tion).
These numbers,
lines
always
in
which, though
nevertheless succeed each other regularity. This kind of with a certain mathematical has therefore received the name of seriesspectrum
Up to spectrum. of the mechanism
luminous
a
knowledge
atom
gas, emitting these seriesspectra, was In 1913 the Danish physicist N. Bohr theory proposed succeeded, with the help of the quantum by the German physicist M. Planck, in throwing
of a complet in
the process of the emission of seriesspectra. A discussion of this theory, which is in many respects
light
on
stillmore
complicated
and
more
mathematical
than
the
FURTHER
theory
of relativity, would
CONCLUSIONS
take
us
85
too
far.
We
shall
merely mention that each of the atoms of the elements is supposed to be a sort of planetary system. In this the socaUed nucleus of the atom takes the part of the
sun
; it is
trons, Elecconstitutes almost the entire mass of the atom. tiny much lighter in weight than the nucleus (i.e. particles, considered to be the atoms of negative elecit in circular or elliptic orbits, around like planets aroimd The physicist Bohr, by the sun. applying the laws holding good for planetary movement
move
tricity
in astronomy to the orbits of electrons in the atom, them with the laws of the aboveand by combining theory mentioned quantum of Planck, arrived at a theory of the seriesspectra of hydrogen and the
helium
which agrees splendidly with experience.^ This is not immediately connected with
theory
of relativity. A closer analysis of these series, however, the following results. The gave single lines of the spectralseries are not in general simple lines ; on the
are
lines consisting of two, comphcated lines very close together, so that when
investigated with a spectral apparatus of small dispersion, they appear to shrink into one single line. example familiar to aU who have
The
best known
The sodium containing with a spectral apparatus. Bohr theory could not explain the appearance of Hnedoublets, triplets, etc., in the spectral series of hydrogen
and
helium. that
case
At
this
in juncture,
1916,
Sommerfeld
atom
occur
showed
1
if the
orbits of
electrons in the
serieslines which
In the
of helium, only
for those
in the socalled
spark
spectrum.
86
SPECIAL
THEORY
OF
RELATIVITY
are not calculated of planetary orbits) ing according to the laws of classicalmechanics, but accordto relativisticmechanics, taking into consideration
to (similar those
the
abovementioned
we
dependency
not
of
the
mass
on
only the general structure of spectral series already elucidated by Bohr, but also the fine structure of the single lines. In addition to this, Sommerfeld was able to predict certain complicated
velocity,!
obtain
from calculations groups of lines in the helium spectrum based on the theory of relativity, which were quently subseby means in Boim confirmed by Paschen of very deUcate spectral measurements. Let us survey the results we have far obtained
so
in
confirmation of the theory of relativity. experimental VI, the First of all, as shown at the end of Chapter two fimdamental foundations the of the theory
"
by and strengthened principles ^have been supported so that we the most experiments, careful and exact should beheve the vahdity of the theory of relativity
"
even
if no
a
further experimental
That
the
immediately
which is due to the fact that all deviations from the old laws of mechanics and electrodynamics, and all the divergencies from our measura of space and time, are imold conceptions of the known processes of small for most It is only for the enormous velocities connected nature. with the orbits of electrons in the atom, and for the
1
The
variability of
case
mass
is much
more
important
here
than
in the
the velocity of orbits, because of actual planetary far surpasses that of planets or fixed the electrons in the atom
stars.
FURTHER
CONCLUSIONS
that the theory leads to
a
87
deviation
fiand cathoderays,
from
on
the
bodies
a
as
matter
enormous
in favour
of the theory
of relativity. Finally, we
must
Einstein stated to 'of the theory result of relativity. As above mentioned, it has a velocity the mass of a body is greater when z; than when it is at rest. Mathematically formulated, be stated thus : If m" is the mass the matter can of a body at rest and ot" its mass when it has the velocity v, by the motion the increase in m^s {m^"m") caused is equal
energy possessed by the body with the velocity v, divided by the square of the velocity This last magnitude to an immense amounts of Ught.i figure in the usual phjrsical units of length and time
to
the
kinetic
900 trillions; thus the increase is immeasurably small for the usual velocities
body
If the velocity of the same possessed by bodies. be again increased, say from v to 2V, its mass again increase by an amount kinetic energy (producedby divided
given by the increase in
velocity)
by the square of the velocity of light. Hence is proportional to the increase in the increase in mass
kinetic energy. Now Einstein
1
was
able to show
of
mass
an
"
.
We
know
^
mass
i,.
m
i
with
"
velocity
wti'
_,
IS
given "
by
"'2
vK
The
mcrease
of
is
therefore
This
/
2C'
is exactly formula valid only for velocities (v) small compared to relativisdc mechanics, the kinetic with c, because, according
energy
for very
large velocities is no
longer given
by
"'.
88
SPECIAL
THEORY
OF
RELATIVITY
increase
mass
increase in the an of kinetic energy produce of a body, but that every increase of energy produces If, for instance, energy such an increase of mass.
to a body, an of heat be conveyed win occur, and this increase of mass
in the form
of mass in the
increase vsdU,
as
be of the kinetic energy before mentioned, divided by the equal to the heat energy absorbed, square of the velocity of light. A direct examination body before and of this law (sayby weighing the same
case
be performed on after the heating) cannot account of the infinitesimal smallness of the effect. Certain considerati however, have led us to suppose that the
theory of relativity will perhaps
best explain
mental funda
the problem of chemistry, namely, problem of the deviation of atomic weights from whole numbers. As is well known, the atomic weights, e.g., of carbon, nitrogen, and oxygen, are almost 12, 14, and 16 times
greater, respectively, than the atomic weight of hydrogen. do not agree quite exactly ; on the But these numbers
less than i contrary, divergencies of somewhat per have been it cent, undoubtedly established. Now be regarded as fortuitous that the ratios between cannot
the each
and
even
atomic
weights
dose 16.
Why
numbers But the theory of relativity If we suppose, supplies us with a possible explanation. for instance, that a Catom consists of 12 hydrogen
atoms,
any
(atomicweight 4),or of of 3 helium atoms of these constituents, then the comarrangement bination of these constituent parts to a single atomic
or
FURTHER
nucleus
amounts
CONCLUSIONS
the rearrangement
is'bound
to involve
small
of electrical energy, from which wiU On the basis of these variavariations of mass. tions of mass required by the theory of relativity,
account
we
can
from
whole
numbers
of atomic weight ratios. At the end of last chapter we that it is mentioned to choose the units of length and time in convenient such a way that the numerical measure of the velocity If we use these natural units of light is equal to imity. in what foUows, then the proportionality factor between increase of energy and increase of mass the (i.e. square of the and we Increase velocity of Ught) will also be equal to unity, law in a simpler way. then formulate our may
of energy
by an equal is always accompanied (Thisdoes not in the least alter the former assertions, because the to our
to a body in the form of heat, etc., energies conveyed if expressed in these natural units, are infinitesimally small.) Now before the introduction of the theory of known that every body, whether hot relativity, it was
of energy. cold, always possesses a certain amount This consists of heat energy, stored within the body, together with the energy of chemical afiinity (suchas
or
but probably that released in the process of combustion), for the main part of enormous amounts of energy situated in the inside of atomic nuclei, and hitherto not rendered of the radioactive elements. evident except in the case We
cannot
say, in
a
contained
only
measure
for example, the total what to, as litre of coal gas amounts which By
energy
we
can
become
free in
chemical
analogy
substances
might
expect
90
SPECIAL
THEORY
Now
OF
the
RELATIVITY
theory
of relativity is equivalent the plausible the
is very teaches
to
an
increase
assumption, total
mass
immediately
by
Einstein, that
is equal to the energy stored within it. Mass and energy, according to this view, become identical. And, as a matter of fact, justfor these two
of
body
of analogous the law of the structure are found to have validity ; i.e. conservation of mass, and the law of the conservation Let us suppose a system of bodies surrounded of energy.
on
magnitudes,
two
fundamental
laws of nature
all sides by
an
impenetrable
nor
neither radiation
of the
conservation mass of all the bodies contained in this envelope remains constant, whatever process they may undergo amongst themselves, in the way of chemical reactions, explosions,
or
which allows heat to pass through it. The law ma^intains that the total of mass
envelope,
combustions, regard
etc.
Exactly
the
the
same
is maintained
with
to energy.
Within
envelope,
chemical
be transformed into thermal energy, and energy may this into mechanical energy ; but the total sum of the
According to the energies always remains the same. fundamental laws of theory of relativity, these two and energy are one nature reduce to one law, for mass thing. and the same
In order
to
energy, and to avoid First of word, let us analyse the idea of mass. empty all,we must state that we have to deal here with a dual idea, which has been treated in physics as a single one of
mass
and
which plays only, owing to an accessory circumstance the second important an r61e m part of this book. In general, the mass of a body is measured with a
FURTHER
balance
;
we
CONCLUSIONS
the
91
determine
force with which gravity i.e. we it with the force compare The result of on the miit of mass. the gravitational The idea of the inertial mass different from this. It is the
this weighing
mass a
can
therefore be designated
of
body
resistance offered to acceleration, i.e. according to the law of Newtonian fundamental mechanics mentioned beginning of this chapter, the quotient between at the force
know We from and acceleration. experience is for all substances always prothat the inertial mass portiona to the gravitational mass body ; hence, if one
is twice
as
"
as
inert
"
as
another, maintain
we
mean
it must
heavy.i
Now,
if
mass,
we
that that
energy
involves
certain inertia and a certain weight. to a body in the form energy is conveyed body will become
"
heavier and more We thus arrive at what at firstsight appears a startling an evacuated result, namely, that even space which is be said to have transmitting energy can weight and inertia.
Thus,
if
we
completely
evacuate
vessel
it (supposing were
of the gas molecules), by dectromagnetic by be permeated radiation (e.g. lightrays, if we are dealing with a glass vessel situated or if this is not the case, in a hghted room, at aU events by heatrays, which
are
at the alwajre present, even But as every kind of lowest attainable temperatures). if it energy, so every vessel, even radiation transmits
'
This
is by
no
means
selfevident,
as we
and
does
not
follow
from in
the
definitions
the second
shall
show
in detail
92
SPECIAL
not
;
THEORY
any
OF
tangible
even
RELATIVITY
contains substance, interior of the empty
does energy
vessel
contain
consequently
the
possesses
gravitation and inertia. This result but it is interesting to note paradoxical, Viennese
before physicist Hasenohrl, developed, and by starting the theory of relativity was different considerations, finally off from completely result, i.e.that inertial mass must arrived at the same
be associated with heat radiation in an empty space. " leave the If we units of measurement, natural to the usual C.G.S. system once more and return of be stated thus : The units, the law in question must
"
energy pUed
of the velocity of Ught, 900 trillions. This figure is stupendous, and it takes breath away to think of what in one's might happen if the dormant to a town, energy of a single brick were be set free, say in the form
suffice to
raze
body
is equal
to its
mass,
mtiltii.e. by
ground.^
as we
know
radioactive
quantities of energy are only liberated with extreme slowness, and of atoms are entirely uninfluenced by human agencies.
enormous
It would Dreadnought
'
imllions
1000 m.
of
battleships
of
the
END
OF
PART
PART
THE
GENERAL THEORY
II
OF RELATIVITY
CHAPTER
XII
ON
INERTIA
led
AND
to
GRAVITATION
the
NECESSITY
Theory
of Relativity.
and the principle of the constancy the velocity of light are held to be right, there can deductions further choice ; all those no made Einstein
follow by compulsion,
as
of be
by
in
mathematical
the
theory
problem. To proceed
did not
seem
with
the
development
a
of
absolutelynecessary from
was
consistency during force here at The main motive the years 19071915. Einstein's philosophical perception ; he saw work was his new theory, and above all, the clearly that even
perseverance and
Newtonian
theory
of gravitation
(accepted without
any
modification up to that time) still possessed all those deficiencies of a philosophical nature, that have been by a number of philosophers clearly and keenly criticised in the
course
"
without
any
of
94
GENERAL
being
THEORY
OF
RELATIVITY
What upon the theory. able to improve we these deficiencies were, shall show in the following.
them At the beginning
"
of this book
we
made
the assertion
to talk of a relative motion It is only reasonable is unbodies ; to speak of absolute motion between reasona because it cannot be proved." When we
"
we meant uniform motion," rectilinear motion. said is not in general, however, For motion this statement motion valid, because the existence of a nonuniform be detected can quite well, without considering the
into brought since inertial forces are surroundings, If, for instance, a train be suddenly stopped, we play. fail to notice this distinctly ; in a railway cannot
collision the inertial forces called forth by the change fatal. According may prove motion absolutely
of
to
Newtonian
to the also according and of the special theory of relativity, these mechanics effects do not depend solely on the relative accelerations tions. of the bodies to each other, but on their absolute accelera
mechanics,
is to
say
if there
were
railwaytrain
nothing else, i.e.if there were nothing itself would the motion relatively to which it moved, not malre itselffelt,but every change of motion would. and In starting phenomena
or
in slowing
occur
down
as
would
Now this means relatively to the earth. less than that the idea of absolute nor
which
"
more
against
successfully
the
comes
theory
motion
relative to absolute space is not compatible with reason, to the principle of and is not perceptible according A change of motion with respect to absolute
just as
incompatible
with
reason,
ON
INERTIA
AND
GRAVITATION
95
^^nd
yet it is to call forth noticeable effects ! However tions, theory, based on such foundacorrectly the Newtonian terrestrial aU astronomical explain and it was phenomena, unable to satisfy the philosophical like Einstein. and scientificframe of mind of a man
could
The
matter
had
been
further
believed special theory of relativity. in an aether as taking the place of absolute space. It might have seemed reasonable to physicists and philosophers that accelerated motion relatively to an cether
be able to call should actually exist) would forth forces of inertia. But after giving up the idea
by
the
(ifaether
a
of
substantial
was
sether, owing
not
to
of
relativity, it
believing
should
caU there is no
experiment and the stars. that forces of inertia should act in an accelerated railway train we supposing it to be quite alone in the world
"
the other hand, for believing this. In order to make of this kind we cannot get rid of the earth Hence, if our intelligence cannot grasp
On
"
to believe it ! Let us compelled take advantage of this fact and recapitulate for this case what experience teUs us on the one haiid, and reason
are
not
in any
way
on
the
other body quite alone in the universe would have no inertia. This last statement is evidently equivalent to the that the capabiUty assumption, of a body to exercise inertial forces {i.e. is caused to possess inertial
is accelerated relatively to other : If a body bodies, inertial forces are called forth; a single
mass)
solely by the presence of other bodies in the universe. inertia, according Hence to this mode of thinking, is not something appertaining to every body of itself;
96
GENERAL
THEORY
OF
RELATIVITY
it is caused, on the contrary, by the interaction between it and the other bodies of the universe, just the weight as of a body is caused by the interaction of the body and the earth. As shown driven thought
to be
an a
in
the
foregoing
developments,
we
are
to these
opinions merely on the grounds of the to plausible imagine inertia process ; it is more interaction of the kind described, than to believe
mass
on
that
own
may possess inertia! single body fact based on One important account.
its
experience,
lends additional support is the fact of the arguments, inertial and gravitational mass.
which
to these purdy
abstract between
and are gravitational mass essentially two perfectly distinct ertial notions, but that according to our ejqjerience the inmass of a body is always proportional to its gravitational
we
explained
This fact of experience occurs in Newtonian law, which mechanics as a perfectly independent has nothing to do with all other laws. The mathematical
mass.
classical mechanics would remain if this law did not claim vaUdity. completely unchanged One a could, for instance, imagine priori that for different substances the inertial ratios between differ from each other, just and gravitational masses as they differ for the specific gravities of different substances.
foundations
of
the
heavy
and
size. Let us suppose iron and platinum to of the same different specific gravities, but have equal specific balls of the same two inertia, so that size, one of iron, offer the same platinum and the other of resistance
to
change
of motion.
This
would
have
been
quite
ON
INERTIA
AND
GRAVITATION
97
possible from the Newtonian point of view, without interfering in the least with the validity of its mechanical fundamental law. This law states: The product of the inertial mass and the acceleration is equal to the abovementioned supposition were fulfilled, the gravitational force acting on the platinum ball would be three times as great as that acting on
force.
If the
to be are since the inertial masses ball would have to fall to earth equal, the platinum with three times the acceleration of the iron ball. Thus between if the law of the proportionaUty inertial and
gravitational
mass
did not
hold
good,
different bodies
"
i.e.in considering air resistance vacuo) fall with different velocities. But that is not the would as we case, can easily convince ourselves with the simple
(even without
in an tube. guinea and feather experiment evacuated between Furthermore, inertial and the proportionahty been has by mass accurately gravitational proved Eotvos'
exactness
e3q)eriments,
of
oooooi
which
were
carried out
with
an
per cent. This empirical fact being known, filed it, but of, registered and made
of it !
to the apply it at once If inertia and gravitation considerations of this chapter. intimately related by the law of proportionso are aUty,
We
can,
however,
the conception strengthen discussed above, according to which inertia and gravitation interaction of bodies. are caused by the mutual In point of fact, we shall see in the following chapter that, in the hands of Einstein, the empirical fact of the this wUl naturally
proportionaUty between inertial and gravitational mass, left imused by physicists for two centuries, became the key to the generaUsation of the theory of relativity.
CHAPTER
XIII
THE
EQUIVALENCEHYPOTHESIS
shall begin
WE
motion
moment
the generalisation of the theory of relativity by discussing the question : Can imagine that the existence of nonuniform we
escapes
our
that it is no motion ?
more
For
the
hopeless, for any change of motion forces of inertia, and these must always reveal produces to us the existence of such change. How curious then that Einstein says : Inertial forces are, of course, always present in nonuniform motions, but their presence does
not
us
necessarily
to conclude
that
change
of
We
simply
persuade
gravitational
to distinguish gravitation
and
illustrate this by
ourselves
the following
example
to
move,
in
lift justbeginning
i.e.
an accelerated rectilinear motion performing upwards. We notice the acceleration by the fact that the pressure bodies on the floor of the lift is sUghtly greater of our fall to than usual ; a body released suddenly would
the
more on a quickly ; a weight suspended ground would etc. spring balance stretch the spring more, From a physiological point of view, the matter becomes
98
THE
more
EQUIVALENCEHYPOTHESIS
99
striking when an accelerated motion is performed If a Kft starts moving downwards downwards. quickly, the force of inertia is opposed to the gravitational force,
and thus diminishes it. Our body appears to be lighter in weight, and, provided the acceleration is suf"dent, we notice a peculiar sensation in the region of the stomach,
as
whilst
other
behave
:
differently
slowly ;
weight
be slightly relaxed, etc. But this would cdl take place in exactly the same would way if the lift reason were or at rest and the earth's gravity, for some suddenly stronger or weaker. of gravitational intensity do, as a matter of fact, take place at the earth's surface, for we are subject to the simidtaneous the earth, the sun, attraction of other, were Fluctuations
and
to
become
this combined effect differs at noon that in the morning and evening. and midnight Since the attraction of the earth far surpasses the other forces, these fluctuations are too small to be felt directly the
moon,
and from
by
distinctly noticeable indirectl in the phenomenon of the tides. Let us Suppose the earth to be so near the sun as to tional enable us to notice the daily fluctuations of the gravitaforces acting at the earth's surface, and let us
our
bodies,
but
they
are
waking up after a long sleep in a liftshut lamp in the interior. We off from dayhght, but Ut up by a suppose the observer to have an exact spring balance by the intensity of the force him, with which he can measure imagine
some
one
If he determine that the of gravitation at any moment. a small tension, he wiU say : shows spring balance " Gravitational force is small now, and since I know that it must now be noon." this is always the case at noon,
100
GENERAL
THEORY
has
OF
RELATIVITY
"
This up, says : need not necessarily be the case ; the intensity of gravitation may possibly be very great at the present moment, but we be moving downwards may with an acceleration."
just woken
may suppose the Uft to be moving in a shaft many kilometres in length, where the accelerated motion be kept up for some length of time.) We see from can
(We
the conversation of the two passengers indeed be doubt the existence about
that there
of
can
accelerated doubts of also ; the question is as to whether motion this kind arise from an insufficient knowledge of facts,
is again the general principle of nature cause of our inabiUty to decide, without reference to the surroundings, which of the two observers is right. We must path as in the first part of this pursue the same
or
whether
book,
when
we
were
discussing
the
special theory
of
relativity.
We
uniform without
stated there that the existence of rectilinear is not discernible by our senses motion
considering the surroundings ; we then went on to say that, by the most exact measurements and observatio in the range of mechanics, we cannot discern
; and
finally, we
extended
the law of relativity to all physical processes. That our two liftpassengers are not able to decide by experiments, viz. by weighings, pendulummechanical and fallobservations, which of them is right, is owing between law inertia! and to the of proportionality which is the kejmote of the general it is not valid, theory of relativity. Let us suppose (a possibiUty indicated in Chapter XII) and assume gravitational
mass,
is three times as that the specific gravity of platinum large as that of iron, and that their specific inertias, on In that case all doubts could the other hand, are equal.
THE
immediately
EQUIVALENCEHYPOTHESIS
be allayed,
as
101
to
whether
the
lift
was
downwards or not. moving motion with accelerated The observers in the Uft would have to replace the iron ball ^which we was will suppose originally hanging
"
on
"
^by
weight).
can
into accoimt.and the platinum ball will strain the spring exactly the same But if the lift move downwards as the iron ball. with
be taken
the spring consists effect on force, diminished by the value of As we have supposed the inertial force acting upwards.
accelerated motion, of the gravitational the
the latter to be weaker for the platinum ball, the tension In of the spring would here be greater in consequence. the same way, pendulumand fallphenomena with
turn out substances would accelerated liftand in a lift at rest.
various
differently in
an
This, however,
between
is not
the
case
the law
of proportionality
inertial and gravitational mass holds that the and its vaUdity guarantees
possibly
an
decide
by
mechanical
accelerated
motion
exists
or
arises
t (analogouso
the
special
theory
can
of
be arrived at.
of uniform
an
empirical
us,
nature
we
had
experiment).
now
when the
began
of
more
general
problem
consideration,
102
GENERAL
THEORY
data
no were
OF
available.
RELATIVITY
Hence, when
no
whatsoever could turn downward out ated with accelermotion and in one at rest, but situated in a weaker gravitational field, he thereby entered the realm of hypothesis, ^whereas the entire structure of the special
experiments differently in a Uft moving
"
he
preeminently a rational worldng Those readers who have followed the foregoing developments carefully will be able to how plausible this very illuminating and appreciate hypothesis
supposed
must
have
that
been
a
to
Einstein.
Could
it be
possible
on principle of relativity) for all physical processes, and its generalisation, on the so a purely apparently necessary from other hand, in demand theoretical point of view, and already
"
law
by
only for mechanical processes, philosophers, as well ? and not for electricaland optical phenomena consist inConvinced that laws of nature cannot contain encehypothes of this kind, Einstein set up his Equival
many
"
at
that
time,
facts. be
though not, explained presently), under the compulsion of any direct empirical Subsequently, experience proved him to
(tobe
entirely in the right, as we shall show in the next For the present, we the shall formulate chapter. In the second chapter of this equivalencehypothesis.
we
book
designated
act at
any
space in which
electric or
netic mag
forces
magnetic
in which field. We
point as an every electric or a field. Analogously, we shall call every space gravitational forces are at work a gravitational understand
by
homogeneous
field a part of
space in every single point of which the gravitational direction and the same intensity. force possesses the same
THE
Every
human
EQUIVALENCEHYPOTHESIS
dwelling
on
103
the surface of the earth can be looked upon to a close degree of approximation as a homogeneous gravitational field, since the variation of the magnitude and direction of the gravitational force from point to point within the walls of a house is
infinitesimally small. Furthermore which inertial forces are present, an lift, for instance,
motion, Now
an
with
accelerated
that the two passengers in maintains the liftcannot decide by any physical experiments whatever, whether the decreased tension in the spring balance is caused
of gravity, or by a downward acceleration of the lift. With reference to described, the ideas of inertial and gravitational field just formulate this statement as follows : With we can
a
Einstein
by
momentary
diminution
respect
to all
physicalphenomena,
homogeneous
an
CHAPTER
XIV
CURVATURE
OF
RAYS
OF
LIGHT
FIELD is
IN
"
GRAVITATIONAL
THE
in
a
equivalencehypothesis
bridge between
how
in
tation. of relativity and the theory of graviTo find the laws of physical processes homogeneous gravitational field,we must calculate these processes take place in a uniformly accelerated
the theory
to the
to
hj^othesis
place in
take
cases.
hypothesis
by
simple
lift moving constant velocity, and upwards with horizontally outside a imagine ray of Ught to move During the lift and enter it through a hole in the wall.
by the hght to required a traverse the chest, the latter moves short distance so that the lightray strikes the opposite upwards, lower than the hole. To the wall at a point sUghtly thus of the lift the path of the hghtray passengers
the minute timeinterval appears
to
(This phenomenon
astronomers
as
inclination of the path naturally increases with the velocity of the chest. Now if the chest possesses a uniform acceleration, its velocity
The aberration.)
CURVATURE
OF
RAYS
OF
LIGHT
105
will increase with time, and the inclination of the ray of light after traversing the chest will be greater than at The so that it describes a curved entrance, path.
inference is clear : Rays of light describe curved paths in an accelerated system, and since an accelerated system is equivalent to a system at rest in a gravitational field, it follows that light will suffer curvature in a gravitational field.1 Rays of light will be curved downwards,
the attracting mass ; the path of a ray of to the path of a light is, therefore, similarly curved is so infinitesimal owing to bullet, only the curvature
"
i.e. towards
magnitude of the velocity of light, that we determine the deviation from caimot rectihnearity in the earth's gravitational field. The matter is different,
the
enormous
however, Einstein
in the far greater gravitational field of the calculated that a ray of Ught travelling
sun.
just
Umb would suffer a deflection of iy".^ past the sun's influences our The way this circumstance astronomical observations is illustrated in Fig. 4, in which the
stellar distances
are
immensely
reduced,
whilst
the
deflection of rays of hght is exceedingly magnified. E is the earth, S a star, and H and H' respectively the As long as the sun in two different positions. is sun
sufficiently distant from the connecting line ES, rays of light travel practically in a straight line, but when the sun arrives at the position H', they are transmitted in the slightly curved the
"
line SPE,
the
star
earth
The
sees
only result in the curvature considerations of field. Calculations lightrays in a homogeneous gravitational in any is the case however, that this teach us, gravitational homogeneous or not. field, whether
foregoing
In
Chapter
XVIII
supplementary
note
will be
added
on
this point.
106
GENERAL
THEORY
OF
RELATIVITY
in order to verify if Einstein is right, we should, for instance, have to photograph a zodiacal constellation
Hence,
the
sun
then is in
again at another
the
sun
The posipart of the sky. tions of the stars on the two graphs photowill not quite coincide, but
will, according
to
Einstein's
thesis, hypo
very
small
indeed
(theyamount
on
to only
used
where
the the
effect
was
finally observed
mical accuracy of astronois great enough measurements to determine the effect in question
(~\ ^^^
H
IT irfi"
The difiSculty lies certainty. the starred elsewhere ; in general heavens in the vicinity of the sun with
cannot
be
the
photographed glaring
at
all,
sxmlight would blurring of the a cause complete during the long time negatives of
because
I
E
Fig. 4.
ejcposure
necessary
to
receive
the plate. picture of the stars on It ^as therefore necessary, in order to carry out observations, to wait for
can which starphotographs by night. The first total
of the bending of solar eclipse after Einstein's prophecy lightrays in the sun's gravitational field took place in
August
1914,
war
broke
out, and
owing
CURVATURE
to
OF
to
RAYS
over
OF
LIGHT
though
107 full
this it had
be
passed made.
unused,
next
Two 1919. to were equipped, under the leadership of Eddington, take the necessary photographs ; one went to Sobral in Brazil, the other to the Principe Islands near the west In both places the photographs taken coast of Africa. during the solar eclipse were successful. A few months later, when the
sun
The
had
moved
sky, control plates of the same instruments, and then the necessary measurements same tion The results confirmed the defleccould be performed. predicted by Einstein. The reader will readily judge what the result of this for the Einstein theory ; it meant successful prophecy signified the last Unk in the chain of proofs for the of rays of light by the amount
validity of
conception
once
First of all
we
had
to
do
With
the
special
referred only to relativity which valid for all rectilinear uniform motion, but which was Then, for reasons of a theoretical processes of nature. a generalisation of were we nature, obliged to demand
theory
of
this principle for arbitrary motions this generalisation for mechanical in the form
also ;
we
processes
of an equivalenceprinciple, since we in this by the empirical fact of the were supported proportionality between inertial and gravitational mass. The extension of the equivalenceprinciple to allphysical processes
was, a at first, hypothesis
only ; there
was
no
of that kind. actual experience to compel an assumption do possess Since the solar eclipse of 1919, however, we in this matter ; we find a natural empirical knowledge
108
GENERAL
which
THEORY
is not only
OF
RELATIVITY
explicable on the basis but the of equivalenceh57pothesis, exactly which coincides with Einstein's predictions ! It is true, those opposed to the theory of relativity tell phenomenon
us
that the deflection of rays of light in the sun's tational gravifield might be otherwise explained, as for instance by the refraction of rays of light in the atmosphere Of course it is always of the sun. possible to
by explain an established natural fact subsequently hypothesis invented ad hoc, but Einstein's original some explanation, based as it is on the compulsion of profoimd thought,
the
sun
Moreover, the exwill be given preference. planati of the deflection as due to the atmosphere of is to be completely for other reasons.
rejected
If the
sun
be needed to cause would deflection of light, certain other phenomena be observable, but this is not the case.
as
atmosphere
of such
immense
CHAPTER
XV
THE
RELATIVITY
OF
ROTATORY
MOTION
BEFORE
and thought
once as
proceeding,
to
more,
we
convenient
take
so as
shall find it necessary train of up the fonner not to lose the general
The theory of relativity survey of logical connections. fiction the idea of absolute tends to banish as an empty is necessary to ehminate space from physics, hence it
both imiform This leads and accelerated motion. rectilinear motion to the further conclusion that inertial forces can only meaningless
for accelerations which are relative to other appear bodies of the universe, and not for absolute accelerations inertial mass ; in other words, the of a body is
"
"
caused
to
us
by its interaction similarly to the gravitational mass with all other bodies. These considerations lead Its validity for the principle of equivalence.
from the beginning mechanical processes was guaranteed by the empirical fact of the proportionality of inertial its validity for optical processes by the successful prewas proved subsequently diction of the deflection of lightrays in the sun's gravitational field.
to
gravitational
mass,
and
for the results of observations made to work during the solar eclipse, Einstein continued confidence in the truth of his theory, with unbounded
Without
waiting
X09
110
GENERAL
that it
was
THEORY
completed
OF
RELATIVITY
so
splendid
structure
for the new room edifice. The make Einstein's train of thought has been made how it was continued will be best made the motion. know We problem
begiiming
dear
of
of the
rdativity
that centrifugal forces appear in rotatory Newton that in the case motion, whence concluded of is reasonable. rotation the idea of an absolute motion devised to show a He this experiment wellknown
empirically : A bucket of water is set into quick rotatory does not partake Owing to inertia the water motion.
but is gradually set rotating of the motion immediately, by friction with the sides of the bucket, until the whole mass rotates of water velocity as the with the same bucket. As
soon as
this is the
in the form of a concave mirror ; water particles rise up the sides of the bucket under the influence of centrifugal forces. At first,when the sides of the bucket rotate, but not the curved
water,
the surface of the water (as remains quite even being a sure ascertained by Newton), ^this proof that no He argued thus : centrifugal forces are then at work. At the beginning of the experiment the rdative motion
"
the sides of the bucket and the water is greatest, and yet no effect is perceptible. Afterwards, however, when there is no relative motion between the bucket and the water owing to the water partaking of the rotation,
between
hence
they
on
must
depend
on
and
not
relative rotatory
RELATIVITY
This motion. first,but when
maintained, Newton's
"
OF
ROTATORY
seems
MOTION
111
as
at plausible enough submitted to rigid criticism it cannot be He says : stated distinctly by Mach.
conclusion
experiment
teaches
us
that
appear for the rotatory motion of the water relative to the sides of the bucket, but they appear as the result of rotation relative to the mass of the earth and the say how the experiment turn were out if the sides of the bucket would made increasingly thicker and more massive, up to, say, several
can
heavenly
bodies.
Nobody
nules known
Only
we
this
one
must
make
of
nature
and
with
our
arbitrary
fictions." arising from the to be proof earth's rotation, and considered by Newton In this case of the absolute existence of this rotation. We
now
proceed
to those phenomena
centrifugal angular
cannot
forces
are
so
minute,
owing
to
the
small
Velocity
(one revolution
on
our
own
doubt. proved without Moreover, their effect appears in the fact of the earth's bodies These centrifugal forces act on oblateness.
can
be
the earth's surface. There is another t3rpe of force, also caused by the earth's rotation, but acting bodies in motion relatively to the earth's surface. on
at rest
on
These
Coriolisforces," and are manifested called freely along in deflections suffered by bodies moving the earth's surface ; these deflections are to the right on the northern hemisphere, in the direction of motion
are
"
and
to the left
If a prothe southern hemisphere. jectil it will for instance, be shot due southwards,
on
112
GENERAL
travel quite
THEORY
OF
RELATIVITY
not
the time
the
Coriolisforce tradewinds
the
following
greater
wear
the
motion)of
the
righthand rails of railway tracks, the greater wearing down river banks, the rotation of the of righthand in Foucault's pendulumplane pendulum experiment,
and view, "the
so on.
Considered
from
the
Newtonian
point
statement,
of
prove
an
that
the and
to
absolute
earth
at
system
round it. Let us again hear what Mach has to say on the subject Let us consider the point on which Newton : concerning the appears to lean with full justification,
"
distinction between
relative and
absolute
the earth performs an absolute rotation their appearance, centrifugal forces will make become flattened, the acceleration of gravity equator
will be diminished, the will be turned, etc. pendulum if the earth be at rest, and will disappear heavenly bodies rotate round it absolutely, way that the
same
relative rotation takes But if so, if we start from the idea of absolute space. keep to the basis of facts, we can we only speak of All motions in the relative space and relative motion.
Universe
are
according
relative to each other, both to the Ptolemaic and the Copernican system. the
same
RELATIVITY
if
we
OF
ROTATORY
MOTION
113
notice of the unknown and unconsidered Both are medium of the Universe.^ systems equally practical. right, only the latter is simpler and more
no
take
The
earth
Universe
at
given
us
twice
over,
with an earth in rotation, once, are and with its relative motions which alone Hence determinable. be we cannot say how it would
rest and
if the actual
earth
case
did
not
rotate.
We
but if
can
interpret
the
in different ways,
to experience,
our
we
interpret it in
mode
of interpretation
The fundamental
be regarded in such a way result also for relative motions." Thus the conflict between the Ptolemaic earth
at
S5rstem
rest)and
the
to
Copemican
earth)
is, according
Mach,
(the (rotating
tljeories
maintain nothing essentially different ^Iheyare merely fact. different interpretations of one and the same In this Mach which clearly sets up that programme was turned into account about thirty year^ later by
Einstein.
this, it was card In order to accomplish necessary to disNewton's with its ideas of absolute mechanics, his theory of gravitation. According acceleration, etc., and the Ptolemaic to Newton, sjretem is not only more
inconvenient quite
impossible.
reconcile the Ptolemaic system forces and CorioUsforces with the fact that centrifugal do not act on the stars act at the earth's surface and
*
Mach
means
by
discarded,
however,
has already this the lightaether, which by the special theory of relativity.
been
114
GENERAL
THEORY
OF
RELATIVITY
scattering them really rotate round the earth), into space ? A relativistic mechanics and theory of No : (i) give us the following answer gravitation must
"
they (if
perceptible centrifugal forces appear in the fixed stars, for acceleration relatively to no more nothing calls forth inertial forces than rotation relatively to ing nothforth centrifugal forces,^ and the mass can call of
"
"
"
"
Coriolisforces
the
must
be
looked
as
forces
overthrow
Newtonian
in view of
mechanics
(whichwe
of
were
the second his theory of gravitation, for acwe overthrow cording law of gravitation,^ the gravitational to Newton's and relativity), bodies
acting on each other depend on distances the masses of the bodies and their mutual Thus fixed apart, but not on the state of their motion.
as stars considered revolving act, according to Newton, with
forces between
stars regarded
as
earth would no other force than fixed being at rest, i.e.with no force at all,
on
a
round
our
for the
"
fixed stars
the
average
case a
are
uniformly
dis
Since rotation
special again
of nonuniform
only
special
case
motion, of inertial
two
It
runs
thus
The
gravitational
for
force
acting
between
to
duct proinversely to the square their masses and proportional of if the earth were Hence, twice as far their distance apart. of as is actually the sun the case, distant from the attracting force
would three be only onequarter of that actually existing ; if it be oneninth, times as far distant, the force would
were
bodies
(sun and
earth,
instance)is proportional
the
and
RELATIVITY
tributed
round
our
OF
ROTATORY
and
MOTION
115
solar system,
of gravitation
must
be
so
the
rotating
gravitational centrifugal
its formulae
and
truly
in mechanics be must general relativistic law of motion in such a way as to admit of inertial forces constructed only for relativeaccelerations, rotations, and so on.
CHAPTER
XVI
THE
NOTION
OF
SPACECURVATURE
AND
OF
WORLDCURVATURE the
the last chapter mark
we
IN
out
more
have
explicitly pointed
Einstein's speculaat by aimed tions, finally attained by him. A was which detailed presentation of the theory which realises here
set
the
demands
up,
can
only
be
with the help of higher mathematics.^ some, help of any mathematics, at least, of the most be elucidated, characteristic traits of the theory can this is best carried out by the application of the special theory of relativity (as far as is reasonably
and
of the rdativity of rotatory problem In the place of discussed in the last chapter. motion huge circular disc let us suppose a our solar system freely poised in space, and so thin that it exerts only
to possible)
the
above very small gravitational forces. Immediately this disc and concentric with it, we suppose a second The centres of both discs disc of equal magnitude.^
are
"
supposed
In
to be connected
theory
by
an
axis, round
demand
which
the
general
accessories physicists. acquired by mathematical " For the sake of simplicity and clearness
"
the
we
speak
of
top
and
bottom
"
disc, though
"
we
"
are
"
aware,
meaningless
to talk of
top
and
xi6
bottom
"
SPACEboth
discs
can
AND
WORLDCURVATURE
The
lower
117
rotate.
disc is to be at rest
relatively to the system of fixed stars, and the upper is to rotate relatively to it. We one to imagme are both discs inhabited by intelligent beings, provided with including, of course, of phjreical apparatus, For the sake of brevity measuringrods and clocks. Reds we will call the inhabitants of the top disc the
aU
sorts
"
"
The Reds Whites." those of the lower disc the will observe the occurrence of centrifugal and Coriolisforces on their disc ; if they have acquired the rdatiand of thought explained in the last chapter, vistic mode they will know that the existence of these forces can be interpreted in two ways, either as inertia! forces, ifthey
"
"
consider their disc in motion, or as gravitational forces by the revolving firmament exerted of fixed stars, if they consider their own disc to be at rest. (If, indeed, their
a
through
in
had
would perhaps not be tions. able to realise that there exist two different interpretaence Perhaps their theory does not recognise the differ
via GaUlei
Newton,
they
between
by
the
way.)
This be in
communication with the Whites, and always to compare their time and lengthmeasurements with those The Reds living in the of their lower neighbours. immediate vicinity of the axis will have a small velocity situated below them, and in relatively to the Whites ence the effects of lengthcontraction and differconsequence
of dockmotion those living at the wiU be immeasurably of the and clocks of both Whites and Reds wiU measuringrods But it is different for those living practically agree.
centre
118
GENERAL
the periphery
THEORY
of the disc.
OF
To
RELATIVITY
plainly speak more that the diameters
near
we shall assume concretdy, equal to the diameter of of the discs are approximately the earth's orbit (300,000,000 km.),the angular velocity The velocity corresponding to one revolution per week.
and
of a point at the edge of the top disc relatively to the lower disc will be about 5,000,000 or 6,000,000 km. per With these velocities hour, or 1500 km. in the second. by the special the contractions in length demanded theory than
of relativity
XTT
P^^
to less
and
the
different motions of the clocks will also be noticeable, if the results of the special theory of relativity (which derived were expressly only for rectilinear uniform
motions) are
motion.
apphable transference
at
all to
A
to
of the
theory
moving nonuniformly Umit the contemplation to very small permitted if we i.e. to smaU spaces and short times. worldelements, a relatively small Thus, if we contemplate region near of the upper
the
case
the edge
the size of the for a short and follow its movement earth's surface) the motion time, e.g. a few minutes, of this r^on region on the lower disc relatively to a corresponding will be practically quite uniform and rectilinear. Hence can we apply the results of the special theory of relativity with a clear conscience. like all other According to this, the measuringrods, belonging to the marginal regions of the top
disc
(say about
objects
disc when
wUl appear in the same
must
from
the
lower
stationary
disc,
in the
the motion
compared
be slower
of clocks
SPACE
AND
WORLDCURVATURE
Measuringrods
119
of the Whites themselves being perfectly equal, and in agreeamong ment with those of the Reds who inhabit the vicinity of the disc's centre, it follows that the lengths of measuringrods and the motions of clocks will be different for the Reds inhabiting the margin ing and for the Reds inhabitWe know the centre. that contraction takes place hence of motion, only those measuringrods of the Reds will be shorter which are laid down to tangentially (parallel the disc's edge); only in
the
direction
to radially (perpendicularly the disc's edge)agree with the corresponding rods of the Whites, to measure When therefore the Reds and Whites come
(inour
the
of their discs, both will arrive at the same km.). But if they 300,000,000 example
of their discs, they will arrive at different results, for the Reds are measuring and will have to lay with shortened measuringrods
drcimiference
them
down
oftener
to
get
round
the
disc than
the
Whites.
For them, the resulting figure for the circumference of their disc will be greater than for the Whites.^
on
VII. An explanatory p. 49, Chapter be added in Chapter XVIII, p. 153. will 2 the following this conclusion Against objection has
'
Gf. footnote
note
only will measuringrods but also the entire circumference suffer contraction, the measuringrods, disc, and in the same as proportion
been
"
raised
Not
laid down
tangentially in the direction of motion. figure for the length of the circumference for the Reds as for the Whites." This
runs
Hence
must
however, objection,
draw from one the cannot conclusions good, because the circumference of the disc special theory of relativity about As mentioned the results of this theory as a above, whole. be applied to cases can only motion, of rotatory when very dealt with. See Supplementary are small spacetime elements
not
hold
Note
on
p. 166.
120
GENERAL
THEORY
OF
RELATIVITY
the ratio between the circumference and of the circle will be a different one for the
for the
Whites.
The
ratio
"
and
this is
remarkable
^will be different for circles of different If the Reds living at the centre of the
a
relatively small circle (witha radius of the ratio only a few kilometres) and then detennine between diameter, they will find circumference and number Ludolphnumber
same as
the
the
Whites
.
the {i.e.
wellknown
once
314159265
letter
.,
designated
and
ir),
only shghtly contracted. intermediate zone a of the top disc draw circle, the diameter is about diameter, of which equal to half the disclarger number they will find a somewhat than TT, and the inhabitants of the periphery of the disc will find still greater number, since the measuringrods laid down by them tangentially will be contracted most
a
of aU. The
inhabitants
of the
top
to
ratio between circumference and diameter of circles will always be equal to ir for the Whites, quite independently of the magnitude of the
as circle (just for the Reds
we
learned at
this school),
only approximately, and those circles of dimensions small as compared with the disc on which they live. Deviations from this law are greatest for circles comparable the disc itself. The Reds
use are
with
might objection
wrong
in their measurements,
contracted
measuringrods
at the
SPACEif reproached
"
AND
WORLDCURVATURE
121
replying
we
:
are
in be justified with this, they would According to a general principle of relativity fully justified taking up the point of in
view that we are at rest, whilst the lower disc and the fixed stars move Hence for us there need us. around be no such thing as a contraction of measuringrods ;
such
and
as
they
are,
they
lead the results derived these measurements for us, us' to that geometry, which is the right geometry because it fitsin correctly with our experience." We of shall be able better to appreciate this mode thinking if we leave the Whites and Reds for a moment Let us turn to fiction again the earth. to be beings of only two dimensions,^ and suppose men from the earth's can who neither raise themselves surface nor penetrate into the earth ^theidea of a third and
return
to
"
give from
us
correct
measurements,
dimension
itself to
to
or
the
mere
them.
To the
them
"
that
earth's
a
not
curved, strange
of curved
to them.
and straight lines would, of course, All great circles on the earth's surface
(for
instance,
be the equator) would meridians and straight lines to them, i.e.lines continuing in the same A line, however, which first ran in a northdirection.
the
south direction, and then turned be a curved line. west, would determined
'
more
and
more
to the
This
by
these
twodimensional
The
have thin
is, of course, height ; we should missing third dimension flat beings somewhat like infinitesimally to imagine these in a horizontal leaves of paper position gliding along the
earth's surface.
122
GENERAL
earth, is for the
are
THEORY
two
OF
RELATIVITY
northsouth and Surface curvature,
curvature
the
dimensions, disposal.
eastwest,
at
their
however,
different matter.
The
be described thus : We earth's surface can a tangential plane at that point of the terrestrial globe in which we are situated (a horizontal ; then we
of our imagine
plane)
touch
earth's surface vanishes beneath the plane (aship receding from us on the ocean below the horizon).Now, if the ideas of disappears
" " "
this plane,
whilst
the
above
and
below
"
are
strange to twodimensional has no significance at aU to them, and the curvature the earth's surface would be quite inconceivable, or
beings least not obvious. Nevertheless, twodimensional would be able, given sufficient progressive development,
to
arrive by
abstract
mathematical
"
reasoning
"
at
the
to the surface curvature conclusion of attributing To understand inhabited by them. this we will suppose them to begin measuring the ratio between circumference for circles drawn on the earth's surface. and diameter
with the circumfere again, for circles small compared of the terrestrial globe, the wellknown it would result, but the ratio would be less than number This is plausible for circles of greater magnitude. IT Here,
enough
Let
us
imagine
the circle in question to be the The real diameter of the circle beings is the chord AB (Fig. 5),
line through the earth's interior connecting two between diametrically opposite points A and B of of the circle. The ratio between the circumference v. this circle of latitude and the chord is, of course,
i.e. the
The notion
of the earth's interior cannot exist, however, beings. They for twodimensional of the only know
SPACE
AND
WORLDCURVATURE
123
opposite
to
greater than that hence the ratio of the circumference of the circle to the line ANB is smaller than the ratio of the circumference
each through
by
us
to be the
were
"
beings
to
determine
resulting from their experience therefore contain the following law concerning would the circumference the ratio between and the diameter This ratio depends on the magnitude of circles : of Geometry
"
the circumference the ratio between and its diameter, the resulting number
of the would
the
and
circle ; for small circles it reaches the hmit for larger circles it decreases, reaching the value
tt,
124
GENERAL
a
THEORY
a
OF
of
RELATIVITY
km."
for
circle with
diameter
20,000
maticians Mathe
beings would the twodimensional amongst be able to show that it is possible to imagine a fictitious the circumference surface for which the ratio between for all circles, of their magnitude, and, furthermore, for which this ratio that surfaces could be imagined than on the earth, e.g.so as to be equal would vary more Lastly, they to 2 for a diameter of i km., and so on.
its diameter
is equal
to
ir
show that surfaces are imaginable for which this ratio increases with increasing diameter of the circle, larger than ir.* i.e.becomes Mathematicians would go distinguishing these to discover that the quaUty on might
various surfaces from each designated as "curvature," with other can for they
of science and simply by calculation, beings know from we threedimensional merely what t surface (plane)he ratio observation : On an imcurved
the
help
between
of
circle and
its diameter
is constant
the
ever, equal to n ; with curved surfaces, howthis ratio varies for circles of various magnitudes, more the greater the curvature so (supposing,f o that the diameter is always measured along the
course,
surface
1
itself).
are on
Saddlesurfaces
curved line
surfaces
an
of this kind.
Let
us
imagine
in ordinary ridingsaddle, and drawn that the shortest distance measured such a way along the saddlefor all surface from the centre of the saddle is exactly the same line would be a circle for twothis line. Such a points on
closed
" " the saddlesurface, and for these circles is greater than t. the ratio between circumference and diameter the ratio between Surfaces where circumference and diameter " " the value ir are exceeds called, in mathematics, circles of
dimensional
beings
on
surfaces
with. negative
curvature.
SPACE
AND
WORLDCURVATURE
125
lead to the following : It is considerations that we are, in point of fact, owing to the circumstance have been abld at a threedimensional beings that we
to early stage (in times of antiquity) acquire the knowledge of the earth's spherical shape. This circumstance, however, was not a necessary condition for the definite attainment of this knowledge.
These
comparatively
Mankind
he had
would been a
have
been
Uke Gauss, for instance, had been great mathematicians cal at his disposal, in order to take the necessary geometrimeasurements.
But of the
in
this
case
the
knowledge
never
earth's
the
pubUc
have
inconceivable
more
remained We
"
or
It would have and tminteUigible. less the sole property of the mathematically
to
now
ask
"
the
question
How
threedimensional
ajid all the stars ?
space,
position
to
the
fictitious twoearth's
are are
We
carmot
a
surface.
not not
able to
able to
imagine
and
we
judge directly
space possesses curvature or not, i.e.whether space is in three dimensions what a is in two dimensions. ing Accordcurved surface or a plane to the foregoing explanations, it is clear that only whether mathematicians
can geometers It will be as follows
and
as (designated
to
our
geometry)are
exactly vahd
any magnitude,
in the universe
we
must
126
GENERAL
THEORY
OF
RELATIVITY
spaces space is
no as
curvature.
(Mathematiciansdesignate such
Euclidean
spaces.)
or
On
"
the
other
"
hand,
be
and diameter, for instance, is for very large circles, or the "" sUghtly sum of the angles of very large triangles is not exactly between circumference different from equal to 180". This must be looked
"
upon
to
clear conception thereof would be as futile as the analogous attempt to give fictitious twodimensional
a
space beings
"
an
attempt
"
beings
of the surface on of the curvature conception ture which they live. Hence, when we talk of spacecurvain the following, we mean more concrete nothing
a
from
EucUdean
measurements
geometry
of the
of this kind might be found the universe had been foreseen nearly a on measuring like Gauss and Riemann. century ago by mathematicians It had become clear to them that the validity of the
The
laws
accepted on the contrary, they saw like asort that other systems of geometrical laws could be set up, differing from logical contradictions free from and
of
EucUdean
could
not
be
Euclidean
geometry.
Finally,
that recognised which of these is suitable to properties of the universe in of fact, Gauss large triangle determine performed
they
with
sides
experimentally actually
amounts
whether
the
sum
of
the
angles
SPACEto
AND
WORLDCURVATURE
no
127
deviation
was
tectabl de
teach
us
such that
could only negative attempt, of course, deviations from Euclidean no geometry the limits of exactitude of our measuring
or
words, that the curvature of space, if it exist at all, must be very slight. It certainly be maintained in no way that such a curvature can be determinable does not exist. It might at some later date with
on more
in other
much
small
(saythe
the kilometres),
EucUdean
aforementioned
their certainly geometry would elude Now, as we out at the end measurements. pointed X, the spatial extent of the scene of our of Chapter in small number activity is given by a comparatively
units of space, and as a matter of fact, the portion of space inhabited by terrestrial beings is an infinitesimally small part of the visible stellar universe. the
natural
"
"
it is quite possible that we are here in the same beings whom position as the fictitioustwodimensional Hence supiposed to inhabit only a small part of the earth's discern the curvature cannot of the space surface ; we Uve in, because all our experience and measurements we
we
refer only to a very small portion of the whole universe. In the discussion of the preceding we paragraphs to have digressed from our theme, and the reader seem has all this to do with the may perhaps ask : What
and with gravitation ? This will relativity of motion The Red : be made clear in the following statement inhabitants of the disc which is rotating relatively to
128
GENERAL
THEORY
arrive
OF
at
a
RELATIVITY
through geometry they of the space
geometry.
the
fixedstar system
Euclidean
long ago considered possible by mathematicians has been reaUsed. The space in of which they Uve does not partake of the character
space; space, but of curved, nonEuclidean This does not mean that the surface their circular of disc is to be thought of as curved upwards or downwards Euclidean That this is not the case they shallow bowl. could readily determine, for if they be threedimensional beings themselves use they can the third dimension
Uke
a
greater than ir, as observed by the Reds.) On turn contrary, the results of their measurements if the whole threedimensional as space in which
not
they
work
embedded
were
cannot
imagine)in which
justas
and
in threeis both
which
visible and imaginable. Let us proceed to consider the following : All those by which the top disc are processes on phenomena,
distinguished from
those
on
(theappearance
of centrifugal and
are spacecurvature),
disc rotates relatively to the firmament of fixed stars, does not. In the foregoing chapter whilst the lower disc
we
saw
clearly, in the
sense
of the MachEinstein
con
SPACE
AND
WORLDCURVATURE
129
of those forces could be ception, that the appearance interpreted in a double way : Either as inertial forces, if we or as gravitational consider the disc as moving, forces exerted by the rotating fixed stars, if the disc be
considered
at
rest.
The
same
can
be
curvature
be an of space ; this too may rotation of the disc, or an effect of a gravitational field due to the rotating fixed stars.
We
were
have
now :
got
to
see
the
point
a
towards
which
we
a
steering
We
that
^avitational
field, of
create
of space.
tional of this special kind of gravitaciiose the example in this case field, because the appearance of
can
be
made
plausible
without
the
The mathematics. mathematical formulae of the theory, however, teach us more than this ; not only the special gravitational field dealt
with here, but every gravitational field causes The gravitational field of our of space.
curvature
sun, a
of the certain
in the universe
causes
This
be
of space, characteristic of the fieldin question. is so slight, that it could not however, curvature, hitherto
determined
by
our
available
means
of
measurement.
great mathematicians prediction of our fulfilled in accordance has been with the Einstein differently from what had theory, though somewhat
Thus
the
been
sUght
What they was conceived mately approxiexpected. the following : The universe in itself has a very (inthe sense of the definition aforecurvature
in
a
mentione
we
similar way
as
the surface
on
which
live possesses
slight curvature.
That
the
presence
130
GENERAL
THEORY
OF
RELATIVITY
their
was
(thefixed
this
stars and
curvature
planets)
hardly
dreamt
of
by
Einstein.^
According
Ues thus : of relativity the matter At a great distance from all gravitational masses, space is almost exactly Euclidean space ; in the vicinity of however, it is curved, the curvature gravitating masses,
depending
masses
imagine i.e.as
as
a
force exerted by the in it clearer, we shall as twodimensional, the universe for a moment Our picture would then be somewhat surface.
on
the
fixed wide regions lying between stars, the surface of the universe would be almost exactly plane, but in the vicinity of every single star there follows
:
In the
shallow convexity, in the midpoint But as the curvature of which the star would be situated. the largest stars is very small, near of space even these convexities would be too slight to be discovered
would
be
sUght
eye, if we had a truetonature model of with the naked before us. this worldsurface The considerations of this chapter referred to the
" "
of space, called forth by a gravitational no field ; there was question of time in connection with Minkowski Now that according to the them. showed
curvature
special theory
of relativity, space itself plays only the Just as the shadow of a body is part of a shadow. different in magnitude according to the surface on which
it falls,so
the space
taken
up
by any
objectis different
of the system following more
to this
to
to the
geometrical
1
which formulation
exception
is equivalent
;
in meaning
seems
With
a
one
foieseen
Hiemann
have
SPACEstatement
:
AND
WORLDCURVATURE
a
131
surface is only a twodimensional part of threedimensional space, so space itselfis not an independent part whole, but only a threedimensional
Just as
of
the
fourdimensional
world.
sense
(For
those
readers
who
we
In explairing the idea of the curvature of space we said : turn out as if the whole The results of measurements in which they threedimensional (the Reds) space
"
work and
and
carry
out
embedded
in
their measurements were curved fourdimensional space (which we the twodimensional earth's
in A threedimensional
cannot
imagine),just as
to
embedded
us
is both
Minkowski,
the
following
plausible enough : Do the ideas of play a similar and space world," in a certain sense, earth's part in the theory of relativity, to the ideas in classical physics and geometry space and surface assumption
" "
appears
"
"
"
"
"
explained : In prerelativistic that the earth's surface was a times it was assumed in a nontwodimensional curved manifold, embedded i.e. (Euclidean)threedimensional manifold, curved
Let
this
be
By analogy, is it not permissible space. that the space in the vicinity of gravitating curved threedimensional manifold
now masses
to
say is a
non
curved fourdimensional manifold, This last sentence is correct in all but one word ; the be omitted, as applied must noncurved attribute According to the results to the fourdimensional world. of Einstein's calculations, not only space, but the entire
" "
spacetimeentity,
called by
Minkowski
the
"worldj'
132
GENERAL
be
THEORY
as
OF
RELATIVITY
must
regarded
is, of
course,
still more
itself,because one fails significance of spacecurvature " between to see the ideas of tirrie any connection " Hence we curvature." will confine ourselves to and
a
Four
IX)
required in nature
ambiguity : the three spatial coordinates relating to the place of the event, and one coordinate the event states the point of time at which which happened. As there explained, the spatial distance without
be calculated by means pointevents can law (47thproposition Pythagorean of the wellknown from the differences of the of the firstbook of Euclid), three spatial coordinates (difference height, length, of between
two
and
breadth). Formerly
this
spatial
distance
was
independent of supposed to be an absolute magnitude, to the theory of the system of reference. According culate relativity this is not the case, but it is possible to cala
with the help of a generalised magnitude law, and using all four differences of coPythagorean ordinat is designated This new the magnitude
"
"
of the
^
"
footnote at the end of Chapter IX for the special theory) is contained in "world
if the laws contained in this worldare analogous to those of planegeometry, geometry for the simplest description of physical phenomena,
geometry."
Now
"
'
Just
as
which
and
deals of
with
corresponding
spacegeometry
(stereometry)for
course,
threedimensional
"
talk of
"
worldgeometry
SPACEand if they
are
AND
WORLDCURVATURE
188
also analogous to those of the stereometry of Euclidean space.^ we then say that the world is Euclidean. When this is not the case, we say that
the world is nonEuclidean (curved).According to the special theory of relativity the laws of the worldgeometry hence the Minkowskiworld still Eudidean, is Eudidean. But according to the general theory, this is no longer valid for parts of the world surrounding
are
hence worldcurvature gravitational masses, there, in the abovementioned sense of the word.
1
exists
Such
law
is that
of
a
used
the
IT,
drcumference quite
independently
before ; The ratio of example is always equal to circle to its diameter of the magnitude of the circle.
as an
CHAPTER
XVII
THE
NEW
THEORY
OF
GRAVITATION
THOSE
Before
readers who, before the perusal of this book, knew the general theory of relativity to be at the same time a theory of gravitation, may
do we perhaps be disappointed and will ask : Where find the explanation for gravitation in this ; why do all bodies attract each other according to Einstein ? this question we must answering consider the following : To explain a phenomenon to trace means it back to a simpler and more general phenomenon. If
was we were
to
which have to
third phenomenon,
manner, answer
"
in this
Why
do
I fall on
"
? tramcar we may answer moving to your inertia ; your body retains its motion after leaving the steps of the car, whereas your feet are suddenly friction with the ground. to rest by brought
jump
off
That
is why
:
questioning
we a can
"
give fundamental
no
further law
reason,
of nature.
admit
There
are
certain ultimate
of
no
NEW
further
these
THEORY
;
OF
they
GRAVITATION
185
explanation
is gravitation : AU be explained, and needs no This fact cannot ; it is simpler than any other phenomenon it might that
we
exist. Amongst simply bodies attract each other. tion explanato which
be reduced.
cannot
This leads
us
to the conclusion
expect a theory of gravitation to give We expect the us an explanationof the phenomenon. theory, however, to describe the phenomenon tion. of gravitaThis
must
description
must
be
quantitative,
i.e. it
possibihty of calculating of a body, for instance of a planet, The gravitational action of other masses. theory of gravitation has done this in a very
with
the
and there would and unequivocal manner, to depart from this theory, did it not show
be
no
those
pecxiliar theoretical defects referred to at the beginning of the second part of this book. To
the defects of
we
must
According
gravitation possesses an infinite The meaning of this will be velocity of propagation. clear in what follows.' The gravitationeil force made
to consist of the attraction planet is known by sun (forming far the greatest part of the entire force)and that of the other planets. These forces depend on the mutual distances of the celestial bodies,
on a
add a to Newton,
acting of the
in the course of time, and wiU be constantly changing Now in accordance with the positions of the planets. that the force acting at a given Newton maintained to be calculated a on say Jupiter ^is planet moment from the instantaneous constellation of the attracting
"
"
masses.
If, however,
gravitation
possesses
finite forces
of the
186
GENERAL
a
THEORY
on
OF
RELATIVITY
have
to be
acting at
Jupiterwill
not
must
in our precise moment formula, but rather other distances corresponding to an earlier instant, i.e.as much earlier as gravitation needs
that
to travel from
those planets to
Jupiter.
formerly the part of physicists, great doubt was cast on the conception of forces propagated to a distance Even Newton himself on one with infinite velocity.
that he could not believe in action occasion admitted Seen from the point of view of that kind at a distance. of the theory of relativity, this conception is utterly impossible. For,
as
On
expounded
in Chapter
V,
one
of the of be
fundamental
the Newtonian
an
the velocity of science, with its progressive beyond was at last obliged to advance theory ; but this theory will ever remain
velocity than for all time.
immortal
work
It
was
the firsttheory
to obtain an exact treatment enabled mankind belonging to natural sciences, and in the of problems future it will remain for aU practical purposes almost the
which
as
theory
of
the astronomer.
of gravitation and Newtonian mechanics be looked upon as the model type of a mathematical can we description of natural phenomena, will first of all
it to exemplify the nature of a description of this kind, and then show how the corresponding description is given by the Einstein theory.
use
The
Newtonian by
means
theory
supplies
we
the
can
apparatus
of which
NEW
gravitational
THEORY
force due
OF
to
GRAVITATION
137
given configuration of at any point in the neighbourhood attracting masses, of This accessory is called Newton's law of these masses. footnote at the end of Chapter gravitation (cf. XV).
any
The description of the motion executed by a body under the action of force is further given by the Newtonian fundamental law of mechanics, which states : If no
will persist in a state If, however, rectilinear motion. of rest or of uniform it, acceleration wiU ensue. The direction a force act on of the acceleration will be parallel to the direction of the force, and the magnitude of the acceleration will be
on a
forces act
body,
that body
equal to the quotient of the force and the inertial of the body. These
laws
were
mass
brought
by Newton
of
differential equations, and in point of fact, by their aid it is possible to calculate the motion carried out by a body under the action of given forces, or vice versa,
a the forces necessary to impart certain How to a body. state of motion great the efl"ciency by the history of the disis shown covery of this theory was,
to
calculate
discovery which is a planet Neptune as one with full justification of the greatest
"
of human
science,
The
French
astronomer
forces of the that the combined hitherto known planets did not suffice to fully explain There remained a small the orbit of the planet Uranus. the motion discrepancy between calculated, and that
Levenier
had observed
actually
performed.
planet
Leverrier
assumed
cause
that
new
undiscovered
and
might calculated by
be the
means
of
where this planet ought to revolve in order that the forces exerted by it would justbe sufficient to explain
138
GENERAL
THEORY
OF
RELATIVITY
the disturbances
As a result of this matheobserved. matical thetical analysis he was able to predict that this hypoplanet would be visible at a certain time, and position in the heavens, and,
as
at
certain
matter
of fact, the planet (whichwas afterwards called Neptune) discovered by the Berlin astronomer Galle in the was
We
now
propose to show
of what follows. which is necessary for the comprehension The position a point in space is given mathematically, of in Chapter IX, by three numbers as shown three co^its ordinat
"
the motion of a point is described unequivocally, if its position is stated for every possible Expressed in mathematical terms, of time. moment this
must
means
Further,
The
of
a
that the value of the three spacecoordinates be given for every value of the timecoordinate. formula representing the solution mathematical
matical problem of motion must theref pre be a certain mathedevice, which how to calculate the us shows for any value of the timecoordinate. three spacecoordinates This device may also be given in a graphical instead of by calculation, i.e.the description of a
be given by a diagram. This is may motion for instance, in the socalled graphical timethe case, by a glance at a charts, which enable railway officials, drawing, trains
us on
way body's
to survey
a
the instantaneous
by
which
shows
a
the
motion
along (particle)
a
of a massstraight line. We
a
vertical
NEW
THEORY
The
OF
GRAVITATION
139
straight line OX
is divided into
equal parts representing seconds of time, and the straight line OY into equal parts representing cms. We then draw vertical straight lines through the division points of OX
and describe the motion of a particle as follows : A mark is made on the straight line going through the divisionpoint I by the
OX
traversed motion
after
Fig. 6,
is made mark going through the division point from OX traversed by the
began.
Another
on
the
straight
line
more suppose the drawing the position accurately made, for instance, by marking for every tenth or every hundredth of a second, and further, if we imagine all the marks connected together,
seconds, and
so
on.
If
we
we
obtain particle in
line which of the represents the motion vertical direction. If the particle move
140
GENERAL
THEORY
OF
RELATIVITY
the path traversed during the vdodty, with unifonn first second will be equal to the path traversed in the The corresponding line will then next second, and so on. be a straight line. But if the motion does not proceed In with uniform velocity, then the line will be curved. Fig. 6 the line A represents motion with uniform velocity
downwards,
C
and
us
the line B
shows
accelerated
other
similar motion upwards. downwards motion and D retarded The straight line E, on the
particle at rest.
a
"
These
or
a
"
lines
represent the story of motion of point, and are designated the particle. make The individual
particle
marks
Any up the worldlines are called worldpoints.^ be can say in celestial mechanics, special problem, considered to be solved, if the worldlines of the planet
or
comet
in question
can
are
known.
only be represented in a plane if the motion of the particle in question takes place in one in dimension only (fi,g. a straight line).In the case of
The worldlines
of a particle (e.g.otion motion m does not sufi"ce to represent in a a plane drawing circle), the worldline, and a threedimensional model is needed.
the
twodimensional
Let
a
us
of the motion
of
in
drde
the surface of the paper a mark that point of the cirde which contains the particleat the Then a mark is put 2 cm. high above the time I second.
'The is nothing the graphical else than worldpoint (Cf.Chapter IX.) pointevent." of a
"
sentation repre
NEW
THEORY
OF
GRAVITATION
141
position in which the particle is situated at the time The connecting line of all these marks 2 seconds, etc. then becomes a a screwline, which resembles spiral
Now the particle runs spring. along a threewhen dimensional curve the (ascrewline, for instance), worldline becomes fourdimensional a can no curve, and longer be represented by a model. In this case it must suf"ce to describe the motion by means of mathematical formulae only, but even so, the corresponding values of the
are
three
one
timecoordinate
the of
sum
and
total of
the worldpoints
"
belonging
to the motion
one
particle,
as
an
exact
description of gravitational
processes
us
if it contains
calculate the
unambiguous rules that liable fluence worldline of bodies under the in
This is justwhat the of gravitational masses. is fundamentally Einstein theory does, and in what a by a suitable generalisation of the very simple manner, abovementioned not of a body
states that
a
Newtonian
law
state of rest
or
of uniform
at rest
or
The worldlines of a body rectilinear motion. in uniform rectilinear motion are straight lines ;
translated
"
of
the Newtonian
not
by any
masses
gravitational
body never free of a move can present, however, forces, because it is always beiag acted upon by gravitational force ; hence there can then be no straight That Einstein's agrees very well with worldlines. assertion, according
to
which
the
world
in the neigh
142
GENERAL
THEORY
OF
RELATIVITY
bourhood
no
There are is curved. of gravitational masses It is impossible, straight lines in a ciurved manifold.
to
an
draw
exact
only in straight
one
all the other points point, whereas Hne would he outside the spherical certain types of unique lines, exactly straight Unes, can nevertheless
"
surface has
not
"
lines, since they show the straightest least deviation from a straight line, as compared with all in the last We mentioned other lines on that surface. chapter that the meridians and the earth's equator would appear
to be straight lines to fictitious twodimensional
these lines, and what have they in common this question, we real straight lines ? To answer follows : If two points are given consider what plane,
an
infinite number
of lines
can
be drawn
on
that
But of all these, point to the other. plane from one the one straight coimecting line is distinguished by being be defined the shortest. The idea of a straight line can
directly in this way : It is the shortest connecting line between two points of a plane, or, more generally, of a Now if, on the other hand, two Euclidean manifold. points are given on a curved surface, they cannot always by a straight line which lies completely connected in the surface, because straight lines can^iot, in general, But here again, of the be drawn on a curved surface. be
numerous
lines that
can
be drawn
hne will always points, one that is the line designated above
two
NEW
THEORY
OF
GRAVITATION
148
In mathematics, lines of that kind are called geodesic lines. On a spherical surface, the geodesic lines are the great circles {i.e.ircles, the diameters of which c
are
of the sphere, such as the ; meridians and the equator in the case of the earth) a plane on the geodesic lines are, of course, straight
equal
to
the
diameter
lines.
of geodesic lines, Einstein of simple law for the motion established a remarkably It a body under the influence of gravitational force. a body situated in a gravithus : The worldline runs tational
By
means
of this notion
of
law
case.
geodesic line. It is obvious that this law of inertia as a special includes the Newtonian forces are For in places where no acting and is field
a
where world
accordingly is Euclidean
no
gravitational
(not curved).
geodesic lines are straight lines, hence the worldlines become straight lines, and that is, as before mentioned, law of inertia in the terminology the Newtonian of worldgeometry. present, however,
lines will be
field is gravitational the world is curved, and the geodesic C and D in curved Unes like the curves
Wherever
Fig. 6.
It is, of
course,
necessary
for
to
complete
description
This law mentioned. curved by the presence of gravitational world becomes for geodesic lines naturally vary according to masses, This second law, however, can the kind of curvature.
law establish another very simple law of motion the tell us in what must way
formulae and not in only be expressed in mathematical designated by Einstein the These formulae were words.
"
Fieldequations
"
of Gravitation.
With
the
advent
144
GENERAL
THEORY
OF
RELATIVITY
theory of these fieldequations the structure of the new It can be charactercomplete. of gravitation became ised
in two
sentences
with
the
foregoing
of
presence
curvature,
gravitational
the depends
the nature
of which
by
world in such a way that their worldlines are In Einstein's theory the fieldequations geodesic lines. curved
part as the law of gravitation in play the same Newton's ever, theory ; the law of the geodesic line, howcorresponds to the law of motion of Newtonian
we
discussed
at all
spoke
of the problem of relativity, although the considerations of space and to that which led to the idea of curvature took their origin in the relativity of worldcurvature of rotatory motion. : Is justifiable the the defects Chapter The
new
theory as presented in of the Newtonian XII, and does it fulfilthe demands set up in Mach's considerations
at
the end
in
of
the the
not
entirely
theory
of the
{e.g.bsolute a
objectedto in do acceleration)
line is strictly vaUd only for the for the purof material points ; that sufQces, however, motion poses for the stars are always dealt with as massof astronomy, It would take us too far to discuss points in celestial mechanics. the laws of mechanics which hold exactly for spatially extended bodies according to the general theory of relativity.
geodesic
NEW
THEORY
OF
GRAVITATION
145
This theory at all in the Einstein theory. appear does not deal with forces or accelerations at all, but geodesic lines and with worldcurvature. only with
By
introducing
these
new
ideas
of their not being the Einstein theory complies with a far more general principle of relativity than that of the special theory in Chapter III. We said there : exactly formulated
disadvantage
"
In different systems
of reference
moving
uniformly
rectilinearly with respect to each other, all natural It take place in exactly the same way." phenomena is true we cannot and generalise this sentence say : In different systems of reference moving arbitrarily and
"
other, all natural processes take For if A and B axe same way." of reference moving acceleration relatively to
each
other
with
an
(e.g. rotating
with
ph5?sical processes will take place respect to each other), In the former example in different ways in A and in B. ball which receives of the two rotating discs, a smooth to roll on impact an with rectilinear will continue
disc, reference to the lower with reference to the top one, for, owing to the the ball will have an centrifugal force in this case,
uniform but not
velocity
with
acceleration. outward however, It is stillpossible, fact that only relative motion has
to take
into account
or
the
(whetheruniform
not)
be done by physical significance. That can into a form in which they putting the laws of motion With the Newtonian are valid for all systems of reference.
any
laws
not
this
was
not
the
case.
The
law,
"
body
acted
upon
of uniform
10
force retains its state of rest or " " rest rectilinear motion," is vahd only if
by any
146
"
GENERAL
"
THEORY
are
OF
RELATIVITY
certain "Ms" (called inertial
to
"
or
motion
spoken of
of relatively to
tinguished
sjretems"
systems in
as, physics),
the
lower
"isc of
however,
our
If the top
"
rest
or
motion,"
disc, the
must
law
is
no
valid, and
in
so
Einstein's
theory.
of
line is universally valid for all the fieldway systems of reference, and in the same of the world may be equations, by which the curvature
of the geodesic
given distribution of gravitating masses, for ss^tems of reference moving retain their form Furthermore, arbitrarily with respect to each other. calculated for
a
the laws of electricity,optics, heat, etc., can be accommodated to the new notions of worldcurvature, and into a form that is valid for any thus be brought
In this way the new theory of reference. systems fulfils a general principle of relativity which be can thus : Laws expressed of nature can be brought into
a
form
of
does
are
not
alter,
even
when
the motions
to referred
any
systems
of reference
at the
we
XV
of
be so constructed that, according to gravitation must firmament its formulae, the revolving of fixed stars a to gravitational field which is equivalent produces
that truly of the centrifugal and Coriolisforces. Also,
a
law general relativistic mechanical of motion be constructed in such a way that inertial forces must appear only for relative accelerations, rotations, and
so
on."
These
demands
are
as also fulfilled,
has been
NEW
proved
^
THEORY
OF
GRAVITATION
147
which
by direct calculation. The problem lishing of estabbasis, the complete system of physics on a new is more satisfactory than the old system from a
therefore be said to philosophical point of view, can have been successfully solved by the Einstein theory.^
*
To
satisfy these
the
to accept
in Chapter
*
views XIX.
See Supplementary
Note
on
pp. 166167.
CHAPTER
XVIII
DEDUCTIONS
FROM
THEORY
THE
GENERAL
THE
place
new
theory
theory in its essential traits. Entirely introduced ; worldlines take the are
or and worldaccelerated motion, of uniform takes the place of forces ^inshort, we have to curvature different description of nature. deal with a fundamentally On the other hand, it was clear from the beginning
"
theory that, with reference to numerical results,any new theory only to a could differ from the old Newtonian For all calculations performed on very sUght extent. the basis of the latter theory almost of any absolute
new
and Einstein, in
that the establishing his ffeldequations, bore in mind tational resulting laws concerning the motion of bodies in gravifields must necessarily agree approximately with
1
those
of the Newtonian
it turned
out
theory.*
that
the
As
regards
the
Subsequently
fieldequations of the
approximately time just the
lead to formulae agreeing which general theory theory, are at the same those of Newton's with
148
GENERAL
THEORY
149
degree of approximation, the matter is similar to that of the special theory of relativity. Here, too, the deviations
from the laws of classicalniechanics and electricity are exceedingly small ; they are apparent only when very bodies are dealt with. When, therefore, swiftly moving falling stone, or the trajectoryf a o under the influence of gravitational force is projectile calculated, the results obtained according to the new theory differ from those obtained from the Newtonian
the
motion
of
theory
that it is absosuch infinitesimal amounts, lutely impossible to detect the difference even with the Only in strong gravitational very finest of instruments. fields are the differences between the results of the old
by
theories within the possibility of measurement. and new So far we know of three phenomena that are bound to turn out differently according to the Einstein theory would according to the older theories. One has already been discussed, and concerns the of them deflection of lightrays in the sun's gravitational field. than
they
To
this
we
must
we
Chapter that
a
XIV
add began
supplementary
our
horizontally outside a vertically accelerated chest, and entering it by a small hole, will describe a curved path with reference to the chest. If, according to the equivalenceh5^thesis, assume we
ray of light moving the path of
ones
ray of Ught in
formal
corresponding
gravitational
which,
for
mathematical
reasons, alone call for tells in favour of Einstein's is not composed of hjrpothesis
On the contrary, starting from the considerations ad hoc. dealt with in the last chapters, them into a and putting form, led necessarily to formulae are we which mathematical agree as well the Newtonian
or
better
with
experience
than
those
drawn
from
theory.
150
GENERAL
THEORY
OF
way,
a
RELATIVITY
then this indicates
ray of light describes the same path under the influence of gravitation as any material body travelling with the velocity of light, in in short, the lightray falls a gravitational field. or,
seen)
Now
for bodies in very rapid motion, the law of motion Einstein's theory differs considerably resulting from law ; in the present case we find that rays of light suffer twice the deflection in a gravitational field according to Einstein's theory that they the Newtonian theory, assuming, according to the Newtonian too, rays of light fall like that in this case, of course, Hence bodies. thus : the matter stands material would from
Conformably
the
sun's
is to equal
be
to
theory of hght, no influence of gravitational field on the propagation of light The deflection would have to be expected.
zero
to Maxwell's
that rays of hght fall in a gravitational field,the result arising from Newton's theory of gravitation is a deflection of the Umb to o85," amounting passing the sun's deflection of 17" results from Einstein's a whereas The observations of both British theory of gravitation. expeditions proved the latter value to be the right one. Ughtrays Another
However,
if
We good that
know
suited to the experimental examinati phenomenon of the Einstein theory is planetary motion. the orbits of planets to be represented with
by Kepler's firstlaw, which states approximation is a the sun planets describe eUipses, of which first empirically established This law, which was deduced Kepler, was theoretically subsequently Newton
focus.
by
by
was
from
his
theory
GENERAL
THEORY
151
fact in t)xe development As we have abready of physics. basis of the exact mathethe fmidamental mentioned, matical treatment of natural sciences by means of
the Now infinitesimal calculus was then laid down. this law of Kepler does not exactly agree with experience, and also from Newton's theory it only follows with exactitude when solely the attraction of the sun is taken
into account in calculating the motion of the planets. But all planets are acted upon not merely by the attraction of the
sun,
Fig.
7.
and when all these are taken into certain sUght deviations from the ellipticorbit account, disturbances of the orbit. termed are result, which One of these disturbances is the socalled motion the
of
periheUonof planets.
planet its position relatively to the system of does not maintain In fixed stars, but rotates slowly in its own plane. the ellipticorbit of a planet, one focus of Fig. 7 is shown S. (For the sake of clearness which contains the sun The the eccentricity of the ellipse is exaggerated.) but planet does not describe this orbit with exactitude,
The
ellipse described
by
only
approximately,
so
that
second
orbit does
not
152
GENERAL
THEORY
with the
OF
RELATIVITY
placed slightly disthird orbit deviates
a
coincide completely
from
some
first,being
with reference to it, and a the second, and so forth. After thousands through
a
of
tmned
as
in Fig. 7.
vertex
the (called periof the ellipse which lies nearest the sun heUon) has then changed its position from P to P'. The motion of the ellipse in its plane is therefore designated
the motion of the perihelion. less in Perihelion motions more or of this kind occur be explained according to all planetary orbits, and can
as
the Newtonian
theory
by
the disturbing
forces due
to
A discrepancy exists only for the planet other planets. viating Mercury, the observed motion of the perihelion defrom that calculated (fromthe disturbing forces) by
an
amount
motion
Now if planetary equal to 43' per century.^ be calculated from Einstein's theory, we
if obtain an ellipticorbit with perihelion motion, even the gravitational action of the sun only be taken into For Mercury this effect amounts account. exactly
to the observed too
small
value of 43" ; for the other planets it is (compared with the disturbances from the
to planets) nearest
it is very strong, and hence the Einstein effect of is noticeable only in the case of this perihelion motion particular planet. The new theory,
on
therefore, gives us the right result ever that very point where the old theory failed. Wherthe old theory, on the other hand, was found correct
measurements, of our within the limits of accuracy Einstein's theory leads to the same results.
'
See Supplementary
Note
on
p. 167.
GENERAL
THEORY
158
In order to understand the last of the three abovedeductions drawn from the general theory of mentioned to the example more must revert once relativity, we
We menof the rotating disc used in Chapter XVI. tioned that the clocks of the peripheral inhabitants of the top disc go slightly slower than the clocks of the inhabitants at the centre. of this statement formulate it more inhabitants
periphery
Let
precisely.
and
We
the
assume
that
the
inhabitants
of the
goes clock which carried out in such a way correctly. that the measurement of the velocity of Ught with the help of these clocks and by use of a standard measuringWhen the rod, must result precisely in the value c. standard inhabitants
receiving these signals, will state that the timeintervals between to their clocks, are the signals, according not exactly
of the rim of the disc give light signals or (say timesignals at equal timeintervals
according seconds of the centre, on
to
their
clocks),
This is the meaning but slightly more. that the clocks of the peripheral statement of our inhabitants go more slowly than those of the central " is the reason What If we inhabitants. are asked :
1000
seconds,
for the
answer
we different motiofi of these clocks ? must is the same as that for the contraccause : The tion
"
and the appearance of centrifugal of measuringrods are forces ; all these phenomena results of the relative the disc and the firmament of fixed rotation between
stars,
in other words, they are results of gravitational Hence, forces exerted by distant revolving fixed stars.
or
by
our
example
become
acquainted
154
GENERAL
another
THEORY
OF
RELATIVITY
effect of the gravitational field : clocks stationed at different points of the field go at different Mathematical further teaches us that treatment rates.
with
in this particular gravitational this is not only the case field, but quite generally, in every gravitational field. Furthermore, we can with the help of our example
discern the
manner
field takes
be assisted
we were
though
were
If, on the other hand, we going downhill. from the periphery towards the centre
the action
travelling
of the centrifugal forces would overcome, and the feeling would be that of going uphill. In the first case we should be travelling into
slowly, and in the second regions where clocks go more into regions where they go more Now the case quickly. equations of the Einstein theory teach us that this rule holds bound This Thus generally. to be accelerated effect, however Even is if
we
go
upMU,
watch
is
theory of
relativity)many
if
tour
one
small to
be perceptible.
a
succeed
of superhuman surface against its attractive force e.g. from the sun's to the distance of the earth's orbit, the effect would be far smaller than the usual daily fluctuations of our best
mountain
in accomplishing dimensions^
In spite of this, its observation is not chronometers. We impossible. XI that the pointed out in Chapter atoms of luminous gases emit rays of fight of quite definite i single lines (spectral lines)n Now a ray of fight the spectrum of the luminous gas. of a particular spectral colour is nothing else (according colour, which
appear
as
GENERAL
to
THEORY
155
the
considerations
wave
of
magnetic electroHence
we
consider an atom which emits sharp spectral lines to be a kind of clock, producing alternate positive and negative electric fieldsin its surroundings at regular
can
timeintervals.
goes
more
Now
if
an
atomic slower
clock
of this kind
slowly,
atom
i.e. emits
substance, of the same spectral lines will be shifted towards the red end of the lines of as with the corresponding compared spectrum
another
(Thisis
because
are
the
visible spectrum, whilst the violet rays are the quickest).On the other hand, we know that the frequency and colour of light are connected length, red rays having the longest with the wave length, and violet rays the shortest. Hence we emit light of greater wave say that retarded atoms In agreement with what
we
oscillations in the
wave
can
length.
said above, clocks the earth, on the sun will go more slowly than on " have to go because one against the uphill would in order to get from the sun to attraction of the sun Hence on the sun's the atoms the earth. surface, if
"
they
presupposed for very good will emit light of longer wave reasons), on the earth. length than the corresponding atoms theory of gravitation concerning The result of the new therefore be verified by comthe motion paring of clocks can
act
as
correct
clocks
(which can
be
length of solar spectral lines with the lines of terrestrial sources length of corresponding wave is so small, however, that it lies of light.i The effect
the
wave
must add that the lines have to deal here are not we with which of the solar spectrum lines ; this does not, however, lines, but absorption emission
'
For
physicist
we
156
GENERAL
THEORY
OF
RELATIVITY
The difference
and
measurement.
corresponding 80 biUionths terrestrial light amounts only to about theless, Neverof a centimetre for the spectral lines examined. the delicacy of our methods optical measuring of sunlight
is
wave
length
between
justsufficient to
permit
us
to perceive
the difference.
of the effect,these measurements are extremely difficult ; hence it has not been possible Whereas to arrive at finAl results up to the present. some observers maintain they have detected the effect, that it does not exist, so that we must others maintain regard this point as not yet ffiially settled. If we survey the results of the experimental tions examinaof the new gravitational theory hitherto obtained,
Owing
to the smallness
be said, in view of the verification of two of its results, that the probabiUty for the truth of the theory is very great ; stillit would seem premature to consider it as confirmed beyond doubt.
it must
alter the absorbing
atoms. matter atoms
the
same
period
way
as
CHA
PTER
XIX
THE
HYPOTHESIS
THE
OF
THE
FINITENESS
OF
UNIVERSE that
is not
THE
but
new
recognition
curved
space
Euclidean
up a of the is
universe. Euclidean,
to assume necessarily compelled longer Eire no we that our universe is infinite. But now bound to believe this. This can be best explained by a
twodimensional
Let us again suppose twoexample. beings on a smooth dimensional XVI), sphere (Chapter to inhabit only a small part and let us imagine them not would of the sphere, so that their measurements the existence of curvature. yet have revealed to them They thus believe the scene of their activity to be a to be asked whether If they were the plane surface.
surface
answer
is finite
"
or
infinite, they
be
would
our
It must
infinite ;
an
assume
ultimate
go
on
of the spherical shape of the by voyages round the world, or have gained knowledge they of something would quite unable to grasp previously, namely, the geo
extending." knowledge
the
158
GENERAL
fact that
a
THEORY
surface
case
OF
can
RELATIVITY
metrical bounded.
This is the
be finite without being with the surface of a sphere ; It is possible to travel over it
in any direction without arriving at a boundary, and yet it is not infinite. Such but not infinite, are are surfaces, which unbounded designated as closed smiaces. We imagine can quite and number of other surfaces possessing this property, as, for instance, eggshaped surfaces, ringshaped surfaces, On the other hand, the following are etc. unclosed
a
surfaces
cylindersurfaces, surfaces of cones, Only curved surfaces can be closed ; paraboloids, etc.^ a into infinity. on plEine has either an edge, or it runs Geometry known long before teaches us (and this was t Einstein)hat dimensional dosed, the
same
planes,
holds good
more
Now
take space,
A curved therefore be space. space can i.e.it can be finite,without having any limit. is curved, we must since our worldspace
"
"
into
account
i.e. our
certainly unbounded. According to Einstein, this assumption, which revolutionises the universe, is not all our views concerning
only that The for this is reason possible, but probable. infinite universe presents certain the idea of an difficulties,independently the Newtonian of whether theory of gravitation or that of Einstein is looked
upon
'
as
right.
Concerning
the
infinite universe,
the
There
a
is
here too
curves.
in the case of onedimensional analogy distinction can be made between closed and
an
figures ;
The
former
an
are
finite, but
unbounded
the
(a
its
own
tail is
instance) the ;
circle and
ellipse belong
to them.
FINITENESS
following two
OF
UNIVERSE
159
alternatives are possible a priori : (i)The entire infinite universe is filled with fixed stars in such a way that the average density of distribution is about
as
systems visible to us represent a kind of solitary island in the universe, whilst in the infinite regions beyond visible space the density of distribution of the stars decreases gradually things, the observed
to
zero.
Now
(As a matter of unfavourable fact, these do not retain really absolute fixed positions in the heavens, but rather travel to and fro like the
individuals of a swarm about takes place at
a
eye.)
the slowness of motion of the stars, that the gravitational forces exerted on each other by fixed stars are very feeble. This could not be the case, however,
an
if the
universe
were
equal or greater mean than in our surroundings. Other tell against the second alternative. objections to exist as an If the whole system of ficxed stars were
island in the infinite universe, this state of things could Rather to exist to all eternity. would not continue After aeons the the stars disperse gradually into space. starry sky would longer be visible in the surroundings sun sohtary ; every star would pursue its own of our ceivable path, severed from its neighbours by distances of inconEven the mutual attractions of magnitude.
no
not
prevent
them
from
dispersing,
as
160
GENERAL
be shown
no
THEORY
by calculation.
OF
Now
RELATIVITY
it is true
that
we
can
have
physical proof against the possibility of such a dissolution of the universe after trillions of years ; but instinctively driven to repudiate this eventuare we ality. be said that the islandhypothesis concerning our universe cannot be upheld scientifically, but we other expedient if shall gladly revert to some Such an be found. it can expedient is the above
Hence
it cannot
According to Einstein, Einstein hs^thesis. mentioned the firstof the two alternatives is the right one : on the average whole, the universe is filled with a uniform But it is not infinite ; it is a closed density of stars.
space in the
sense
chapter.
Hence
against the
We
can
first alternative
can
be
dismissed.
appreciate the idea of a closed universe by translating the whole matter into two dimensions, as We said did before at the end of Chapter XVL we
only extensive regions that lie between fixed stars, the surface of the universe would be almost but in the vicinity of every single star exactly plane, at the centre of there would be a sUght shallow hump
"
there
In
the
which
would
our
as
whole small shallow humps, having and is studded with many (Thisis no contradiction the stars as their centres. distances was to what said before, for the average
between
be the star itself." We must now ment supplepicture as follows : The world's surface taken is a spherical surface of immense extension^
parts
stars
fixed stars are very small compared neighbouring with the girth of the universe, and hence those neighbouring of the world's surface between
in point of fact, be looked
can,
plane.)
In such
way,
we
might
upon as imagine
almost twoa
FINITENESS
dimensional
would three
OF
UNIVERSE
; the
161
picture the of
to
be,
universe Einstein,
curved
above against the idea of an infinitely extended universe had nothing to do with the problem of relativity in itself. There is another argument,
in favour of the idea of a finite universe, intimately connected which is most with the idea of oped relativity. It is as follows : The considerations develin Chapter however,
is not according to whichinertia body in itself,but, like gravitation, is interaction of bodies, guided Einstein
XII,
in his task of establishing the equations of motion and the field equations. These comply with the general in Chapter XVII, principle of relativity formulated and
with
Mach's
demands
mentioned
the relativity of rotational motions. must analysis, however, shows that we interpret these
now
necessarily
completed equations, subsequently, as meaning that the inertia of a body is actually caused only by the interaction between it and the other masses There are many physicists who conof the universe. sider formulation of the new the mathematical theory correct, but who do not agree with the abovementioned
conception of the nature of inertia. The theory would then be robbed of its most
"
profound
general theory of relativity has been repeatedly threedimensional to, because spherical space of that be grasped by our as curved cannot space generally) well
The But
kind
objected (as
that
our
tion. imaginait is only fair to say that it is not Einstein's fault of imagination fails in this point. powet
162
GENERAL
THEORY
OF
RELATIVITY
conceptual
If it is to be in truth a completely nucleus. consistent theory of relativity, and not merely a mathe facts of the greatest matical description of astronomical be interpreted possible accuracy, the inertia of bodies must by it in
the
way
indicated.
Now
Einstein
is this interpretation that mathematically only possible if we assume universal space to be a closed " Hence if the radical relativistic world." spherical MachEinstein point of view be accepted, we shall have
showed
to beheve
the universe
to be unbounded,
but finite.
CONCLUDING
REMARKS
THE
the
purpose theory
connections
It wiU be useful, relativity. therefore, if we collect together in the form of a genealogical ideas discussed in detail table the genesis of the
in the text.
done
in the table
on
p. 167,
which hardly needs any further explanation has already been said. It is intended to map
after what
serve as a
of the regions of the special and general theory of relativity, for those who have lost their bearings in the mist of mathematical and geometrical difficulties. After faithfully following the discussions of this book the reader may throughout, judgefor himself as to the
or
otherwise
too
a
were
of the theory of relativity. If the to express his own opinion, it might perhaps cause StiU, we and mistrust. exuberant words objective
shall add
with
few
concerning the criticisms of relativity, thanks to its reputatio than any other richly endowed
"
physical theory. The truth of the If critics take the point of view : to merit theory is far from being sufficiently proved its creator our alongside Galileo or Newton," ranking against it, unless it be that a ifit has nothing of ideas can be admirable even structure to do with the reality of things. nothing
can
be
said
163
164
GENERAL
there
are
THEORY
many These
OF
declare
RELATIVITY
the whole
to
But
who
be
logical nonsense.
people either do not understand they are not clear as to the limits of
It has been said, for instance ray of light possesses the same
:
The
statement
that
velocity c with reference to two systems moving rectilinearly and uniformly with respect to each other is logically wrong." In this case the critic,though understanding
what the theory of relativity is intended to of what belongs to the convey, is nevertheless unaware realm of logic. In point of fact, the statement quoted
has nothing at all to do with logic ; it merely ujKets That cannot the traditional ideas of space and time. be denied.
the theory of relativity critic reproaches being the most and mathematically confused with That is to be explained dif"cult theory ever set up. third A
layman generally has not much average to do with higher mathematics, and he is quite happy standing about it. If,in attempting to find the way to an underenof the theory of relativity, he becomes tangled
thus
The
in mathematical
at
not
The other branch of mathematics. functions, etc., contain algebra, the theory of difficult than those chapter which is far more differential geometry
that form
many
parts of
and absolute differential calculus matics the basis of Einstein's calculations. Mathe
is verily no child's play I On the other hand, the theory of relativity is frequently overrated, as regards the extent of its importance. It supplies
us
with
new
in respect of
SUPPLEMENTARY
geometry,
"
NOTES
165
physics, and perhaps also of philosophical But it has nothing to do with what we call science. Worldconception in the human sense general
"
be interesting The Einstein may man of the word. from this point of view ; his theory, however, must not be mixed up with this.
of the theory is solely to approach the ideal of the rational description of physical processes as nearly be judgedat present, that as possible ; and as far as can aim purpose has been
The
fulfilled.
SUPPLEMENTARY
NOTE
PAGE
NOTES
TO 2
here used, and accordance with the method of treatment have already by the desire for brevity and simplicity, we guided ^ongside of each other, two statements ranged unhesitatingly the arouse perhaps criticism of philosophically may which feel the For those trained (and only for those) who minds. ideas involved, formulation we shall of the need of a stricter By following the the of note : motion supplementary add in the body a (in general, and not only abovementioned the alteration we of its position, understand restricted sense) is only given by its distance of a body since the position and is in its essence a bodies, the concept from of motion other does not Thus the first of the above statements relative one. In
new of opinion ; it is an analytical expression express anything to the idea belonging a that already property characterises This significance of the in virtue of its definition. of motion has been the phoronomic "motion" (kinematic) called word But the idea of motion we can regard of motion. conception " " by the in that we in yet another motion understand way, the presence a of. which might condition, physical of a body be established ence referalso without circumstances certain under is in a state of instance, if a body For bodies. to other by the be recognised the existence of this state may rotation, forces, without of the occurrence consideration of centrifugal in Newtonian idea This of motion surroundings. physical The is thus not of a relative one. of the nature mechanics has (with the above reference mentioned statement second there made) to the physical conception limitation of motion.
166
GENERAL
THEORY
OF
RELATIVITY
and is therefore not selfevident, but a statement of a physical fact. From half of this book the discussion it of the second be recognised that the tendency the theory of will of relativity is to weld the phoronomic and the physical conceptions of motion into one, in such a manner that we can only speak of a physical it also carries out a phoronomical state of motion of a body when motion
(i.". relative
to other
bodies).
NOTE
TO
PAGE
IIQ
We take into account, as that must previously emphasised, the results of the special theory of relativity can be applied only if very small world to a case of nonuniform motion, elements being We infer, therefore, are that a conconsidered. may traction individual but takes it is measuringrods of place ; the basis of the special theory not possible, on of relativity, the circumference to draw the disc as a conclusions about of On the other hand, on the any whole. circle drawn concentric top disc will always a coincide with concentric corresponding disc ; the circumference of the top disc, for circle on the lower
instance, will permanently lower disc, i.e. one wiU
along the circumference of the the other But cover it is completely. figures which to regard two cover reasonable obviously each being equal. the transference as (Cf. other of measuringrods, to the direction VII, p. Chap. of motion. situated normally 53.) it is thus quite reasonable In this sense to say that each single but that the circumference contracts, measuringrod of the disc does not contract. a whole as
run
NOTE
TO
PAGE
147
fulfilling the requirements in Chapter of Mach mentioned XV, Einstein's a it as meory represents reconciliation, Ptolemaic Copemican the were, and worldsystems (the of by Mach, be the most latter, as pointed out will always useful We for all practical can one only say that the earth By
purposes).
a out of fixed stars carry rotatory motion and the firmament is no in maintaining there point relatively to each other, and " in truth that of the two is in motion, only one and the other at rest. from believed derived It was that one the special argument had been found, indicated that only theory which of relativity be correct, viz. " the can one assertions of the abovementioned the firmament As and rotates of fixed stars is at rest." earth XI, it necessarily follows from in Chapter the special explained body can move a theory material of relativity that no with
"
SUPPLEMENTARY
NOTES
167
The opponents greater velocity than that of light. of the general follows : If the earth theory thus as were argue of relativity at rest, and the firmament of fixed stars were revolving round it, even fixed stars would the nearest attain a velocity greater distant be revolving than that ones must very of light, and c, in order that with velocities many million times greater than in a the earth their enormous cover they may orbits round To this we single day. reply that the theory of relativity deals to the special theory According only with relative motions. of for instance, that at it is quite out of the question, relativity body the solar system could traverse any time an outside cosmic that of light. On the other hand, than a greater with velocity from imagining us a system of reference nothing prevents to north moving, say, in the direction of the earth's axis from km. The a 400,000 per second. south earth of with velocity have bodies then velocities, relative to would and all heavenly
be than the which would greater of reference, the abovementioned light, without thereby violating velocity of be no law of the special theory of relativity ^for there would These c. than only with appear relative velocities greater fictitious coordinate axes to the our of peculiarly reference The case of a coordinate system system of reference. chosen the firmament the earth of fixed stars rest in it and at with Here, too, we have rotating relatively to it is quite analogous. distances of the fixed The c. no relative velocities greater than from the centre as from as of the earth other, well each stars the globe, do not or alter with velocities other point of any higher than c ; these only relatively velocities appear greater is fixed to the to the axes system which of that coordinate that is no more i.e. purely and structures conceptual earth the example a of relativity than violation of the special theory this system
" " "
aforementioned.
NOTE TO
PAGE
152
have been raised by astronomers doubts concerning recently has been discovered error a numerical this result, because Newcomb astronomer in the fundamental of the American work Inner Planets, whence the Four this the orbital Elements of on It is possible, therefore, that the taken. been had statement between the value coincidence of the striking wonderfully by Einstein Mercury the out and worked of motion perihelion by in the an error was caused accidentally value observed Be it may, by Newcomb. that as of observations reduction to agree seem Einstein's of the orbit of Mercury calculations than those performed better with current according observations A full explanation law. of this question wiU to the Newtonian on the Elements be possible, after Newcomb's great work only Inner Planets has been revised. of the Four
Quite
S
o
1
s
.s "S."'
5"
(""0
"Sjt^
I
'5 S
a
Bi P"
"6i"l
"5S o
s
"
'S
"""1
"
5.1
I Si
H
"l
/
PKINTED
BV
MOKKISON
AND
GIBB
LTD.,
EDINBURGH
SELECTION
FROM
Messrs.
Methuen's
PUBLICATIONS
This Catalogue contains only a selection ol the more important books A complete catalogue of their published by Messrs. Methuen. publications
may be obtained
on
application.
OF ART ClnttonBroek (AJ, WHAT (W. W.). THE IS THE KINCCRICKET. Cr. Sto. 6s. net. DOM OF HEAVEN? Fitth Edition. Bain W.)" Fcap. 8110, ss. net. A Digit of thb Hoom : A Hindoo Love ESSAYS ON ART. Second Edition. Fcab. Story, Thb Dsscbht Sum : A 8"o, is. na. of thb Cycle of Birth. A Hbifer of thb Daww. ESSAYS ONBOOKS, Third Edition. In the Great God's Hair. A Deaughx Fcap. Sun, 6s, net. Bi,ue. OF the Ah Essence Dusk. MORE of the ESSAYS ON BOOKS, Fcap. Soo. Ah Incarnation Show. A Mihb of the 6s. net. OP Faults. The Ashes SHAKESPEARE'S Gon. of a HAMLET, Fcap. Svo. Bubbles of the Foam, A Stbup op the SS. net. Bees, The Liveky Eve, stance Conrad The Subof MIRROR OF (Jaseph).THE of a Drbau, All Fcap. 8vo. 55, THE SEA ; Memories and Impresaioiis, Echo Spheres. of the Wide Fourth Edition. Fcap. Svo. 6s. net. net. Ah Demy, izs. 6d. net. Drever (James). THE PSYCHOLOGY OF Baker IUusEVERYDAY LIFE. Cr. Svo. (C. H. CoUIbs). CROME, 6s, net. THE PSYCHOLOGY trated. Quarto. "5 5J, net. OF INDUSTRY, BaUonr fSir Grsbsm). THE LIFE OF Cr. 8ra, ss. net. ROBERT TieenLOUIS STEVENSON, Einstein THE (A.). RELATIVITY: In one Cr, "00. Volume, Uetk Edition. SPECIAL AND THE GENERAL Buchram^ ys. 6d. net. THEORY, Translated by Robert W, OF DRAWINGS. Bateraan (H. H.). A BOOK Lawsoh, Seventh Edition. Cr.Svo. ss.net. Royal ON RELATIVITY, Fifth Edition. Two ^to. SIDELIGHTS 6d. net. Lectures by Albbkx 10s. Eihstbih, Cr. Svo, SUBURBIA. Demar \to. 6s. net. 3$. 6d. net. HISTORY Other Boohs on the Elnflteln heory, BeU (Mary \. M.). A SHORT T PAPACY. Demy 8110. 21J. m*, SPACE" OF THE TIME" MATTER, By Herh ahh Weyl, Demy 800, sis, net. BeUoe (H.)" EINSTEIN the Sea, 6s. Pakis, 8s, 6i. net. Hills ahd : His Work TflE SEARCHER Kindred jects, SubOn Noisihg and BXPLAINBD IH DIALOGUES ElNSTEIN. WITH net. 6s. net. Oh Everythihg, 6s. net. By Alexander Moszkowski, Demy 6s. net. First and Last, 6s. Svo, 12s, 6d. net. Oh Somethihg, Other, 6s. AN INTRODUCTION This and That and the TO THE THEORY na. Antoinette, OF RELATIVITY. By Lyndon Boltoh. net. Maris iSs. net. TENNIS Cr. Sffo. 5s. net. Blaekmore (8. FoweU), LAWN Demy 8"o. RELATIVITY GRAVITATION, lUustrated, UPTODATE, AND By Various Writers. Edited by J, Malcolm I2S. 6(2.net. Bird, MATION, TRANSFORCr. Sua. ys. 6d. net. Carpenter (Q. H.). INSECT RELATIVITY Demy Svo. lis. 6d. net. AND THE UNIVERSE. By Dr. Harry Schmidt, D Second Edition. Chandler (Arthur),.D., late Lord Bishop of Cr, Bvo. ss. net. Bloemfontein THE OF EINSTEIN'S IDEAS THEORY, Ara C("LI : An Essay inMyatical Theology, Experience, Cr. Svo. ss. net. By J, H. Thirrihg, ss. net. 5S. net. Faith and By Herbert Passing Moment, 6s. RELATIVITY FOR ALL, The Cult of the union, ReDingle, English Church and Fcap. Svo. as. net. net. The Scala Mundi, Evans (Joan). ENGLISH JEWELLERY. 45. 6d. net. 5s. net. Royal ito. "z zas. 6d. net. Chesterton (G,K.)" All White Horse, Fylemao AND NEYS. CHIMof the The Ballad (Rose). FAIRIES Trenendous Ttaellth Edition. Cohsidered, Things Fcap. Svo. A Discursions, Alarms ahd Trifles, 35. 6d. net. THE Uses of GREEN. Sixth Men, The FAIRY Edition. of Miscellany All Fcap. Svo. 3s. 6d. net. Diversity. Ecap. "vo. 6s. net. FAIRY FLUTE. Second Edition. Soho. Water, ahd WiHB, Fcap. Bvo. THE Fcap.Svo. 3s. 6d. net. IS. 6d. net.
Armstrong
JF.
"
Messrs.
Methuen's
Publications
SEVEN
Svo.
GIbUna IN (H. "e B.). INDUSTRY ENGLAND : HISTORICAL OUTLINES. With Maps Tfnth Edition. and Plans. Demy Svo. 125. 6d. net. THE HISTORY OF INDUSTRIAL With 5 Maps ENGLAND. and a Plan. Cr. Svo. 55. Twentyseventh Edition. Gibbon (Edward). THE DECLINE AND FALL OF THE ROMAN EMPIRE. Edited, with Notes, Appendices, and Maps, by J. B. Bury. Seven Volumes. Demy 8p0. Illustrated. Eacit 6d, 12s. net. Also in Seven Volumes. UniUustrated. Cr. Svo. Each Js. 6d. net. Glover (T. K.)" The Cohflict of Religiohs in the Early Rohan Eufirb, zo$. 6d. net. Poets and Puritans, ios. 6d. net. Froh Pericles Philip, ids. 6d. net. Virgil, 105. 6d. TO Christian Tradition and its net. The Verification (The Angus Lecture for 65. net. igi2}, Orahame WIND IN (Kenneth). THE THE WILLOWS. E Tuiellthdition. Cr. Svo. 7s. 6d. net. ANCIENT Hall (H. R.). THE HISTORY OF THE NEAR EAST FROM THE EARLIEST TO THE BATTLE TIMES OF SALAMIS. Illustrated. Fifth Edition. Demy Svo. 21s. net. SCARLET Hawthorne THE (Nathaniel). LETTER. With Illustrations in 31 Colour, by Hugh Thouson. Wide Royal Svo. 315. 6d. net. OF Holdsworth (W. S.). A HISTORY Seven Volumes. Demy ENGLISH LAW. Soo. Each 25s. net. MYSTICISM. Inge (W. H.). CHRISTIAN Lectures of 1899.) Fi/th (The Hampton Cr. Soo. 7s. 6d. net. Edition. OP ENGLISH Jenks (B.). AN OUTLINE GOVERNMENT. LOCAL FifthEdition, revised. Cr. Svo. 5s. net. OF ENGLISH LAW A SHORT HISTORY : From Earliest Tiues to the End the Edition, Second Year of the igii. Svo. 125. 6d. net. revised. Demy TIONS JnUan (Lady) ol Horwleh. REVELALOVE. Edited by OF DIVINE Grace Warrack. Seventh Edition. Cr.
THE Cr.
net.
Fcap.
Also a Square
SEAS. Thousand. iJTJ* Buckram, 7s. 6d. net. Also Svo. Cloth, 6s. net ; leather, is. 6d.
Service Edition. Tao Volumes. Each 35. net. THE FIVE NATIONS. 126th Thousand. Cr. Svo. Buchram, 7s. 6d. net. Also Fcap. Svo. Cloth, 6s. tut 1 leather,71. 6d.
Fcap. Svo.
net.
Service Edition. Two Volumes. Each 3s, net. DEPARTMENTAL DITTIES. I02ni Thousand. Cr. Svo. Buckram, 7s. 6d. Also Fcap. Svo. Cloth, 6s. net;' net. leather, 7s. 6d. net. Also a Service Edition. Volumes. Two Square Fcap. Svo. Each 31. net. THE YEARS BETWEEN, gsth Thousand. Cr. Svo. Buckram, 7s. 6d. net. Fcap. Svo. Cloth, 6s, net ; leather,7s. 6d. net. Also a Service Edition. Two Volumes. Square Fcap. Svo. Each 3s. net. HYMN BEFORE ACTION. lUuminated. Fcap. 4to. IS. 6d. net.
Also a Square
Fcap, Svo,
RECESSIONAL.
is.
Illuminated. Fcap. 4I0. 6d. net. TWENTY POEMS FROM RUDYARD KIFLINS. Fcap. Svo. 313M Thousand.
is.
net.
G.). ('Evoe* of
REGAINED. Morrow.
Punch.)
PLETE COMMary). THE WORKS. Edited by E. V. Lucas. A New and Revised Edition in Six Volumes, With Frontispieces. Fcap. Svo. Each 6s. net. The volumes are : I. Miscellaneous Prosb. 11. Elia and Last Essay the Elxa. of hi. Books Children, for iv. Plays PoBva. and V. and VI. Letters. ESSAYS THE OF ELIA. With an Introduction by E. V. Lucas, and 28 Illustrations by A. Garth Jones. Fcap. Svo. Ss. net. Lankester (Sir FROM AN Ray). SCIENCE Svo. 55. net, CHAIR. EASY Illustrated. Thirteenth Edited, with Introduction Keats (John). POEMS. Edition. Cr. Svo. 7s. 6d. net. and Notes, by E. ds SelinFROM AN EASY CHAIR. gravure. SCIENCE With Frontispiece in Photocourt. a Second Series. Illustrated. Third Edition. Fourth Edition. Demy Svo. Cr. Svo. ys. 6d. net. I2S. 6d. net. DIVERSIONS OF A NATURALIST. SCIENCE OF Kldd (Benjamin). THE Illustrated. Third Edition. Cr. Svo. Edition. Cr.Svo. ys.6d.net. POWER. Ninth 7s. 6d. net. DemySvo. 8s. 6d.net. SOCIAL EVOLUTION. SECRETS OF EARTH AND SEA. Cr. NATURE. WITH A PHILOSOPHER Bvo. Ss. 6d. net. Cr. Svo. 7s. 6d. net. Secona Edition. Lodge Kipling THE (Sir OUver). MAN AN" (Rudyard). BARRACKROOM Cr. Svo UNIVERSE : A Study of the Influence BALLADS. 215th Thousand. Advance OF the Scientific in Buckram, ledge Know7s. 6d, net. Also Fcap. Svo. UPON Understanding our Cloth, 6s. net ; leather, 7s. 6d. net. of 'J'co Volumes, Christianity. Ninth Edition. Cr. Svo. Also a Service Edition. Square Fcap. Svo. Each 3s, net. 7s. 6d. net.
"
Lamb
(OharlM and
"
Messrs.
THE SURVIVAL Unekcochizsd
BIethuen's
Pubucations
Guest, Wrack Miracle
Act,
OF MAN : A Stodt m Huhah Faculty. Settnth Edition. Cr. tvo. fs. 6d. net. MODERN PROBLEMS. Cr. Sua. js. 6d. net. RAYMOND; Lifi oit Death. amd lUiutrated. Twelfth Eiition. Demy 6vo. los. 6d. na. hating (F. H.). ATOMIC THEORIES. Demy Sm. I2s. 6i. net. Lneu (E.V.)" The Life of Chaklbs Lahb, a vols.,sis. im Holland, ios. 6d. net. A Wahderee London, in xos. 6d. net. A Wanderer Revisited, ios. 6d. net. A net. London Wanderer in Paris, ios. 6i. net and 6:. in Florence, ios. 6d. net. A Wanderer in Venice, ios. 6d. net. net. A Wanderer The Open Road: A Little Book for Wayfaieis, 6s. 6(/. net. The Friendly Town: A Little Book for the Urbane, 65. net. Fireside Sunshine, and 6s. net. Character Cohedy, 6s. net. and The Gentlest Art : A Ciioice of Letters by Entertaining Hands, 6s. 6d. net. The Second Post, 65. net. Her Infinite Vabiety A Feminine Portrait Gallery, : 6s.net, Good Cohpan y ; A Rally of Men, 6s. net. One Day 6s. net. and .Another, Old Lamps for New, 6s. net. Loiterer's 6s. net. Cloud Harvest, Silver, 6s. and of Baghdad, and A Boswell other net. Essays, Eagle and 6s. net. *Twixt Dove, 6s. net. The Phantom Journal, 6s. EsSAYS AND DIVERSIONS, AND OTHER Specially Selected : A Choice of net. Essays, 7s. 6d. net. Urbanities. trated Illusby G. L. Stampa, 7s. 6d. net. The School An Anecdotal British : Guid" to the British Painters and Painting"=.in the National Gallery, 6s. net. Roving West : Notes Roving East and gatheied in India, Japan, and America. Austin Abbey, R.A. 55. net. Edvih Delft, of Verkeee 2 vols. "6 6s. net. 6d. net. IOS. INGS. PAINTHeIdrBm(D.S.). REMBRANDT'S Wide Royal Svo. "z 2s. net. OF ANTHOLOGY Methnen (A.). AN Introduction With VERSE. MODERN Edition. Seventh Lynd. Robert by Pcap. Svo. 6s. net. Thin paper, leather, Js. 6d. net. TION INTRODUCMcDongaU (WllUam). AN PSYCHOLOGY. SOCIAL TO Cr. Svo. 8s. 6d. net. Seventeenth Edition. AND BODY Defence Svo. Demy
of a and A History MIND: Animism. Fifth Edition. I2S. 6d. net.
6s. net.
op the
Poems, Storm,
:
Burgomaster of Stxlehonde Play in Three Acts, 5s. : A The Betrothal "". ; or. The Blue Bird Chooses, 6s. net. Mountain Paths, 6s. net. The Story of Tyliyl, ais. rut. Hllne (A. A.)" Not that it Matters. Veap. Svo. 6s. If I May. Fcap. Svo. 6s. net. net. HeTfll (Bslph). MAYFAIR AND MONTMARTRE. Demy Svo. 15s. net. AND Horwood (GUbert). EURIPIDES SHAW Essays. Cr. Svo. other : With Js. 6d. net. AND Osborn (B. B.). LITERATURE LIFE. Cr. Svo. ys. 6d. net. Oxenham (John) Bees Amber Little Book in of ; A Thoughtful Verse. Pott Svo. SmaU Well; All's Stiff Boards, zs. net. A Collection of War Poems. The King's High The Splendid. Way. Vision The Fiery Cross. High Altars : The Record of a Visit to the Battlefields of France Hearts geous. Couraand Flanders. All Clear ! AU SmaU Pott Svo. Paper, is. 3d. net ; cloth boards, 2s. Dawn. of the Gentlemen net. Winds The King, 2s. net.
" "
5s. ntt.
The
NATIONAL DECAY.
AND
6s. net.
NATIONAL
Haeterlinek (Hanriee) The Blue Bird : A Fairy Play in Six Acts Play in Magdalene : A 6s. riel. Mary
Three Acts, 5S. nrt. Death, 6(. net. The Odr Eternity,
3s. 6d. net.
A OF Petrie (W. H. Flinders). HISTORY Illustrated. Six Volumes. Cr. EGYPT. Svo. Each gs. net. XVIth Vol. 1st to the I. From the Dynasty. 6d. net.) Ninth Edition. (los. XVIIIth II. The and XVllTH %'OL. Dynasties. Sixth Edition. Dynasties. Vol. hi. XXXth XIXth to Second Edition. Ptolemaic Vol. Egypt ihe IV. undkb Dynasty. J.P.Mahaffy. Second Edition. Roman Rule. Vol. V. Egypt under Second Edition. J. G. Milne. Middle Ages. Vol. VI. Egypt in the Second Edition. Lane Poole. Stanley TELL THE FROM AND EGYPT, SYRIA Cr. Svo. EL AMARNA LETTERS. 5s. na. Translated from the TALES. EGYPTIAN to First Series, ivth Papyri. xiith Third Edition. Dynasty. Illustrated. Cr. Svo. 5s. net. Translated from the TALES. EGYPTIAN Second Series, xvnith to xixth Papyri. Edition. Illustrated. Second Dynasty. Cr. Svo. 5S. net. PolUtt (Arthur W.). THE ENJOYMENT Second Edition. Cr. Svo. MUSIC. OF 5S. net. OF HISTORY Price (L. L.). A SHORT IN ENGLAND ECONOMY POLITICAL TO ARNOLD SMITH FROM ADAM Eleventh Edition. Cr. Svo TOYNBEE. 55. net.
Unknown
Messrs. Methuen's
"
Publications
LIFE. AND Belsni (Edmniil) Toner (W. J.). MUSIC ToHHY Cr. Svo. ys. 6d. net. Tohuy Smith's Ahihau. Smith's Other Animals. Tommy Smith A Underbill (E?elyn). MYSTICISM. Smith agaih Zoo. Tommy AT at THE Study in the Nature and Development of Zoo. THE Each 2S. gd. Jack's Iksects, Ninth Spiritual Consciousness. Man's Iksbcts, $$, 6d. 3S. 6d. Jack's Other Svo. is". net. Demy Edition. With Shelley (Peroy ByHhs). POEMS. GOLF. TO PLAY Tardon (Harry). HOW an Introduction by A. CluttokBrock Edition. Cr. 8". Illustrated. Fifieenth Notes by C. D. Lococe. Two and Ss. 6d. net. Volumes. Demy 8vo. "i is. net. TESTAMENT Wade Smith THE OF WEALTH (O. W.). NEW (Adam). Demy Svo. i8". net. HISTORY. Edited by Edwih Cahnah. NATIONS. Two Volumes. Third Edition. Demy A BOOK Waterhense (Elliabeth). LITTLE Bvo. "i zos. net. DEATH. OF LIFE AND Twentyfirst AT Smith (S. 0. Kalnei). LOOKING Edition. Small Pott Svo. it. 6d. net. Uustrated. PICTORES. Second tion. Edi8vo. 6s. net. Fc"p. OF SHORT HISTORY Weill ".). A DRAMA. ELIZABETHAN SpSM (Janet). ROME. Eighteenth Edition. With 3 Cr. Zvo. 65. net. Maps. Cr. Svo. 5". BtevenBon JR. L.J. THE LETTERS OF OF OSCAR THE WORKS WUde (Oseat). LOUIS STEVENSON. ROBERT Edited Fcap. Svo. Each 6s. 6d. net. by Six SiDHEY COLVIN. A New RearrangedWILDE. Savile's Crime and Arthur I. Lord Edition in four volumes. Fourth Mr. W. H. 11. The Portrait or thb Edition. Svo. Each 6s. net. Fcap. Padua, hi. Poems, iv. Duchess of Snrtees (R. 8.) Fan. Lady Windermere's v. A Woman Crom, Handley 7s. 6d. net. Mr. band, No Importance, vi. An Ideal HusOF Sporting Sponge's Tour, 71. 6d, net. Importance Being op vii. The Ask Mamma ; or. The Richest Commoner of granates. PomeEarnest, A House viii. in England, ?s. 6d. net. Jorrocks's Prox. De IX. Intentions, Jadmts and Jollities, 6s. net. Mr. says. EsPrison Letters, xi. fundis and Hounds, Facey Romford's 7s. 6d. net. Florentine Salohb, a XII. Hawbuck Grange The Sporting ; or. Courtisahb. Tragedy, and La Saints Scott, Esq., 6s. Adventures of Thomas Pall Mall. A Critic in xiv. XIII. Ringlets ? 7". 6d. net. net. Plain or op Prose Oscar Wildb. Selected Hall, 75. 6d. Het. Hillingdon Decoration. Art and XV. OF LAWN ART Tllden (W. T.). THE OF POMEGRANATES. lUnsA HOUSE Illustrated. Fourth Edition. TENNIS. trated. Cr. ^to. 315. net. Cr. Bvo. 6s. net. STRENGTH OF BOOK IRISH Teats (W. B.). A Tlleston (Mary W.). DAILY NEEDS. DAILY Twentyseventh Fourth Edition. Cr. Svo. js. FOR VERSE. Edition. Medium i6mo. 3s. 6d. net. net.
"
Part
II.
"
^A
Selection
of
Series
With
Numerous
Paihtrd England. Glass in Amciekt False Antiquities. and Archaologv Thb Brasses The Bells op England. Walled Thb Castles and OF England. England. Celtic Art in of Towns Christian Times. wardens' ChurchPagan and Domesday Accounts. The Furniture. English Church Inquest. Monastic Costume. English English FolkLore Seals. English Life. as Science. The Gilds and an Historical The London. Hermits of Companies ENGLAND. THB AMCBOBITES OF AHD
Messrs.
The
Methuen's Arden
Editor. Svo.
Publications
Shakespeare
R.
General
Demy An
H.
CASE
in Single Plays ; each edited with a full Introduction, edition of Shakespeare Textual Notes, and a Commentary the foot of the page. at
Classics ol Art
Edited by Dr.
With
The Art
of
J. H.
W.
LAING
numerous
Greeks, 21s. net. The Romans, i6j. net. Chardih. 15s. net. Donatsllo, i6s. net. Gbqrgb RoHNEY.xss. Ghiklahdaio, 15s.net. net, Lawsbmcb, 25*. net. Hichelangelo. 151.
of
the
Art
the
The
Thb
Complete
Series
Svo
The
Complbtb Lawn Tennis 55. net. Player, Complete 6d. net. The 12s. Motorist, Complete ios. 6d. net. The Mountaineer, 16s. net, Thb Complete The Oarsman, Cpmplete 155, net. Photographer,i2s. 6d. net. The Complete Footballer, New land Zeaoh Rugby the System, 6d. net. The plete Com12*. Shot, Complete 165. net. The Swimmer, 6d. net. Thb Complete ios. Yachtsman, i8s. net.
The
With
numerous
Connoisseur's Library
"1
lis.
Ivories.
Miniatures.
Jewellery.
Sculpture.
Mezzotints. Sbals.
Wood
Handbooks
Demy
Incarnation, of the 15s. The Doctrine Early Christian of History net. A to Doctrine, i6i. net. Ihtroductioh 6d. net. 125. Religion, of History the History of thb to Introduction Am
of Theology
Svo
THB
of 12s.
the
125.
in
6A. net. The Philosophy England America^ and Thb XXXIX Articles of Ehclahd, of 155. net.
The
The pREVBif tion of thb Cohhom Throat Staying Plague. the Tuberculosis. Ear Troubles. Health Child, a". net. of vhe
Messrs.
The
Handy
Methuen's
Publications
Library of Devotion
Books, well edited Notes (where necessary)
With Small
Little Boolis
With many Each
on i6mo.
Art
5s. net each volume 30 to 40
Illustrations. Demy
zoo
a
Albrecht DOrer. The Arts of Japan. Bookplates. BurhbBotticelli. JoHES. Cellini. Christian Symbolism. Art. Christ Claude. Constable. in Corot. Early English WaterColour. George Enamels. Frederic Leigbton. and RoMHEV. Greek Art. Grbuze
Holbein. Boucher. Illuminated Manuscripts. Jewellery. John HofpMillet. Sir Joshua Reynolds. NER. Art. in Miniatures. Our Lady Turner. Vahdycx. Raphael. Rodih. Watts.
many
Illustrations by E. H. New
photographs
Small Pott 8wo. Guides to the English and Welsh
^s. net to js. 6d. net.
Counties, and
some
wellknown
districts
form; a The main features of these Guides are (i) handy and charming by wellknown illustrations from photographs ; (3)good artists and (2) an plans and maps ; (4) adequate but compact presentation of everything that is interesting in the natural features, history, archaeology, and architecture of the town or districttreated.
The
Edited by W. Pott i6mo.
CRAIG.
Leather, priceis. gd. net each volume 40 Volumes. Cloth, IS. 6d. net.
Arnold Bennett.
Edition. Fijth
Edition.
Messrs.
Methuen's
Publications
The
Westminster
Demy
Commentaries
LOCK
8vo
x6s.ntt. Job, 8". 6d. net. Tsa Pastoral Epistles, 85. 6d. net. The Philippiahs, 85. 6d. net. St. Jahes, 8s. 6d. net. Si, Matthews, 155. net. St. Luke, 135. net.
General
Taa
Acts
Editor, WALTER
Apostles, 121. 6i. net. of *bx 85, 85. td. net. I Corinthians, 6d. net. Exodus, 15s. net. Ezbkixl, 225. 6d. net. Genesis, 165. net. Hebrews, 81. 6d. net. Isaiah, i6t. net. Jereuiah,
Amos,
Part
"
III.
"
^A
Selection
of
Works
of
Fiction
BniuU (Arnold) Lessways, 8". net. Hilda Clayhanger, The Card. Twain. 8f. 6d. net. These Story of A Five Towns : Regent The Price of The in London. Adventure A Man Alive. proh Buried Love. Five of the The Matador North. the God hath Wkoh Joined. A Towns. He. Prohack. Great Man : A Frolic. All 75. 6d. net. (GeorgeA.)" Birmingham Party. Search The Gold. Spanish Up, Lalagk's Lovers. *The Bad Times. Lawyer. All Lost The Rebels. THE 75. 6d. net. Inisheenv, 8s. 6d. net.
Bnrrouglis (EdgarHice)" The 6s. net. Apes, the of Tarzah 65. net. The Beasts of Taezan, Return 6s. net. The Son of Tarzan, of Tarzan, 65. net. Jungle Tales of Tarzan, 65.
^ ^
"
Opar, of 7s. 6d. net. 65. net. The Gods of Mars, A Princess of Warlord The 65. net. Mars, of Mars, 6s. net. Thuvia, Maid of Mars, net. 65 net. Tarzan the Terrible, 25. 6ii. out withMan 65. net. The Mucker, The 6s. net. a Soul,
and net. Tarzan the 6s. net. Tarzan the
Jewels
Corelll(Marie)" A Romance op Two Worlds, 75. 6d. net. gotten, Vendetta: Story of One Foror, The A Norwegian 85. net. Thelha ; Princess, 85. 6(2.net. Ardath : The Story of a Dead Self, 7s. 6d. net. The Soul of LiLiTH, 75. 6d. net. Wormwood : A Drama : A Dream of of Paris, 85. net. Barabbas the World's Tragedy, 8s. net. The Sorrows MasterSatan, of 7s. 6d. net. The Power: Christian, 8s. 6d. net. Temporal A Study in Supremacy, 6s. net. God's Good Man : A Simple Love Story, 85. 6ii. The Tragedy Holy Orders : net. of a Quiet Life, 8s. 6d. net. The Mighty Atom, 7s. 6d. net. Boy : A Sketch, 75. 6d. net. Cameos, 6s. net. The Life Everlasting, 85. 6d. net. The Love of Long Ago, and Other Stories,' 8s. 6d. net. Innocent, Power: Secret A The 75. 6d. net. Romance of the Time, 7s. 6d. net.
Untamed,
HIchens (Rohert)
"
Conrad (Joseph)
"
ho.
05
Six, 7s. 6d. net. Victory .An The Cr. ivo. 95. net. Tale. A Simple Tale. Cr. Svo. : Agent Cr. Eyes. Western net. Under Cr. ivo. gt. net. Chance. gs. net
of
Tongues Conscience, of 7s. 6A. net, Felix : Three Years in a Life, 7s. 6d. net. Fan, 7s. 6rf. net. The Woman with the 85. 6d. net. The Allah, Garden of 85. 6d. net. The Call of Blood, the Threshold, The Dweller on 7s. 6d, the of Ambition, 7s. 6d. net. net. The Way In ihe Wilderness, 7s. 6d. net.
Messrs.
Hope
A
Methuen's
Publications
(Jobn)
"
(Antbonr)
"
Air. A Man Mark. ov Simon Dale. King's Thb Mirror. The Dolly Dialogues. Mrs. Maxon Protests. A Young Man's Year. Beaukaroy Hohb Wars. from vbb All 7s. 6d. net.
oy
Change
Oxenbam Profit
Thb Song of Hyacinth, Loss. and Other Stories. The Coil of Carhb. The Quest of the Golden Rose. Mary AllAlohb. All 7s. 6d. net,
and
Farkei Mrs.
of
(OUbert)"
'
St. net. Sea Urchins, A Master or 5s. net and 3$. 6d. net. Craft, 6s. net. Light Freights, 5s. net. The Skipper's Wooing, 5s. net. At SunwicH Port, 5s. net. Dialstonb Lane, 5s. net. Odd Craft, 5s. net. The Lady OF THE Barge, 5s. 5s. net. Saltbaveh, Knots, 5s. net. Short net. Sailors' Cruises, 6s. net. Ntnth
Translation Falchion. The Savage. When Valmond came PoNTiAC : The Story, ^f a Lost TO Napoleon. Am of ADVBNTtjRE the " North : The Last Adventures of Pretty Pierre.' The Seats of the Mighty, Thb Battle Strong ; op the A Romance Kingdoms. Thb Trail of the oi Two Sword, Northern Lights. All 71. 6i. net.
a
London (Jaok"" WHITE FANG. Edition. Cr. 8vo. ys. 6d. net.
"
Lucas (E. V.) Listener's Lure : An Oblique Narration, 6s. net. Over Bemerton's An Basy: going Chronicle, 6s. net. Mr. Inglesidb, 6s. net. London Lavender, 65. net. The Vbrhilioh Landmarks, 65. net. Box, 6s. net. Vbrsha Midst, ih the Rose, 6s. net. 85. 6d. net. Rose and
Pblllpotti(lien)" Mist. Children Thb River. of the Dehbtbr's Daughter. The Huuan Boy and the Wab. All 7s. 6d. net,
Robmer (Sax)" The Orchard Tales of Secret Egypt. Scorpion. Thb Golden OF Tears. AU Devil The Doctor, 7s, 6d, net. Dr, Ft^;MANCHU. The Mystery of The Yellow Claw. All 3s. 6d, net,
and Houses. Swlnnerton (F.) Shops The Happy Family. Septehber. On Coquette. The Chaste Thb Staircase. The Merry Wife. All 7s. 6d. net. Thb Casehbht, The Young Heart, AU 6s. net. Idea.
HoKenna (Stepben) Between SoNiA : Two Worlds, 8s. net. NiNETY^Six Hours' Leave, 7s. 6d. net. The Sixth Sense, 6s. net, Midas " Son,
"
8s. net.
Halet (Lucas) The History Calmadv of Sir Richard : A Romance. los. Carissiha. net. The The Dbadhau Gateless Barrier. Hard. All 7s. 6d. net. The Wages of Sin. 8s. net. Colonel Enderby's Wife, 7s. 6d. net.
"
Fourth Edttton,
Mason (A. E. W.). CLEMENTINA. BdUion, Cr, Svo. Illustrated. Ninth 7s. 6d. net,
Hllne (A. A.)" Tub Holiday Round. The Day's Play. All Cr. Bvo. 7s. 6d. net. Once a Week. Cr. Sao, Side. The Sunny 6s. net. Hystert. The Red House Cr. 8"a.
6s. net.
WUllamson (0.n. and A. U.)" Conductor The Lightning : The Strange Adventures of a Motor Car. Lady Betty Water. It Happened the across in Egypt. The Shop Girl. The Lightning My Friend Conductress. the Set in Silver. The Chauffeur. Sbcret. The Pearl Great Love AU Cbucifiz Pirate. 7s, 6d, net, CoRNEB. 6s, net.
for CompleteList
Fcap. Svo
J22