"HE UNIVERSE _

rom Flat Earth To Quasar ‘*

Were Asimov Has made the stupendous mysteries of space and time and
hu incredible dimensions of astronomy thrilling and comprehensible to *
;ae ae and interested reader’ JAMES S. care ae PEN Cie eue ein)
i :
ae Me
iN ;' :
Fy
i vt
¥Sae
' 7
a
he we by ai
rat
eee 3 3
* . O es
{ .
let i i o

ie” :
t: a3 is
\ Py.
rh U _* i “a af ry
\ - y ry
i f ° Seiad Pe
i = iJ a
} Oo
A
.J °
A

‘ Ps
\ i i ©
Hi Al ‘

f beraeas °:
a Va
Panne
; } 4
Vie
ear
esi:
i eee:
a iia
Ii ag e bd
\ Ni i . e

4! |
jie

{ {

K Le Ae
Ft biel
f tlk
J =e
\ ig 1

‘
7
e A
Meee nce feo
oe 2 E
cr A
ap
eT ee
_

ISAAC ASIMOV is undoubtedly America's

foremost writer on science for the layman.
An Associate Professor of Biochemistry at
the Boston University School of Medicine,
he is the author of such standards of science
reportage as THE CLOCK WE LIVE ON,
THE GENETIC CODE, and the three-volume
UNDERSTANDING PHYSICS, in addition to
many popular works of science fiction. Born
in Russia, Isaac Asimov came to this country
with his parents at the age of three, and grew
up in Brooklyn. In 1948 he received his Ph.D.
in Chemistry at Columbia and joined the fac-
ulty at Boston University. The astronomer
James S. Pickering has said of him, and of
this book, “Isaac Asimov is a skilled writer
of great experience, and he has made the stu-
pendous mysteries of space and time and the
incredible dimensions of astronomy thrilling
and comprehensible to any intelligent and
interested reader. He has covered all the es-
: eentials of astronomy in THE UNIVERSE.”
 Other Avon Books by
Isaac Asimov

THE FOUNDATION TRILOGY 26930 $4.95

ASIMOV’s GUIDE TO THE BIBLE ;
THE NEw TESTAMENT 24786 4.95
ASIMOV’sS GUIDE TO THE BIBLE
THE OLD TESTAMENT 24794 4,95
ADDING A DIMENSION 22673. 4b
FROM EARTH TO HEAVEN 292314 Sie
LIFE AND ENERGY ; 31666 1.95
THE NEUTRINO 25544 1.50
SOLAR SYSTEM AND BACK 10157.. Tae
VIEW FROM A HEIGHT 24547 1.25
FOUNDATION 29579 - tae
FOUNDATION AND EMPIRE 30627 = 1.50
SECOND FOUNDATION 29280 1.50
 From Flat Earth to Onueee

Isaac Asimov

= © A DISCUS BOOK/PUBLISHED BY AVON BOOKS

 AVON BOOKS
A division of
The Hearst Corporation
959 Eighth Avenue
_ New York, New York 10019

Copyright © 1966 by Isaac Asimov

_ Published by arrangement with Walker and Company.
_ Library of Congress Catalog Card Number: 66-22515.
ISBN: 0-380-01596-X
_ Allrights reserved, which includes the right
_ to reproduce this book or portions thereof in
_ any form whatsoever. For information address
_ Walker and Company, 720 Fifth Avenue,
~ New York, New York 10019.

First Discus Printing, February, 1968

Eleventh Printing

Cover photograph courtesy of Mount Wilson and Palomar

‘Observatories. Copyright by California Institute of Technology
and Carnegie Institute of Washington.

DISCUS BOOKS TRADEMARK REG. U.S. PAT. OFF. AND

FOREIGN COUNTRIES, REGISTERED TRADEMARK—
MARCA REGISTRADA, HECHO EN CHICAGO, U.S.A.

Printed in the U.S.A.

 Table of Contents:
'
>>

CHAPTER 1— The Earth

Introduction
The Flat Earth
The Spherical Earth
The Size of the Earth
}~
be

CHAPTER 2 — The Solar System

: The Moon
The Sun
Parallax
The Size of the Solar System

he CHAPTER 3 — The Stars

The Vault of the Sky
A Multiplicity of Suns
, The Search for Stellar Parallax
The Distance of the Nearer Stars

CHAPTER 4— The Galaxy

: : Olbers’ Paradox
: Herschel’s Lens
- The Moving Sun
; Star Clusters
; Variable Stars
 : CHAPTER 5 — The Size pywhe Galaxy
- The Doppler Effect
- The Spectrum
Spectral Lines
The Galactic Center
Galactic Dimensions

CHAPTER 6 — Other Galaxies

The Andromeda Nebula 94
Novae
The Andromeda Galaxy 102
The Spiral Galaxies 106
CHAPTER 7 — The Age of the Earth
Angular Momentum 110 ©
The Conservation of Energy itsins
Nuclear Energy 119
_ CHAPTER 8 — The Energy of the Sun
The Planetesimal Hypothesis 124
Constitution of the Sun 12%
Surface Temperature of the Sun 131.
Internal Temperature of the Sun tel 134

_ CHAPTER 9 — Types of Stars “@F"

Constitution of the Solar System 138
Spectral Classes 141
Giant Stars and Dwarf Stars 144 —
The H-R Diagram 148 |

CHAPTER 10 — Stellar Evolution

The Mass-Luminosity Relation 151
Interstellar Gas 156
Beyond the Main Sequence 161
 ~ White Dwarfs
_ Supernovae 172
eee Dying Stars 177
Second-Generation Stars aR 181

_ CHAPTER 12 — Galactic Evolution

Bs The Question of Eternity
— 184
Classes of Galaxies 186
- Stellar Populations 189

_ CHAPTER 13 — The Receding Galaxies

- The Galactic Red Shift 1937
Relativity 198
Clusters of Galaxies 203

HAPTER 14— The Observable Universe

Olbers’ Paradox Again 207 =
= Hubble’s Constant 210
_ The Cepheid Yardstick Revised eee

PTER 15— The Beginning of the Universe

The Big Bang 219
The Formation of the Elements 224
Before the Big Bang 227
Continuous Creation 231

‘CHA TER 16 — Particle Bombardment

- Massless Particles 240
Cosmic Rays | 244
Cosmic Ray Sources 248
: Le P aes
an ae ‘ Soap tees (het
ean -s:, ape ar A
CHAPTER 17 — Energetic Photons
The Electromagnetic Spectrum 253
Neutron Stars 256
Antimatter 261

CHAPTER 18 — Radio Astronomy

The Sun 268
The Planets 270
The Stars 275
‘The Galaxy 278

CHAPTER 19 — The Edge of the Universe

Colliding Galaxies 283.
Exploding Galaxies 286
The Distant Radio Sources 2a
Quasars é 295
 fem
Palas
E a
3
8 2
ie
2 >
fo
2 z
ee es
 CHAPTER 1

a The Earth

; _ Introduction

S. In the last few years, astronomers have excited them-

_ selves and the public enormously by the discoveries they
_have been making in unimaginably distant outer space.
fe Phrases like “quasi-stellar objects” (abbreviated as —
_ “quasars”, and “blue stellar objects” (BSO’s) are making
headlines. Points of light billions of billions of miles away
set scientists to wondering about the far past and far
future of the Universe.
_ Does the Universe extend forever or is there an end
somewhere? Does it expand and contract like an ac-~
cordion, with each in and out motion lasting billions of
_ years? Was there a time when it exploded once and for all
_ and will the flying fragments separate until our own frag-
‘ment isvirtually alone in the universe? Does the Universe
renew itself and is it eternal, unborn, and undying?.
_ We are a fortunate generation, for we are watching a
_ period of astronomy in which the answers to such questions
_and to many others equally intriguing may actually be at
hand.
The situation is an unexpected one, too. The celestial
Eebjects that are opening new vistas for astronomers were
not known before the 1960’s. The rockets and satellites :
_ that are now feeding them so much data were not blasting
off before the 1950’s. The radio telescopes that disclosed
‘unexpected wonders of the Universe were not in existence
_ before the 1940’s.
In fact, if we go back 2500 years to, say, 600 B.c., we
_ will find that the entire Universe known to man was but a
patch of flat ground, and not a very large patch either.
That is still all that man is directly aware of today; just
13
 14 The Universe
a patch of flat ground and, of course, the sky overhead,
with small luminous objects shining in it. Nor does the sky
seem to be very far above our heads.
By what process of reasoning then, did the narrow
7 surroundings visible to ourselves fade outward and out-
_ ward and outward until no man’s mind can possibly grasp
the size of the Universe we now speak of, or imagine the
tiny insignificance of our physical surroundings in compar-
ison with it?
In this book I want to trace the steps by which man’s
_ grasp of the Universe as a whole (“cosmology”) and of its
origin and development (“cosmogony”) widened and
deepened.

The Flat Earth

In 600 B.c., the Assyrian Empire had just fallen. At its

height, it had extended from Egypt to Babylonia, for an
' extreme length of 1400 miles. It was soon to be replaced
by the Persian Empire, which extended from Cyrenaica to
Kashmir, for an extreme length of 3000 miles.
Undoubtedly, the common folk of these empires had
only the vaguest notion of the extent of the realm and
were content to live and die on their own few acres or, on ~
some occasions, to travel from village to neighboring
village. Travelers and soldiers, however, must have had
some concept of the vastness of these empires and of the
still greater vastness of what must lie beyond.
There must have been in the ancient empires those who
occupied themselves with what might be considered the
first cosmological problem facing scholars: Is there an end
to the Earth?
To be sure, no man in ancient times, however far he
traveled, ever came to any actual end of the Earth. At
most, he reached the shore of an ocean whose limits were
beyond the horizon. If he transferred to a ship and sailed
outward, he never succeeded in reaching the end either.
Did that mean there was no end?
The answer to that question depended on the general
shape one assumed for the Earth.
All men, before the time of the Greeks, made the
assumption that the Earth was flat, as indeed it appears to
be, barring the minor irregularities of the mountains and
valleys. If any pre-Greek ancient thought otherwise, his
name has not come down to us and the record of his
thinking has ‘not survived.
 The Earth a AS
‘Yet if the Farth were indeed flat, an end of some sort
would seem an almost foregone conclusion. The alterna-
tive would be a flat surface that would go on forever and
forever—one that would be infinite in extent, in other
words. This is a most uncomfortable concept; throughout
history, men have tended to avoid the concept of endless-
ness in either space or time as something impossible to
grasp and understand and therefore something that cannot
_easily be worked with or reasoned about.
On the other hand, if the Earth does have an end—if it
_is finite—there are other difficulties. Would not people fall
off if they approached that end too closely?
Of course, it might be that the dry land was surrounded
_by ocean on all sides so that people could not approach
the end unless they deliberately boarded a ship and sailed
out of sight of land; far out of sight. As late as the time of
é‘Columbus, in fact, this was indeed a very real fear for
- many seamen.
_ The thought of such a watery protection of mankind
_faised another point, however. What. was to prevent the
ocean from pouring off the ends and draining away from

_ One way out of this dilemma was to suppose that the

sky above was a solid shield as, indeed, it appears to be’,
and that it came down to meet the Earth on all sides, as it
appears to do. In that case, the entire Universe might be
thought to consist of a kind of box, with the sky making
up the curved top and sides, while the flat bottom is the
sea and dry land on which man and all other things live.
‘andmove.
What might the shape and size of such a “box-
“Universe” be?
- To many, it seemed a rectangular slab. It is an interest-
ing accident of history and geography that the first civili-
zations on the Nile, the Tigris-Euphrates, and the Indus
Rivers were separated east and west, rather than north
-and south. Moreover, the Mediterranean Sea runs east
and west. The dim geographical knowledge of early civi-
lized man therefore expanded more easily east and west
_ than north and south. It seems reasonable to picture the
_ “box-Universe” then as considerably longer east-west than
north-south.
The Greeks, however, seemed to have a stronger sense

1 The biblical term “firmament” attests to the primitive belief of

the sky as a “firm” object, a solid substance.
“
 16 The Universe
of geometric proportion and symmetry. They tended to
think of the Earth as a circular slab, with
Greece, of
course, in the center. This flat slab consisted chiefly of
land, with a rim of water (“the Ocean River”) from
which the Mediterranean Sea extended inward to the
center.
By 500 B.c., the first scientific geographer among the
Greeks, Hecataeus of Miletus (birth and death dates un-
known), considered this circular slab to have a diameter
of perhaps 5000 miles at most. This would make the area
of the flat Earth about 20,000,000 square miles. Such a
figure would certainly have seemed ample, and even enor-
mous, to the men of Hecataeus’ time, but it represents
only a tenth of the Earth’s actual surface.
Then, too, what kept the box-Universe, whatever its size
and shape, in place? In the vision of a flat Earth, which
we are considering, “down” means one particular direction
and all things that are heavy and earthy fall “downward.”
Why does the Earth itself not do so?
One might suppose that the material of which the flat
-Earth is composed, the land we stand on, simply continues
downward forever. If so, we are faced once more with the
concept of infinity. To avoid that, people might suppose
instead that the Earth was standing on something. The
Hindus placed it on four pillars, for instance.
But that only postponed the difficulty. On what were
the four pillars standing? On elephants! And on what were
the elephants standing? On a gigantic turtle! And the
turtle? It swam in a gigantic ocean! And this ocean
In short, the assumption of a flat earth, however much
it might seem to be “commonsensical,” inevitably involved
one in philosophic difficulties of the most serious sort.

The Spherical Earth

Indeed, the flat Earth did not even appear to be com-

monsensical, if one used one’s eyes properly. If the Earth
were really flat, then the same stars ought to be visible in
the sky from all points (with some minor differences due
to foreshortening, perhaps). Yet it was the universal ex-
perience of travelers that if one traveled north, some stars
disappeared beyond the southern horizon and new stars
appeared from behind the northern horizon. If one
traveled south, the situation was reversed. This could most
easily be explained by supposing that the Earth curved in
a north-south direction. (Whether there was a similar
 er.

The Earth 17
east-west effect was obscured by the general east-west
“motion of the entire sky, which made one complete turn
every twenty-four hours.)
The Greek philosopher Anaximander .of Miletus
(611-546 B.c.) suggested therefore that men lived on the
surface of a cylinder that was curved north and south. He
was the first man, as far as we now know, to suggest any
shape for the Earth’s surface other than flat, and the
suggestion was perhaps made about 550 B.c.
Yet a cylindrical Earth was insufficient. It was the
experience of men who lived on the seashore and dealt
with ships that vessels heading out to sea did not merely
grow smaller and smaller until they disappeared into “an
infinitesimal point, as would be expected if the Earth were
fiat. Instead, they disappeared while still perceptibly larger
than points and did so hull-first as though they were
moving beyond the top of a hill. This would be exactly
what was to be expected if the surface of the Earth were
curved. What is more, ships disappeared in much the same
fashion no matter toward which point of the compass they
moved. Therefore the Earth was curved not only north-
south, but in all directions equally; and the only surface
that curves in all directions equally is that of a sphere.
It also seemed to Greek astronomers that an eclipse of
the Moon could best be explained by supposing that the
Moon and Sun were on opposite sides of the Earth and
that it was the Earth’s shadow (cast by the Sun) that fell
_ on the Moon and eclipsed it. This shadow was always seen
st to be circular in cross section, no matter what position the
Moon and Sun held with respect to the Earth. The only
solid that casts a shadow with a circular cross section in
all directions is a sphere.

oe
The disappearing hull

Zz sails still visible

a hidden bd
of the Earth ER
we t
theEarth

—
 Meera

Eg x The Universe
_ Close observation, then, would show that the Earth’s
‘surface is not flat but spherical. It appears flat to the
casual glance only because it is such a large sphere that
the small portion of it visible to the eye has a curvature
too gentle to detect.
_ As far as we now know, the first person to suggest that
the Earth was a sphere was the Greek philospher Phi-
lolaus of Tarentum (480- ? B.c.), who made the sugges-
tion about 450 B.c.
The concept of a spherical Earth at once put to rest
any problem of an “end” to the Earth, without introduc-
ing the concept of infinity. A sphere had a surface of finite
size but one without an end; it was finite but un-
bounded.
About a century after Philolaus, the Greek philosopher
Aristotle of Stagira (384-322 B.c.) summarized the im-
plications of a spherical Earth.
One had to consider “down” not as a definite direction
but as a relative one. If it were a definite direction as we
sometimes think it is when we point to our feet, then the
whole spherical Earth might be expected to fall downward
forever—or until it comes to rest on something that is
‘solid and has an infinite downward extension.
_ Suppose, however, we merely define “down” as the
direction pointing to the center of the Earth. When we say
things naturally “fall downward,” we mean they naturally
fall toward the center of the Earth. In that case, nothing
falls off the Earth and the people on the other side of the
globe from ourselves have no sensation of standing upside
down. -
The Earth itself cannot fall because every part of it has
_ already fallen as much as possible and has reached as
closely as possible to the Earth’s center. Indeed, that is
why the Earth must be a sphere, for the sphere has the |
property that the total distance of all parts of itself to its
center is less than that of any solid of the same size but of
a different shape.
By 350 B.c., then, we can say that it was firmly estab-
lished that the Earth was a sphere. This concept has been
accepted ever since by the educated of the Western
world.
The concept was so satisfying and so free of paradox
that men accepted it even in the absence of direct proof.
It was not until 1522 a.p., eighteen centuries after Aris-
totle, when one surviving ship of an expedition originally
led by the Portuguese navigator Ferdinand Magellan’
 The Earth 19
(1480-1521) sailed into port, that anyone actually suc-
ceeded in sailing completely around the Earth, thereby
establishing direct proof that it was not flat.
Today we have established the sphericity of the Earth
on an actual “seeing-is-believing” basis. Rockets climbed
high enough in the late 1940’s to take pictures of large
sections of Earth’s surface that show its visible spherical
curvature.

The Size of the Earth

With the Earth established as a sphere, the question of

its size becomes more meaningful than ever. It would be
difficult to establish the size of a finite flat Earth without
actually pacing it off. A spherical Earth, however, pro-
_ duces effects that vary with the size of the sphere.
If the sphere of the Earth were extremely large, for
instance, the effects produced by its sphericity would be
too small to detect easily. The view of the stars would not
change perceptibly as one traveled a mere few hundred
miles north or south; ships would not disappear hull-first
while they were still large enough to be seen; the cross-
section of Earth’s shadow against the Moon would seem
straight, for its curve would be too gentle to detect.
The mere fact, then, that the effects of sphericity were
noticeable meant not only that the Earth was a sphere but
that it was one of rather limited size—large, perhaps, but
_ not enormously large.
Still, how was one to measure that size with precision?
_ Greek geographers could set a lower limit. By 250 B.c.,
they knew from experience that there was land to the
_ west somewhat beyond the Strait of Gibraltar and other
land to the east as far as India, an extreme distance of
some 6000 miles (well above Hecataeus’ apparently gen-
erous estimate of two and a half centuries earlier). Over
that extent, Earth’s surface clearly had not doubled back
on itself, so the circumference of the Earth- had to be
greater than 6000 miles; but how much greater one could
not say. F
The first to suggest an answer based on observation w:
the Greek philosopher Eratosthenes of Cyrene (276-196
B.c.). He knew, or was told, that on the summer solstice,

2 To be sure, the Earth is not an exact sphere, but the departure

from the exactly spherical is too small to be visible to the eye.
Viewed from space, Earth would seem a perfect sphere.
 The Unters ae
- June 21, when the noonday Sun was as near the zenith as
it ever was, it was actually at the zenith at the city of
Syene, Egypt (the modern Aswan). This could be demon- |
strated by the fact that a stick placed upright in the
ground would then cast no shadow. At the same time, a
stick placed upright in the ground in Alexandria, 500
_tniles north of Syene, would cast a short shadow, one that
indicated the noonday Sun was a little over 7 degrees
south of the zenith. —
If the Earth were flat, the Sun would be virtually
overhead both at Syene and at Alexandria simultaneously.
The fact that it was overhead at one but not the other was
proof, in itself, that the Earth’s surface curved over the
space between tlie cities. The stick in the ground at one
city did not, so to speak, point in the same direction as did
that at the other. While one pointed at the Sun, the other
did not.
_The greater the curvature, the greater the difference in
direction in which the sticks pointed, and the greater the
difference in the shadows. Eratosthenes carefully proved
his calculations by geometry, but we can take the proof
for granted and simply say that if a difference of a little
over 7 degrees is 500 miles, then a difference of 360
Be degrees (the complete turn about a circle’s circumfer-
ence) must be about 25,000 miles to keep the proportion
the same. 3
If the circumference of a sphere is known, then its
diameter is known, too. The diameter is equal to the
circumference divided by yz (“pi”), a quantity equal to
about 3.14. Eratosthenes. therefore concluded that the
_ Earth has a circumference of about 25,000 miles and a
diameter of about 8000 miles.
_The surface area of such a sphere is about 200,000,000 —
square miles, at least six times the largest area of the
known world in ancient times. Eratosthenes’ sphere was
apparently a bit too large for the Greeks, and when later
astronomers repeated his observations and obtained small-
er figures (18,000 miles in circumference and 5700 miles
in diameter for a surface area of 100,000,000 square
miles), those smaller figures were accepted eagerly. The
smaller figures prevailed throughout medieval times, and
Columbus used them to prove that the westward journey
from Spain to Asia was practical for ships of that time. In
fact it was not; his voyage was a success only because the
eure occupied the space where he thought Asia would
stand, —-
ae Pees 4, ok
ape The Earth 21
It was only in 1522 with the return of the single
remaining ship of Magellan’s fleet that the true size of the
Earth was established beyond question and Eratosthenes
was vindicated.
The latest determinations make the circumference of
the Earth at the Equator 24,902.4 miles. The diameter of
the Earth varies slightly in different directions because the
Earth is not an exact sphere, but its average length is
7917.78 miles. Its surface area is 196,950,000 square.
miles.
 CHAPTER 2

The Solar System

The Moon

If the Earth were all the Universe, the Greeks would

have solved the essential problem of cosmology 2000
years ago. The Earth, however, is not all the Universe as
the Greeks well knew. There is also the sky overhead.
_As long as the Earth was assumed to be flat, it was
perfectly possible to consider the sky to be a solid dome
that closed down on that flat-Earth on all sides. The
_ enclosure it would form would not need to be terribly
high, either. If it were ten miles high, for instance, that
_ would suffice to enclose the highest mountains and the
_ clouds.
If, however, the Earth were a sphere, the sky had to be
a second, larger sphere enclosing it. It was the sphere-of
the sky (the “celestial sphere”) that bounded the Uni-
verse, and it would therefore be of great interest to know
its dimensions.
For all that could be told by casual observation, the
_ celestial sphere might still hug the spherical Earth and be
removed only ten miles from its surface in all directions.
If the diameter of the Earth were 8000 miles, then that of
the sky might be 8020 miles.
But let us not be satisfied with merely casual observa-
tion, for the Greeks—and the Babylonians and Egyptians
before them—certainly were not.
The celestial sphere appears to revolve about the Earth
once in twenty-four hours. In so doing, it seems to carry
the stars with it “all in one piece.” That is, the stars do
not shift position relative to one another but remain fixed
in place year after year and generation after generation
(hence the “fixed stars”). It seemed natural to believe that
22
 The Solar System 23
the stars were attached to the vault of the sky like so
many luminous pinheads, and until the seventeenth centu-
_ ry that was, indeed, the common opinion.
However, even prehistoric man must have noticed that
some of the heavenly bodies moved in relation to the stars
and were near one star at one time and near a different
star at another. These bodies, therefore, could not be
attached to the vault of the sky but must be closer to the
Earth than was the sky itself.
There were seven such bodies known to the ancients,
and the names we know them by are, in order of bright-
ness, the Sun,1 the Moon, Venus, Jupiter, Mars, Saturn,
and Mercury. These seven bodies were called “planetes”
(“wanderers”) by the Greeks, because they wandered
among the stars. The word has come down to us as
“planets.”
It was possible to make a stab at judging which of the
planets were closer and which farther. For instance, the
Moon passed in front of the Sun at every Solar eclipse, so
the Moon must be closer to the Earth’s surface than the
Sun is.
To make other distance judgments, the ancients relied on
the relative speeds of the planetary motions among the
stars. (We know from experience that the nearer a mov-
ing object is, the faster it seems to move. An airplane near
the ground moves with frightening velocity, whereas the
same airplane a mile high scarcely seems to move at all,
although it might actually be speeding more quickly then
_than when it Was near us.)
From the relative speeds of their motion against the
stars, the Greeks judged the Moon to be the closest of the
seven planets. The remainder, in order of increasing dis-
tance, were considered to be»Mercury, Venus, Sun, Mars,
Jupiter, and Saturn.
If, then, one wished to determine the distance of the
heavenly bodies, it stands to reason one must start with
the Moon. If the distance of the Moon cannot be deter-
mined, there is little chance of determining the distance of
any other heavenly body.

1 The position of the Sun relative to the stars cannot be observed

directly. It can, however, be observed that the stars visible at mid-
night differ somewhat from night to night. This is taken to mean
that the Sun slowly moves among the stars, blanking out a slightly
different part of the sky each day and leaving aslightly different part
visible at night. It takes 365% days for the Sun to make a complete
circuit of the sky in this fashion.
 ; 24 : The Universe :

‘The first to make a serious attempt to determine the

distance of the Moon was the Greek astronomer Aristar-
---chus of Samos (320-250 B.c.). He made use of observa-
tions during a Lunar eclipse. As the Earth’s shadow fell
across the Moon, one could tell from the curvatures of its
- edge how large its cross section was compared to the size
of the Moon. If it is assumed that the Sun is much farther
away than the Moon, then Aristarchus could use ordinary
geometry to tell how far the Moon must be from the
Earth in order to allow Earth’s shadow to shrink down to
the observed dimensions.
This method was improved and refined a little over a
century later by another Greek astronomer, the greatest
of ancient times, Hipparchus of Nicaea (190-120 B.c.).
Hipparchus came to the conclusion that the distance of
the Moon from the Earth was equal to just about thirty
times the diameter of the Earth. If Eratosthenes’ figure
for Earth’s diameter—8000 miles—is accepted, then the
Moon’s distance from the Earth is 240,000 miles.
This is an excellent figure considering the state of the »
art at the time. The best modern figure for the average
distance of the Moon from the Earth, center to center, is
238,854.7 miles. This is only an average figure because the
- Moon does not move around the Earth in a perfect circle.
It is a little closer to the Earth at some times than at
others. The closest it ever gets to Earth (“perigee”) is
221,463 miles and the farthest (“apogee”) is 252,710
- miles.
Knowing this distance, one can calculate the actual
diameter of the Moon from its apparent size. It turns out
- to be 2160 miles through and to have a circumference,
therefore, of 6800 miles. It is distinctly smaller than the
Earth but is still of respectable size.
The distance of the Moon, once determined, disposed
once and for all of any notion that the sky might be fairly
Close overhead. By Greek standards it was already seen to
. be an absolutely enormous distance. Even the closest
heavenly body was a quarter of a million miles away and
all the others had to be farther, perhaps much farther.
Could one probe further? What about the Sun?
Aristarchus realized that when the Moon was exactly at
the first quarter (or last quarter), it, the Sun, and the
Earth were at the vertices of a right triangle. By measur-
ing the angle separating the Moon and the Sun as seen
- from the Earth, one could then use simple trigonometry to
get the ratio of the distances of the Moon and the Sun.
 me Uae A >See ae aver ed So mee ER ee
ae e.% j ay Le Ie
oPa Tere ec4
‘The Solar System —s-.25.
ors,

ie Then, if the distance of the Moon were determined, that

of the Sun could be calculated.
Unfortunately for Aristarchus, measurements of angles |
in the heavens are rather difficult to obtain without good
instruments, and determining the exact time of the Moon’s
first quarter is far from easy. The theory he worked with
was mathematically perfect, but his measurements were a
trifle off, enough to give him widely inaccurate results. He
concluded that the Sun was twenty times as distant as the:
Moon, If the Moon is 240,000 miles from Earth, the Sun
would then be a little less than 5,000,000 miles away,
which is a gross underestimate (but which served as addi-
tional evidence for the unexpected largeness of the Uni-
verse).
By 150 B.c., then, we can say that in four centuries of
careful astronomy, the Greeks had accurately determined
the shape and dimensions of the Earth and the distance of
the Moon, but had not managed to probe very far beyond
that. They could conclude that the Universe was a huge
sphere that was at least several million miles across, and
at its center they placed an Earth-Moon system possessing
dimensions we still accept today.

The Sun

For 1800 years after Hipparchus, man’s understanding

if of the dimensions of the Universe proceeded no further. ©
‘There seemed no way to determine the distance of any of
the planets other than the Moon, and although a number
of guesses were made as to the distance of the Sun, none
was of any value...
One reason for this lack of progress after Hipparchus
_-Wwas that the Greeks had developed a model of the plane-
_ tary system that was of only limited use. Hipparchus and
those who foilowed considered the Earth to be the center
of the Universe. The Moon and the other planets circled
about the Earth (in a rather complicated fashion) and
beyond them the vault of the stars also circled the Earth.
This system was preserved in detail for posterity in the
works of a late astronomer, Claudius Ptolemaeus, who
lived in Egypt and wrote about 130 a.p. He is popularly
known, in English, as Ptolemy; the “geocentric system”
(“Earth-at-center”’) is often known as the “Ptolemaic sys-
tem” in his honor.
Such a system made it possible for astronomers to
Sa
/
oi.

RP Eg toy
. 26 The Universe | °
calculate the apparent motions of the planets against the
background of the stars with sufficient accuracy for the
needs of the time. It did not, however, supply a model
accurate enough to be of help in the determination of
distances beyond the Moon.
The beginnings of a new model for the heavens was
established by the Polish astronomer Nicolas Copernicus
(1473-1543) who suggested, in a book published in
1543, on the very day of his death, that it was the Sun
and not the Earth that was the center of the Universe. The
planetary system was, according to this view, truly a “Solar
system,” Sol being the Latin word for the Sun.
This had actually been suggested by Aristarchus nine-
teen centuries before, but it had then seemed a radical
notion, too radical to accept. According to such a “helio-
_centric system” (“helios” is the Greek word for the Sun),
the Earth would circle the Sun as the other planets did,
and the whole vast mass of solid rock beneath man’s feet
would be flying through space without his being aware of
it. There would then be six planets rather than seven:
Mercury, Venus, Mars, Jupiter, Saturn, and Earth. The
Sun would no longer be a planet but a motionless center.
The Moon would not be a planet in the same sense as the
rest, for it would he circling the Earth and not the Sun
even in the heliocentric systém. A body circling a planet
was eventually named a “satellite” and that name de-
scribes the Moon.
The Copernican system began to make slow headway in
the minds of astronomers, for by this time the geocentric
view had been revealed to have many deficiencies. The
mathematics involved in calculating planetary positions
was cumbersome by the older system and yielded results
that did not satisfy the painstaking observations of the
new generations of astronomers of early modern times.
The heliocentric system yielded somewhat better results
and simplified the mathematics as well. It was not an
accurate model, however, for Copernicus still imagined
planetary orbits to be combinations of perfect circles,
which proved to be an inadequate point of view.
In 1609, an accurate model was finally advanced. The
German astronomer Johannes Kepler (1571-1630)
studied excellent observations of the position of the planet
Mars, made by his old mentor, the Danish astronomer
Tycho Brahe (1546-1601), and decided at last that the
only geometric figure that would fit the observations
 aE Y Eee ets
‘i og ; s : za s .
fergie mo The Solar anes a 27
would be an ellipse.2 Kepler showed that the Sun was at
one focus of Mars’ orbital ellipse.
The same eventually was shown to be true for all the
planets revolving about the Sun, and for the Moon in its

Ellipse
&. high eccentricity e

minor axis

focus major axis

B. loweccentricity

revolution about the Earth. The orbit was always an

ellipse and the central body was always at one focus of
_ that ellipse.
In 1619, Kepler went on to find that the average
distances of the planets from the Sun were related in a
simple mathematical fashion to their times of revolution
about the Sun. It was easy to measure the times of
revolution and, by comparing them, it was easy to calcu-
late the relative distance of the different planets.
In short, a highly accurate model of the Solar system
could now be drawn, with the orbits laid out in precise
proportion. Unfortunately, however, by comparing times
of revolution, you could only say that one planet might be
twice as far from the Sun as another planet was, but you
could not tell exactly how far either planet was from the
Sun in miles. The model was there but not the scale on
- which it was constructed, Even this, however, gave a new
2 An ellipse is a kind of flattened circle which possesses two foci.
If from any point on the ellipse, a straight line is drawn to each
focus, the sum of the lengths of the two lines is always the same.
 shia Sm te a ae
28 - The Universe
_ idea as to the size of the Solar system, for it could be seen
that the farthest planet known to the Greeks (or to
Kepler), which was Saturn, was nearly ten times as far
_ from the Sun as the Earth was.
If the distance from the Earth to any one planet could
be determined at any moment, it would give the scale.
The distance of all the planets could then be determined.
It remained only to determine one planetary distance
correctly.

Parallax

To determine the distance of a planetary body, use

could be made of a phenomenon known as parallax. This
can be demonstrated most simply by holding one finger
out in front of your eyes against some variegated back-
ground. Hold both finger and head steady and view the
finger first with one eye and then with the other. You will
see the finger shift position against the background as you
switch from one eye to the other. If you bring the finger
closer to your eye, it will shift position across a wider
stretch of the background.
This happens because your two eyes are several inches
apart, so that the line from the finger to one eye forms
a perceptible angle with the line from the finger to the
_ other eye. The two lines, extended toward the back-
ground, will show the finger to be appearing in two
different positions. The closer the finger the greater the
angle and therefore the greater the apparent shift. If your
eyes were wider apart, that too would increase the angle
formed by the two lines from your finger and create a
larger shift of the finger against the background. (The
background is usually so far away that lines from some
point on it to both eyes make an angle too small to
measure. The background can therefore be considered
fixed.)
_ This same principle can be applied to a heavenly body.
The Moon is so far away, of course, that switching eyes
makes no difference. But suppose the Moon is viewed
against the starry background of the heavens from two
observatories that are hundreds of miles apart. The first
observatory might see the Moon as having one edge a
certain angular distance from a particular star; the other
observatory might see the Moon at the same moment with
that same edge a different angular distance from that
particular star.
 | The lassi” oo 2s
or the extent of the Moon’s shift against the starry
background (assuming the stars to be so far away as to
remain fixed regardless of the change in position of the
observatory) and the distance between the observatories
are known, it is possible through the use of trigonometry —
to calculate the Moon’s distance.
This could, in fact, be done, for the Moon’s apparent
shift against the stars with change in the position of the
observer is quite large. Astronomers standardize this shift
for the case in which one observer is viewing the Moon on
the horizon while another is viewing it directly overhead.
The base of the triangle is then equal to the radius of the
ee
er
Pe
Earth, and the angle at the Moon is the “equatorial
horizontal parallax.” Its value is observed to be 57.04
minutes of arc or 0.95 degrees of arc. This shift is
_
indeed sizable, for it is equal to twice the apparent diame-
ter of the full Moon. It can be determined with sufficient
accuracy, therefore, to give a good value for the Moon’s |
distance. This distance, calculated by parallax, gave good
“agreement with the distance calculated by the earlier meth-
od that. involved the Earth’s shadow during a Lunar
eclipse.
' Unfortunately, the planets were so far distant that the
change in position from observatory to observatory under
conditions prevailing in 1600 produced a shift against the
starry background that was too small to measure accur-
ately.
- Then, in 1608, came the invention (or reinvention) of
the telescope by the Italian scientist Galileo Galilei (1564-
1642). The telescope had the effect of magnifying the
small shifts involved in parallax. A shift too small to
_ detect with the naked eye became large enough to measure
easily by telescope.
In 1671 the first good telescopic determination of a
planetary parallax was made. One of the observers was
Jean Richer (1630-1696), a French astronomer who led
a scientific expedition to Cayenne in French Guiana. The
other was the Italian-French astronamer Giovanni Dom-
enico Cassini (1625-1712), who stayed behind in Par-
is. Each observed Mars at as nearly the same time as
possible and noted its position in relation to nearby stars.
ne
Se
ee
Me
ee
Oe
eo
Sy
re
From the difference in position that was observed, and the
® One degree (1°) is equal to 1/360 of the circumference ofa circle.
There are sixty minutes (60’) in one degree, and sixty seconds (60”)
in one minute.
 “> <A e

30 ‘The Universe
Parallax
background of stars
* * * * * *« *

AY Soy
\ :5 ty
Hy farge parallax ,
‘iy en . t
a oe J
e ‘
t \ / / apparent
@ !/ positions
area ak * of Venus
Moon against \ rf Venus
x a / against
stars
\ , el stars
‘\\ / wi
Ss
\ A I
j

2 ae
\ 7

Moon a l

\ he i
ba I
/ XN It

/
CAN
4

4 7, '

4 ie x t
las ‘ :
observatory ea \ pobservatory
No.1 /7 \ 1 No.2
f \l e
surface of the Earth

known distance from Cayenne to Paris, the distance of

Mars at that moment could be calculated.
Once that measurement was made, the scale of Kepler’s
model was known, and all other distances in the Solar
system could be calculated. The distance of the Sun from
the Earth, for instance, was calculated by Cassini to be
87,000,000 mles. This is about six million miles short of
the actual figure, but it was excellent for a first attempt
and may be considered the first useful determination of
the dimensions of the Solar system.*
4 The Sun’s parallax cannot be determined directly with any accu-
racy, for to do so would require an astronomer to work with one
fixed point on the surface or edge of the globe. There are no fixed
points on its blinding surface, and the tiny shift involved in its
parallax (only about 8.8” or 1/400 that of the Moon) cannot be made
out. The planets, with much smaller and dimmer globes, are more
easily and accurately observed in some specific position, and a small
parallax is much more easily detected. From that, the Sun’s distance
can be calculated much more satisfactorily than by any direct deter-
mination.
 ae PE a Len ae} ar ht es , 7

vrais F “The Selar Seem | 31

During the two centuries after Cassini’s time, somewhat
more accurate determinations of planetary parallaxes were
made. Some of them involved Venus which, on occasion,
passes directly between the Sun and the Earth and can be
seen as a small, dark, circular body crossing the Sun’s
glowing disc. Such “transits” took place in 1761 and 1769,
for instance. If the transit is carefully observed from.
different observatories, the time at which Venus first
makes apparent contact with the Sun’s disc, the time it
_ ]eaves, and the duration of the transit all will be found to
vary. From the amount by which they vary and from the
distances between the observatories, the parallax of Venus
can be calculated; from that, its distance; and from that,
the distance of the Sun.
In 1835, the German astronomer Johann Franz Encke
(1791-1865) used the data from Venus transits to
deduce a distance of 95,370,000 miles for the Sun. This
was a little too high but only bya little over two million
miles.
The difficulty of getting still more accurate values lay in
the fact that Mars and Venus did show up as tiny gobes
in the telescope and that this slightly blurred attempts at
_ fixing the precise position of the planet. Venus was partic-
ularly disappointing because it had a thick atmosphere
which produced optical effects that slightly obscured the
exact moment of contact with the Sun’s disc during a
transit.
Then came an unexpected break. In 1801, the Italian
astronomer Giuseppe Piazzi (1746-1826) found a small
planet circling in an orbit between Mars and Jupiter. He
called it Ceres; it proved to be somewhat less than 500
miles in diameter. As the century progressed, hundreds of
even smaller plants were found, all circling between the
orbits of Mars and Jupiter. These were the “asteroids.”
Then, in 1898, the German astronomer Karl Gustav Witt
(1866-1946) discovered Eros, an asteroid which strayed
out of the “asteroid belt.” Part of its orbit passed within
that of Mars and approached LEarth’s orbit rather
closely.
In 1931, Eros was scheduled to approach Earth to a
distance of only about two-thirds that of Venus, the
nearest of the large planets. This close approach meant an
_ unusually large and easily measured parallax. Further-
more, Eros is so small (only an estimated fifteen miles in
 32 - .The Universe
its longest diameter) that it held no atmosphere to fees ‘its
outlines and, despite its closeness, remained a mere point
of light. This meant that its position could be determined
with a great deal of accuracy.
A vast international project was set up. Thousands of
photographs were studied and eventually, from the paral-
lax and position of Eros, it was determined that the Sun
was just a bit less than 93,000,000 miles from the Earth.
This is an average distance, since the Earth moves about
the Sun in an ellipse and not a circle. At its closest
approach to the Sun (“perihelion”), the Earth is at a
distance of 91,400,000 miles from the Sun; at the farthest
(“aphelion”), it is at a distance of 94,600,000 miles.

The Size of the Solar System

In recent years, something even better than parallax has

turned up. Techniques have been developed whereby very
short radio waves (“microwaves”) of the type used in-
tadar beams, can be sent out into space, bounced off a
planet such as Venus, and the reflection received and
detected. The microwaves move at a velocity that is ac-
curately known, and the time lapse between emission and
reception can be measured accurately. The distance
traveled by the microwave beam on the round trip and,
therefore, the distance of Venus at a given time, can be
determined with greater precision than will be yielded by
any parallax determination.
In 1961 such microwave reflections were received from
Venus. Using the data thus obtained, the average distance
of the Sun from the Earth was calculated to be 92,960,-
000 miles.
Making use of the Keplerian model, the distance of all —
the planets, either from the Earth at some particular time
‘or from the Sun, can be calculated. It is more convenient
to give the distance from the Sun since this changes less
with time, and in a less complicated way, than does the
distance from the Earth.
The distance can Bs expressed in four ways of particu-
lar interest.
First, it can be given in millions of miles. In the United
States and Great Britain, miles are the common unit for
measuring long distances.
Second, it can be given in millions of kilometers. The
 The Solar System _ 33
kilometer is the common unit for measuring long distances
in civilized nations other than the Anglo-Saxon ones, and
it is used by scientists everywhere, even in the United
States and Great Britain. A kilometer is equal to 1093.6
yards or 0.62137 of a mile. Its length can be placed, with
reasonable accuracy, at 4 of a mile.
Third, to avoid millions of miles or kilometers, the -
average distance of the Sun from the Earth may be set
equal to an “astronomic unit” (abbreviated A.U.). Dis-
tances can then be given in A.U., where 1 A.U. is equal to
92,960,000 miles or 149,500,000 kilometers. It is reason-
ably accurate to say: 1 A.U. = 150,000,000 kilome-
ters.
Fourth, the distance may be given in terms of the time
it takes light (or any similar radiation, such as mi-
crowaves) to cross that distance. Light travels, in a vacu-
um, at a velocity of 299,792.5 kilometers per second, and
this can be set at the even value of 300,000 kilometers per
second without too much inaccuracy. This velocity is
equivalent to 186,282 miles per second.
A distance of approximately 300,000 kilometers can
therefore be set equal to “1 light-second” (the distance
traveled by light in one second). Sixty times that, or
18,000,000 kilometers is “1 light-minute” and sixty times
that or 1,080,000,000 kilometers is “1 light-hour.” We are
not too far off, if we think of a light-hour as equal to a
billion kilometers.
With this in mind, let us consider the planets known to
the ancients and prepare a list of their average distances
from the Sun in each of these four units:

Average Distance from the Sun

Million Million §Astronomic Light-
Planet miles kilometers units hours
Mercury 35.9 57.9 0.387 0.0535
Venus 67.2 108.2 0.723 0.102
Earth 92.9 149.5 1.000 0.137
Mars 141.5 227.9 1.524 ~ 0.211
Jupiter 483.3 178.3 5.203 0.722
Saturn 886.1 1428.0 9.539 1.321
ED

From the time of Cassini on, then, it was known that

the diameter of the Solar system from one end of Saturn’s
 <e ah ara
Pc =
a
CP

34 The Universe oe
orbit to the other, was nearly three billion kilometers (or |
- two billion miles). The diameter of the imaginary sphere
that included the planets known to the Greeks was not a
matter of only millions of miles as the Greeks of Hippar-
chus’ time had suspected, but was thousands of millions.
Even this turned out to be inadequate. The diameter of _
planetary orbits was doubled at a stroke in 1781, when the
German-English astronomer William Herschel (1738-
1822) discovered the planet Uranus. The diameter was
doubled again, in two steps, when the French astronomer
Urbain Jean Joseph Leverrier (1811-1877) discovered
Neptune in 1846 and the American astronomer Clyde
' William Tombaugh (1906- ) discovered Pluto,
The distances from the Sun of these outer members of-
the Solar System are given below:

Average Distance from the Sun

Million Million Astronomic Light-
Planet miles —- kilometers units hours
Uranus 1782 2872 19.182 2.66:
Neptune 2792 4498 _ 30.058 4.26
Pluto 3671 5910 39.518 - $47

If we consider the orbit of Pluto, rather than that of

Saturn, we see that the diameter of the Solar system is not
_ three billion kilometers but twelve billion. A ray of light,
_ Which can travel a distance equal to the circumference of
_ the Earth in 1/7 of a second and pass from the Earth to
the Moon in 1% seconds, would take nearly a half a day to
span the Solar system. The sky has indeed receded unut-
terably since Greek times.
In fact, there is every reason to believe that Pluto does
not mark the boundary of the Sun’s dominion. Nor does
this mean we must postulate still further undiscovered
planets (although it is quite possible that small, very
distant ones may exist). There are known bodies, which
are easily seen on occasion, that undoubtedly recede to
distances greater than that of Pluto at its farthest. ;
This fact was known even before the discovery of
Uranus extended the boundaries of the strictly planetary
portion of the Solar system. In 1684, the English scientist
Isaac Newton (1642-1727) had. worked out the law of
universal gravitation. This law explained the existence of
 The Solar System 35
Kepler’s model of the Solar system in a straightforward
mathematical manner and made it possible to calculate
the orbit of a body revolving about the Sun even when
that body was only visible through part of its orbit.
This, in turn, made it possible to deal with comets,
fuzzily luminous bodies that appeared now and then in the
heavens. Through ancient and medieval times, astrono-

Size of the Solar System

mers had thought that comets arrived at irregular intervals

in motions bound by no natural law—and the population
in general was sure that their only purpose was to foretell
disaster.
Newton’s younger friend, the English astronomer
Edmund Halley (1656-1742) tried, however, to apply
gravitational calculations to the comets. He noted that
certain spectacular comets seemed to show up in the skies
at intervals of seventy-five or seventy-six years. He made
the assumption, in 1704, that these comets were actually a
single object moving in a regular orbit about the Sun—an
orbit so elliptical that most of it was enormously far from
the Earth. When the comet was distant from the Earth it
was not visible, but every seventy-five or seventy-six years
 ned. Pen «i {ye a ee
: : As eet

36 The ‘Universe
-it was in the potion of its orbit near the Sun (and the
Earth) and was then visible.
Halley calculated the orbit and predicted that the comet
would once more be visible in 1758. The comet did indeed
appear then (sixteen years after Halley's death) and it has
been called “Halley’s Comet” ever since. —
At its closest approach to the Sun, Halley’s Comet is
only about ninety-million kilometers from the Sun, so that
it moves slightly inside the orbit of Venus. At its farthest

Halley’s comet

recession from the Sun, however, Halley’s Comet moves

out to a distance some 31% times that of Saturn. At its
aphelion, it is 5300 million kilometers from the Sun, which
means that it recedes to a point well beyond Neptune’s
orbit. By 1760, then, the astronomers were perfectly
- aware that the Solar system was much larger than the
Greeks had imagined and did not need the discovery of
additional planets to tell them so.
Indeed, Halley’s Comet is one of those that are relative-
ly close to the Sun. There are some comets that move
about the Sun in such fantastically elongated orbits that
they return only after intervals of many centuries or even
millennia. They recede from the Sun to distances not of
mere billions of kilometers but, in all likelihood, hundreds
of billions of kilometers. According to a suggestion made
 The Solar System 37
in 1950 by the Dutch astronomer Jan Hendrik Oort
(1900- ), there may even be a vast cloud of comets
that remain at such immense distances throughout their
orbits, and are therefore never seen. ;
It follows, then, that the Solar system may well eve an
extreme diameter of something like a thousand billion
kilometers; that is, a trillion (1,000,000,000,000) kilome-
ters or even more. It would take a ray of light forty days
to span this distance. The diameter of the Solar system
can thus be viewed as more than 1 “light-month.”
Nor is the relative insignificance of the Earth a matter
of distances only. The four outer planets: Jupiter, Saturn,
Uranus, and Neptune expand into globes of measurable
size when viewed through the telescope. Once the distance
of these bodies came to be known, the apparent size of the
globes could be turned into absolute measurements. Each
of these outer planets turns out to be a giant compared
with the Earth. And the size of the Sun makes it a giant
compared with even the largest of the planets.

Equatorial Diameter
; ‘ Earth’s diam-
Object Miles Kilometers eter—1 -
Earth 7927 12,753 1.00
Neptune - 27,700 44,600 3.50
Uranus 29,200 47,000 3.68
Saturn 75,100 121,000 9.5
Jupiter : 88,700 143,000 a
Sun 864,000 1,392,000 1

Then, too, each of the giant planets has a satellite

system that dwarfs the Earth’s. The first of the outer
satellites to be discovered were the four largest of Jupiter,
which were seen in 1610 by Galileo through his first
primitive telescope. The last large satellite to be discov-
ered was Neptune’s satellite Triton, detected in 1846 by
the English astronomer William Lassell (1799-1880).
' Additonal small satellites were discovered; a second satel-
lite of Neptune, Nereid, was noted as late as 1949 by the
Dutch-American astronomer Gerard Peter Kuiper (1905+

). The total number of known satellites in the Solar

system, including our Moon, is now thirty-one, but it is
very likely that additional small ones will be discovered.
 ~ 2 ea. ; tS years

38 The Universe _
A notion of the size of some of the satellite systems
compared with that of the Earth is given in the following
table:

~ Average Distance of
Farthest Satellite from
its Planet
Number of Farthest
Planet Satellites Satellite Miles Kilometers
Earth 1 Moon ~ 238,900 385,000
_ Uranus 5 Oberon 368,000 591,500
Neptune 2 Nereid 3,461,000 §,540,000
Saturn 9 Phoebe 8,053,000 12,905,000
Jupiter 12 Hades* 14,700,000 23,600,000
© Unofficial name
—

CHAPTER 35

The Stars

The Vault of the Sky

If the Solar system were all there was to the Universe,

the essence of the problem of its size would have been
solved by 1700. However, the Solar system is not all there
is to the Universe. There remain the stars,
In 1700, it was still possible to believe that there was a
solid vault bounding the Universe, one that contained the
stars as luminous dots, and that this solid vault lay (pos-
sibly) not far outside the confines of the Solar system.
Kepler’s views on the matter were something like that, for
instance. ;
Parallax measurements of the sort that had revealed the
scale of the Solar system in the seventeenth century, were
useless in connection with the stars and did nothing to
disturb this “solid-sky” view. The separation of two neigh-
boring stars never varied measurably no matter which
observatories on Earth’s surface made the measurements.
A shift in base over the full width of the Earth caused no
detectable change in the position of any star. This is not
surprising, since the stars, even if barely outside the orbit
of Saturn, would be too distant to show a parallax large
enough to be measured in 1700.
The Earth’s surface was not, however, the final resource
of the astronomer in this respect. The Earth’s diameter -
was not, quite 8000 miles, but the entire globe moved
through space as it pursued its revolution about the Sun.
One side of that orbit was 186,000,000 miles from the
other side. If, then, the positions of the stars were charted
on some particular evening and then again on a particular
evening a half-year later, the astronomer would be
viewing the stars from two positions separated by 23,600
39
wie
 POPS a Oe Se a Tye the, Sa rere
3 . veo ;

40 ~+~°~—-*‘The *Universe
times as great a distance as that represented by the full
width of the Earth’s diameter. The expected parallax
would be increased by that same ratio. Indeed, ‘the posi-
- tion of a particular star might shift slightly each night as
the Earth moved and, in the course of the year, the star
- would mark out a tiny ellipse in the sky—a kind of image
of the Earth’s orbit. The angular distance from the edge
of the ellipse to its center would be the “stellar paral-
lax.”
This device cannot be used for the planets because in
the course of the year, each planet marks out a rather
complicated path of its own across the sky, one that masks
any parallactic shifts caused by the Earth’s motion. Trying
to sort out the planet’s own motion from that imposed on
it by the Harth’s would be extremely complicated and
would yield results that were not as accurate as those
arrived at by ordinary parallaxes. The stars, however,
remain virtually fixed in place over the course of the year
so that a useful parallactic shift might, conceivably, be
obtained.
Such a shift, however, was not obtained. Indeed, the
1800’s dawned without astronomers having been able to
detect the stellar parallax of a single star. -
Several reasons might be advanced for this.

Stellarparalledt

It might be, of course, that Copernicus and Kepler were

wrong after all and that the Earth did not move around
the Sun, but was thé motionless center of the Universe. If
so, no parallax over the course of a year would be
expected. Indeed, when Copernicus first advanced the heli-
ocentric theory, the absence of stellar parallaxes was used
 The Stars 41
by his opponents as a strong argument against him. How-
ever, there were too many other reasons for accepting the
heliocentric theory and at last it became firmly embedded
in astronomic thought, despite the absence of stellar paral-
lax. The Earth does move and the absence of parallax
must be explained in other ways.
Despite the Earth’s motion, parallaxes could still be
absent if all the stars were virtually at the same distance.
_ A parallax is visible only when the position of some
_ Telatively near object is sighted against some relatively
distant one. If there were really a solid vault of the sky,
the stars would all tend to shift in identical fashion with a
comparatively small change in the position of the observer
(and the shift of Earth in its orbital positions is small
' compared with even the small Universe some men ac-
oo in 1700). There would then be no visible paral-
ax.
But can the theory of the solid vault of the sky really be
accepted? Arguments could be advanced for believing that
stars might be at varying distances from us, even at widely
_ varying distances. They might be distributed through a
broad expanse of space, and there might be no solid
boundary to the Universe at all.
For one thing, the stars vary in brightness. This is
_ obvious to anyone who looks at the night sky. Hipparchus
_ had been the first to attempt to reduce this difference in
_ brightness to some sort of system. He divided the stars
_ into six classes or “magnitudes.” The brightest stars he
_ considered to be first magnitude, the next brightest second
Magnitude, and so on down to the sixth magnitude which
_ Yepresents the dimmest stars that can be made out by the
naked eye.
Modern astronomers measure the brightness of individ-
- ual stars with instruments unavailable to the ancients and
have defined the magnitudes with mathematical precision.
A difference of five magnitudes (say between 1 and 6)
represents a ratio, in brightness, of 100. In other words a
star of magnitude 1 is 100 times as bright as one of
magnitude 6. A difference of a single magnitude in bright-
hess represents a ratio of 2.512, therefore, since
2.512 X 2.512 X 2.512 X2.512 X2.512 is equal to 100.
Accurate measurements of brightness make it possible
to define the magnitude of a star to fractions and even to
a tenth of a magnitude. Thus, the bright star Aldebaran
has a magnitude of 1.1, while Regulus, slightly dimmer,
has a magnitude of 1.3. The still dimmer Polaris (the
 42 The Universe
North Star) has a magnitude of 2.1, while Electra, one of
. the stars of the Pleiades, has a magnitude of 3.8. .
There are a number of stars brighter than Aldebaran
that have magnitudes less than 1.0. Procyon has a magni-
tude of 0.5 and the still brighter Vega one of 0.1. The
very brightest stars in the heavens must be assigned nega-
tive magnitudes. Canopus has a magnitude of —0.7 and
Sirius one of —1.4.
It is even possible to fit the planets, the Moon, and the
Sun into this system. Venus, Mars, and Jupiter are all
brighter at times than even the brightest star. Jupiter can
attain a magnitude of —2.5; Mars a magnitude of —2.8;
and Venus a magnitude of —4.3. The full Moon has a
magnitude of —12.6 and the Sun one of —26.9.
Working in the other direction, stars dimmer than the
sixth magnitude exist, too, although they are invisible to
the naked eye. Galileo, when he first turned his telescope
to the heavens in 1609, found hundreds of stars he could
not see before. Stars of magnitudes 7, 8, 9, and so on, up
the scale of numbers and down the scale of brightness, are
recorded and studied. Our largest telescopes can make out
myriads of stars with magnitudes more than 23.5.
If all stars were of the same intrinsic brightness (or
“luminosity”), we could suppose that the difference in
apparent brightness was entirely a matter of distance.
Closer stars look brighter than distant ones, just as closer
lamp posts seem more brilliantly lit than distant ones.
There was no reason in 1700 to assume that the stars
were all of the same intrinsic brightness. It might just as
well be that the stars were all at the same distance from
the Earth and that the difference in brightness was a real
one; bright stars just happened to be more luminous than
dim ones, as some light bulbs are intrinsically more lumi-
nous than others.
Another factor, however, dented the equal-distance hy-
pothesis with considerably greater effectiveness.
The ancient Greeks had recorded the relative positions
of the visible stars. The first to do so in a systematic way
was Hipparchus who, about 134 B.c., had recorded the
positions of over 800 stars. His was the first important
“star map,” and it was preserved for posterity by Ptolemy,
who increased the number of stars it contained to more
than a thousand.-
In 1718, Halley, studying the positions of the stars,
noted that at least three stars, Sirius, Procyon, and Arc-
turus, were not in the spots recorded by the Greeks. The
 eS : oa he So es By Fu ae = Sree < i =a

‘The Stars Mae

_ difference in position was so great that it was unlikely that
either the Greeks or Halley could have made a mistake.
Halley found Arcturus, for instance, to be a full degree
(twice the apparent width of the full Moon) away from
the position recorded by the Greeks.
It seemed clear to Halley that these stars had moved.
They were not truly fixed stars after all but had “proper
motions” of their.own. The proper motions of the stars
were exceedingly slow compared with those of the planets
and ‘did not make themselves evident from day to day or
even from year to year. But from generation to gener-
_ ation, the slow proper motion of the stars succeeds in
displacing them perceptibly against the sky.
The mere existence of proper motions among the stars
was a terrible blow to the solid sky hypothesis. At least
some of the stars were not attached to the vault, and the
feeling grew at once that none of the stars were; that,
indeed, there was no vault.
Nevertheless, although the stars were not actually at-
tached to any solid object, it was possible that they were
all at virtually the same distance. There might still be a
rather narrow shell of space through which stars might be
_ distributed without being attached to anything.
This possibility was rendered unlikely by the fact that
only a small minority of stars were found to display
_ measurable proper motions. To be sure, a star might move
without its motion being apparent to us, even over long
_ periods, if it moved in a direction parallel to the line of
_ sight. Yet if stars moved in any direction, at random, then
at least as many ought to move more or less at right
angles to the line of sight as parallel to it. In that case, if
any stars at all showed a measurable proper motion, at
_ least half of them ought to. Still, the keenest search
revealed measurable proper motion to be very much the
exception.
But what if we abandon the assumption that all stars
_ are roughly at the same distance? Let us suppose, instead, .
that they are at widely varying distances. If all of them
_ are moving at the same velocity, or over some reasonably
narrow range of velocities, and in any direction at ran-
dom, we can come to certain conclusions.
Of those stars moving more or less parallel to the line
of sight, none will show measurable proper motions,
whether near or far. Of those moving more or less at right
angles to the line of sight, those that are nearer will have
larger proper motions than those that are farther.
 44 The Universe
This association of comparatively large proper motion
with comparative closeness is borne out by the fact that it
is the bright stars that are most likely to show such proper
motion. The first three stars found to possess proper
motion—Sirius, Procyon, and Arcturus—are among the
eight brightest stars in the sky. It is clear that a nearby
star is apt to appear bright as well as to possess a proper
motion. According to this view, it makes good sense to
find a measurable proper motion only among very few
stars. It may well be that only the nearest stars are close
enough to show even a tiny proper motion and that
beyond them lie vast myriads of stars too distant to show
any perceptible motion at all, even over many centuries.
By the mid-eighteenth century, then, it was perfectly
clear that there was neither a solid sky nor even a rela-
tively narrow shell of stars. Rather the stars were widely
spread out through an indefinitely vast expanse of space.

Proper motion and distance

small angle
equivalent
to small proper
motion
 The Stars 45
This view had, in fact, been suggested by a few medieval
scholars such as the German philosopher Nicholas of Cusa
(1401-1464), but what had then been speculation had
now become deduction from careful observation.

A Multiplicity of Suns

But then, if the Earth is moving, and if the stars are

distributed at all distances, why do not the nearer ones
show a stellar parallax with reference to the farther
ones?
One explanation of this was so forcefully pieioas that it
received almost instant acceptance. Even the nearest stars
were so distant that their parallaxes were too small to
measure with the instruments of 1800. (Indeed, Coperni-
cus had maintained just this in answer to those critics who
held up the absence of stellar parallax as evidence against
the revolution of the Earth about the Sun.)
In fact, it was possible to make rough estimates of the —
distances of the nearer stars by several logical lines of
argument.
Suppose, for instance, we assume that the stars moved,
in reality, as rapidly as do the planets. If so, we can
estimate the distance at which such a motion is reduced in
appearance to the tiny, almost immeasurably small crawl
represented by a star’s proper motion.
Thus, the star with the greatest of all proper motions is ~
“Barnard’s star” (so named because it was discovered in
1916 by the American astronomer Edward Emerson Bar-
nard [1857-1923]). Its proper motion is 10.3 seconds of
arc per year.
~ To understand how little this is, remember that a circle
is divided into 360 degrees, each degree into 60 minutes,
and each minute into 60 seconds. A single second of arc
represents 1/1,296,000 of the circuit of the sky. Since the
Moon is 31 minutes of arc in diameter, a single second
fTepresents 1/1860 the diameter of the Moon. Jupiter
looks like a mere point of light to us, but the diameter of
its globe as seen by the naked eye is from 30 to 50
seconds of arc, depending on how close to us it is.
When we say, then, that Barnard’s star moves 10.3
seconds of arc a year, we mean that it moves less than a
hundredth the diameter of the Moon, or only about a
quarter the diameter of the Jupiter pinpoint in one year.
And yet so comparatively large a proper motion is this
that the star is sometimes called “Barnard’s runaway
 46 The Universe |
star.” It is much more usual for proper motions to be 1
second of arc per year or’less.
Suppose, then, that Barnard’s star actually ote as
quickly across our line of sight as does Earth in its journey
about the Sun—18.5 miles per second. In one year, it
would travel 585,000,000 miles. For such a length to
mark off only 10.3 seconds of arc, Barnard’s star would
have to be something like ten trillion miles from us—
thousands of times as far away from the Sun as Pluto is.
And if this were so, the parallax: Barnard’s star would
display would be only about 1 second of are. If Barnard’s
star moved more quickly than does the Earth (as, in
actual fact, it does), then it would be still more distant
and would have a correspondingly smaller parallax.
If a star makes a small ellipse in the sky, one that is
about 1 second of arc across or less, astronomers would
be hard put to detect the fact. The size of the ellipse
would be roughly that of a twenty-five cent piece viewed
from a distance of four miles.
To be sure, proper motions of 1 second of arc per year
can be observed without too much trouble, for proper
motions go constantly in the same direction and pile up
year by year. After a century, a star moving at 1 second
of arc per year will have crawled nearly 2 minutes of arc
across the sky and that shift is easy to observe by tele- .
- scope. A parallactic motion, on the other hand, is eternally
back and forth and does not pile up with the years.
If the stars are, in fact, tens of trillions of miles away at
the very least, it becomes-a matter of interest that we can
see them at all. Even as supernally bright an object as the
Sun would be seen as only a tiny point of light from a
distance of ten trillion miles. It would appear to be merely
a star. Conversely, any star seen from the distance of our
Sun would increase so enormously in brightness as to
appear another sunlike object itself.
We must, therefore, look on the Sun as a star which
differs from all the other stars chiefly in that we see it
from a distance of millions of miles, rather than trillions
of miles as in every other case. We must also look on the
Universe as a vast assemblage of suns of which our own is
merely one.
Suppose, then, that the star Sirius were actually as
luminous as the Sun and that its lesser light is only the
result of its enormous distance from us. Sirius has a
magnitude of —1.6 and the Sun a magnitude of —26.9.
The Sun is 25.3 es brighter and each magnitude
ae eee
A . The Stars 47
represents an increase of brightness by a ratio of 2.512.
This means that the observed brightness of the Sun is
13,200,000,000 times that of Sirius.
The brightness. of a light source varies inversely as the
square of the distance. That is, if a light source is placed
at two times its former distance, its brightness is reduced
to (4%)? or %; if it is placed at five times its former
distance, its brightness is reduced to (4%)? or 1/25.
For Sirius to shine with only 1/13,200,000,000 times
‘the brightness of the Sun, it must be 115,000 times as far
as the Sun, since 115,000X115,000 is equal to 13,200,-
000,000. We know the Sun to be 93,000,000 miles from
us and by this line of reasoning, Sirius ought to bealittle
over ten trillion miles away. In short, whether we argue
from the proper motion of stars or from their brightness,
we end with the same colossal distance and are forced to
expect the same heartbreakingly small parallax. —
To express a distance as trillions of miles is mean-
ingless. Instead we can turn to units of distance as meas-
ured by the length traveled by light in a particular unit of
time. I have already said that a light-hour is equal to
1,080,000,000 kilometers, which comes to about 670,000,-
000 miles. Suppose we consider the “light-year” instead.
At 300,000 kilometers per second, light will travel 9,-
440,000,000,000 kilometers (or 5,880,000,000,000, miles)
in one year. We can say, then, with rough accuracy, that a
ees isequal to ten trillion kilometers or six trillion

Making use of this unit, we can see that Sirius (by the
line of argument we have used above) is ten trillion miles,
or nearly 2 light years, from us. Since Sirius is certainly
one of the closer stars, whether one uses the criterion of
brightness or of proper motion, one must conclude that
stellar distances must all, at the very minimum, be meas-
ured in light-years.
Consider this in comparison to the Solar system. A ray ©
of light that travels from the Sun to the Earth in 8
minutes and from the Sun to distant Pluto is 5% hours
could reach even the nearest stars only after a journey of
_ years!

The Search for Stellar Parallax

As telescopes grew larger and better with the passing

_ decades, hopes grew that the tiny stellar parallaxes might
_ be detected and that the distances of at least some of the
 48 The Universe
nearer stars might actually be determined directly rather
than merely deduced from more or less rickety -assump-
tions. Unfortunately, the closer stars were observed, the
more complex the problem seemed to grow.
For instance, the close stars, which might have the
largest parallaxes, were also those most likely to display
considerable proper motion. This meant that a parallactic
shift would have to be disentangled from the proper
motion.
The situation became more complex in 1725 when the
English astronomer James Bradley (1693-1762), in the
course of careful determination of stellar positions, noted
small shifts that did indeed cause a certain star to mark
- out a tiny ellipse in the sky in the course of the year. The
trouble was that the star did not change position in accord-
ance with what one would expect of a parallactic shift.
Although the star should have appeared farthest south in
December, it reached that point in March instead. The
three-month delay continued all round the ellipse.
By 1728, Bradley had been able to show that this was
the result of Earth’s motion into the beams of light falling
on it from the stars.
The analogy is usually made to a shower of rain. If the
rain is falling vertically and a man with an umbrella is
standing still, he need only hold the umbrella directly
overhead. If, however, the man walks forward, he will
move into some raindrops that have just cleared his um-
brella. He must, therefore, angle the umbrella foward

Aberration

2
ae
ee
—
ew
ee
Se
——KK
eee
See
we
me
em
eo
ee
ee
ae
a

Td
ee
et
em
OE
em
'
ose keepinga
clear region
SS
eS
wee
eS Doe

@otionless walking fipping the

; into the umbrella
eaindrops
 The Stars - | 49
slightly. The faster he moves the farther forward he must
angle his umbrella, and if he changes direction, he must
change the direction of the umbrella.
In the same way, since the Earth is moving through a
“rain” of light-rays, the astronomer must angle his tele-
scope very slightly by an amount that depends on the
velocity of the Earth compared with the velocity of light.
As the Earth changes its direction in revolving about the
Sun, so the astronomer must change the direction of his
telescope. In the end the star marks out an ellipse, but it is
not a parallactic one. :
This phenomenon is called the “aberration of light,” and
its effect on stellar position is larger than would be
produced by parallax. A star can be displaced over a
distance of as much as 4 seconds of arc by aberration,
and if parallax is to be detected, it must be disentangled
from this considerably larger effect.1
Bradley also discovered that the direction of the axis of
the Earth relative to the stars moves slightly back and
forth in a period of 18.6 years, as though the Earth were
nodding. The motion is called “nutation” (from the Latin
word meaning “nodding”). This motion is reflected in
slight changes in the apparent position of the stars, and
this, too, must be disentangled from any parallactic. shift
that might exist.
The search for stellar parallax, which led Bradley to the
phenomenon of aberration, lured Herschel (the discoverer
of Uranus) into a still more exciting discovery.
Herschel thought, he might detect the tiny changes pro-
duced by stellar parallax with greater ease if he chose two
stars that were extremely close together. The assumption
was that such stars just happened to lie very close to the
same line of sight but that one might be enormously more
distant than the other. In that case, the nearer star would
seem to move with respect to the other. (The suggestion
that this be done had first been made by Galileo.)
1 Although Bradley was trying to detect parallax when he detected
aberration, this certainly cannot be considered a failure. The exist-
ence of the aberration of light is as good a piece of evidence for the
motion of the Earth about the Sun as the detection of stellar paral-
lax would have been. If the Earth were really a motionless body at
the center of the Universe, there would be neither parallax nor aber-
ration. By the eighteenth centry, the geocentric theory scarcely
needed another coffin nail, but Bradley’s discovery provided one. In
addition, the discovery of aberration allowed Bradley to estimate
the velocity of light with better accuracy than had been possible
previously.
 rests kn,tage rk eee Te ie ees 7

50 The Universe
Herschel found shifts almost at once, but the shifts
‘could not be due to stellar parallax. A parallactic shift
(after you subtract the effects of proper motion, aberra-
tion of light, and nutation) would have to produce a
closed ellipse in the space of one year, but the shifts
Herschel observed did not do this. Herschel did find his
stars to be. marking out ellipses, but they did so over an
interval of time much longer than a year.
By 1793, he was convinced that what he was observing
was the case of two stars circling each other about a
-.common center of gravity. The shifts he saw depended
only on the gravitational attraction of one star for the
other and had nothing to do with parallax. The two stars
were not independent entities that seemed close together
only because both happened to be near the same line of
sight (although this is certainly possible in some cases),
but they were stars that were close together in actual fact

The motion of Sirius

1912

19 1930 & 1980

197 1940
& 1990
1960
1950
 The Stars 51
as well as in appearance. Herschel had discovered “binary
stars.” Before the end of his career, he had discovered 800
such binaries.
Indeed, the motions produced by virtue of the existence
of a close neighbor star could be detected even when the
neighbor star itself could not. In 1844, the German as-
tronomer Friedrich Wilhelm Bessel (1784-1846), who
was studying the star Sirius, found that its proper motion
was not the straight line it ought to be. It was a wavy line. »
Had there been a-companion star, it would have been easy
to account for the wavy line as an elliptical motion about
the companion, superimposed on the straight line of the
proper motion. However, no companion star was visible.
Bessel suggested that there was a “dark companion,” a
burnt-out ember, perhaps, of a once-bright star; a com-
panion that could no longer be seen but that still existed
and made its gravitational effect felt.
Bessel’s suggestion was borne out in 1862. In that year
the American astronomer Alvan Graham Clark (1832-
1897) detected a dim spot of light near Sirius. This turned
out to be Bessel’s “dark companion.” It was not complete-
ly dark, after all, but had a magnitude of 7.1.

The Distance of the Nearer Stars

As astronomic instruments continued to improve, and as

the possible roadblocks in the form of complicating mo-
tions were better and better understood, hope continued to
grow that stellar parallax might yet be detected. Attempts
to achieve this goal grew more determined and sophisti-
cated, and in the 1830’s three separate attacks were made
on the problem.
In South Africa, the Scottish astronomer Thomas Hen-
derson (1798-1844) was carefully plotting the position
of Alpha Centauri. This is the third brightest star in the
sky (but is too far south to be seen from North Temper-
ate latitudes) and therefore, he hoped, one of the closer
ones.
In the Baltic provinces of Russia, the German-Russian
astronomer Friedrich Georg Wilhelm von Struve (1793-
1864) was plotting the position of Vega. This is the fourth
brightest star, and it, too, might be one of the close
ones.
In K6nigsberg in East Prussia, Bessel (who in the next
decade would discover Sirius’ companion) was taking an-
other tack. It was not brightness he was using as a
 z : ne ern
52 The Universe
criterion for closeness, but rapid proper motion. For the
purpose, he selected a star known as 61 Cygni (in the
constellation Cygnus, the Swan) which had been found to
‘have a proper motion of 5.2 seconds of arc per year. This
was, at the time, the largest proper motion known.? :
Bessel judged that 61 Cygni, despite its lack of bril-
- liance, must be close. He measured the distance between
61 Cygni and each of two very dim (and therefore, he
hoped, very distant) neighbor stars with a new instrument
called a heliometer which was capable of making very
accurate measurements of angular distance. He continued
this for over a year.
Eventually, all three astronomers succeeded in determin- .
ing the parallax of the stars they were studying. Bessel ©
was the first to announce his results, in 1838. Henderson,
who had completed his work even earlier than Bessel,
waited for his return to England and did not announce his
results until 1839. Struve joined in 1840. wv

’ Tt turned out, following certain improvements in the

original measurements in later years, that Alpha Centauri
(really a triple star, with two members of the system
sizable and very close to each other, and a third very
distant and very ‘dim companion) has a parallax of 0.760
second of arc. This parallax, only three-fourths of a sec-
ond of arc, has turned out.'to be the largest one ever
found. Therefore the Alpha Centauri system is the closest
known system to ourselves outside our Solar system. It is —
4.29 light-years away.
As for 61 Cygni, that, too, is a double star. Its parallax
. turned out to be 0.29 second of arc, and its distance 11.1
light-years. Vega was the hardest of the three targets to
handle because it was the most distant. It is about 27
light-years away.
With parallaxes determined, a second unit of distance
came into vogue. The distance at which a star would have
a parallax of one second would be a convenient unit to
use. This unit is a “parallax-second,” universally abbrevi-
ated to “parsec.” One parsec is equal to 3.26 light-years,
or 200,000 astronomic units, or 19 trillion miles, oree
trillion kilometers.

2 It is still the largest proper motion known among the stars visible
to the naked eye. The half-dozen or so stars that have been found to
move still more rapidly are all so dim as to be visible only in tele-
scopes—but they are dim because of their tiny size, not because of
their great distance.
 ‘ The Stars : §3
The distance to some of the nearer stars is tabulated as
follows:

Distance
Star Light-years Parsecs
(a ES

Alpha Centauri 4.29 1.32

Barnard’s star §.97 1.84

a a aa
Wolf 359 7.74 2.38

eni 3 ‘|
Procyon 11.3 3.48
Kapteyn’s star 12.7 3.87
Van Maanen’s star 13.29 4.06
Altair So 4.82

It is clear then that the Solar system is isolated within a

vast emptiness. The rough estimates made earlier in the
chapter turn out to have been overconservative. Barnard’s
star, estimated to be ten trillion miles distant on the
assumption that it moves as quickly as the Earth, actually
moves more quickly and is thirty-five trillion miles away.
Sirius, estimated to be ten trillion miles distant on the
assumption that it is as luminous as the Sun, is actually
considerably more luminous and is fifty trillion miles
away. Even the very nearest star, Alpha Centauri, is 25
. trillion miles away.
The nearest star is nearly 7000 times as far from the
Sun as Pluto is. Imagine a circle drawn with the Sun at
the center and Alpha Centauri at the circumference. If
this is scaled down to the point where the radius is 9 yards —
in length (so that a good-sized house can be placed within
it), the orbit of Pluto about the Sun will be a tiny ellipse
with an extreme diameter of 1/10 of an inch.
Once we know the actual distance of a star, we can
calculate its actual luminosity from its apparent magni-
tude; or, conversely, determine how bright it would ap-
‘pear at any distance. The brightness of a star at the
arbitrary distance of 10 parsecs (or 32.6 light-years) is
called its “absolute magnitude.”
If the Sun were at a distance of 10 parsecs (instead of
at 0.000005 parsec, as it is), it would have a magnitude of
4.9 and would be only a dim star. If Sirius were placed at
10 parsecs instead of the 2.67 parsecs it actually is, it too
would appear dimmer than it does, but not by very much.
Its absolute magnitude is 1.4.
 yi
SN SePO SIS is een see? eae eke © =
54 "The Universe - 3
At equal distances from us, Sirius would be 3.5 magni-
tudes brighter than the Sun. Since 1 magnitude means a
ratio of 2.512 in brightness, we can say that Sirius is
. (2.512)8-5 or twenty-five times as luminous as_ the
Sun.
Figures on the absolute magnitude and luminosity of
some well-known stars are given below:

Star Absolute Magnituds Luminosity

: (Sun = 1)

at0D 27 ie.
"
Altair 2.3 10.9
Sirius 1.4 25.0
Vega . 0.5 57.5
=— 03 120
Capella 0.3 120
Regulus — 0.7 173
Aldebaran ~ Se P on
Beta Centauri — 5.2 12,000
Antares - — 5.4 13,000
Rigel = 7.1 25,000
~ Deneb —7.1 25,000

In other words, the Sun, which is the most glorious

object in our heavens and which Copernicus thought to be
the center of the universe, is not only merely a star but
merely an ordinary star. There are other stars that are.
thousands of times as luminous as the Sun.
However, we need not be too abashed. While the Sun is
not the brightest star, neither is it the dimmest.
Indeed, of the fifty stars closest to the Sun, only three—
Sirius, Procyon, and Altair—are distinctly more luminous
than the Sun. The two larger stars of the Alpha Centauri
system (Alpha Centauri A and Alpha Centauri B) are
each about as luminous as the Sun. The remaining forty- ~
five are all dimmer than the Sun and some are much
dimmer, ;
CHAPTER 4

The Galaxy
Olbers’ Paradox

By 1840, astronomers had finally plumbed-the distance

of the stars, at least of the nearest ones, and found them
to be something more than a parsec distant.
_ The next question, a rather inevitable one, is: Where do
the stars end? How far is the farthest star? After all, the
Earth has a finite surface, the Solar system takes up a
finite region of space. In graduating to a new “plateau,”
are we still in the realm of the finite? Or must we finally
find ourselves face to face with the infinite—that concept
that has troubled scholars from the beginning.
If we restrict ourselves to that portion of the Universe
that can be seen with the unaided eye, the Universe is
certainly finite. It is now known that, in the region nearest
ourselves at least, the average distance between stars is
about 3 parsecs (or 10 light-years). We also know that
there are about 6000 stars visible to the naked eye. Sup-
pose that those stars are all that exist and that they are all
separated by the average distance. In that case, a sphere
with a diameter of about 100 parsecs, or 330 light-years,
would enclose all 6000 stars.
This is large enough, certainly, by any ordinary human
standards. A sphere 100 parsecs across has a diameter of
nearly two quadrillion (2,000,000,000,000,000) miles, and
its size would have shocked and flabbergasted any astrono-
mer who lived before 1600—and many who lived after-
ward.
However, 6000 stars are not nearly all there are. As
soon as Galileo turned his first telescope on the heavens in
1609, he found large numbers of dim stars that were
invisible to the unaided eye. Every time the telescope was
55
 56 The Universe —
improved, a new crop of still more copious and still
dimmer stars was uncovered.
There seemed, at first, no sign of any end, and in 1800 ©
one might well have veered from a Universe of 6000 stars
and a 100-parsec diameter to one of an infinite number of
stars and no final boundary at all. In the latter case, the
question “How far is the farthest star?” would have no
answer because one might say, “There is no farthest
star!”
As usual, though, the thought of the infinite was repul-
sive to mankind. The attack on a possible infinite universe
of stars was carried on along two fronts, one of theory
and one of observation.
The theoretical reasons for doubting the existence of an
infinite universe of stars arose from suggestions made by
the German astronomer Heinrich Wilhelm Matthaus Ol-
bers (1758-1840). In 1826, he advanced what has come
to be called “Olbers’ paradox.” To explain that, let us start
with the following assumptions:
1. The Universe is infinite in extent.
2. The stars are infinite in number and evenly spread
throughout the Universe. ;
_ 3. The stars are of uniform average luminosity
throughout the Universe.
Imagine the Solar system at the center of such a Uni-.
verse and consider that Universe to be divided into thin,
_ concentric shells like those making up an onion.
The volume of such thin shells would increase as the
square of the distance. If Shell A were three times as far
from us as Shell B, Shell A would have 32, or nine times
the volume of Shell B. If the stars are evenly spread
through all the shells (Assumption 2 above) then Shell A
with nine times the volume would have nine times the
stars that Shell B would have.
On the other hand, the light of the individinat stars
would fall off as the square of the distance. If Shell A is
three times as distant as Shell B and contains 9 times as
many stars as Shell B, each individual star (assuming
uniform average luminosity throughout—Assumption 3
above) in Shell A-is only (4%)? or 1/9 as bright as the
individual stars in Shell B.
We conclude then that Shell A has nine times as many
stars as Shell B, and that each star in Shell A is-1/9 as
bright as each star in Shell B, so that the total light
delivered by Shell A to the Solar system is 9 X 1/9 that
 soi yt es
The Galaxy 57
of Shell B. In short, Shell A and Shell B deliver an equal
amount of total light to the Solar system.
The same can be argued for every other shell. It follows
then that if there are an infinite number of shells (As-
sumption 1 above), the total light reaching us would be
infinite, except that the nearer stars would block the light
of the farther ones. Even allowing for such shielding, the
whole sky would glow like the surface of one huge, bright
sun, Yet this certainly is not the case.

Olbers’ paradox

tev dtbuton ofsrthough inte

mamber
ofshold

Olbers suggested that one way out of this paradox

might be found in the existence of dust clouds in space,
clouds that absorbed the light from really distant stars, so
that only the light from the relatively nearby stars reached
us. This, however, is not good enough. If the dust clouds
absorbed the light, they would gradually heat up until they
were emitting as much light as they were absorbing. The
amount of light reaching us would still be infinite.
There must be something wrong with Olbers’ assump-
tions. The Universe must not be infinite in extent, or, if it
 58 ; The Universe
is, the stars must not be infinite in number. One would
expect, instead, that there would bea finite (although very
large) number of stars, spread out over a finite (but very
vast) space.
This conclusion, based on Olbers’ reasoning, jibed per-
fectly well with the careful astronomic observations being
carried on at this time by William Herschel.

Herschel’s Lens

At least one of the assumptions behind Olbers’ paradox

seems shaky on the face of it.
We might assume that the stars are evenly spread
throughout space, but certainly observations of that por-
tion of the Universe that can be seen from Earth do not
seem to support this view.
Stretching completely around the sky through the con-
stellations Orion, Perseus, Cassiopeia, Cygnus, Aquila,
Sagittarius, Centaurus, and Carina is a softly luminous
band. It fades out in the bright lights of a modern city,
but on a moonless night deep in the country, it is a
beautiful sight.
The ancients, who did not possess the dubious blessing
(from an astronomer’s point.of view) of the electric light,
were well aware of this luminous band. To them it looked
like a milky cloud. The Greeks called it “galaxias kyklos”
(“milky circle”) and the Romans called it “via lactea,”
which translates directly into the English “Milky Way.”
From the Greek version of the name comes the English
“Galaxy.”

In 1610, Galileo looked at the Milky Way through his

primitive telescope and found it to be not a featureless
luminous cloud but a vast collection of very dim stars, as,
indeed, some philosophers of pretelescopic days had
speculated it might be.
It seems clear then that there must be many more stars
in the direction of the Milky Way than in any other
direction. Indeed, even the bright visible stars occur in
somewhat greater profusion in the direction of the Milky
Way than in other directions. This certainly goes against
the assumption of evenly spread-out stars.
Herschel was well aware in his systematic survey of the
heavens that there were more stars in some directions
than in others, but he was not satisfied with a merely
qualitative judgement. In 1784, he decided to count the
 The Galaxy 59
stars and see exactly how they varied in profusion from
place to place.
To count all the stars all over the sky was, of course, an
impractical undertaking, but Herschel realized it would be
quite proper to be satisfied with sampling the sky. He
chose 683 regions, well-scattered over the sky, and
counted the stars visible in his-telescope in each one. He
found that the number of stars per unit area of sky rose
steadily as one approached the Milky Way, was maximal
in the plane of the Milky Way and minimal in the direc-
tion at right angles to that plane.
How could this be explained? Possibly the stars just
happened to be spaced more closely as one approached
the Milky Way. But why should this be so? There seems
no simple way of explaining why this steady change in
spacing should occur. It seemed to Herschel much more
reasonable to suppose that the stars were spread out with
uniform separations, but over a volume of space that was
not spherical and therefore not symmetrical in all direc-
tions.
Suppose the stars were evenly distributed through a
volume of space shaped like a lens or a grindstone and
that our own Sun were near the center of the mass. If we
sighted along the long diameter of the grindstone, we
would see a number of bright stars close to us, and behind
them large masses of distant and therefore dim stars, and
still farther beyond even more numerous masses of still
more distant and still dimmer stars and so on. The numer-
ous stars we would see in the far distance would be too
dim to be visible individually, but en masse they would
lend a pale milky luminosity to the sky and make up the
Milky Way. ;
On the other hand, if you looked away from the long
diameter of the grindstone, you would look through small-
er and smaller thicknesses of stars. You would then see
only the nearby rather bright stars but beyond them would
be no mass of distant stars and no milky luminosity.
To Herschel, then, it seemed that the stars of the
Universe formed a finite “sidereal system” possessing a
definite shape. (Gradually, the word Galaxy came to
mean the entire sidereal system, rather than. merely the
visible Milky Way, so that we could speak of our Sun as
one of the stars of the Galaxy.)
From the number of stars he could see in the various
directions and making use of his assumption as to even
separation, Herschel even tried to make some rough deci-
 60 The Universe
sions as to the size of this sidereal system. He suggested
that the long diameter of the lens-shaped Galaxy was
about 800 times the average distance between two stars
(which he took to be the distance between the Sun and
Sirius—and here he happened to be correct, a least for
those stars in the neighborhood of the Sun). The short
diameter of the Galaxy he took to be 150 times the mean
distance.
What Herschel was envisaging, then, was a Galaxy that
could hold 300,000,000 stars, 50,000 times as many as
could be seen with the naked eye. Moreover, if the aver-
age distance between stars is taken as 10 light-years (al-
though it must be remembered that actual stellar distances
were not determined till sixteen years after Herschel’s
death), the Galaxy was 8,000 light-years in its long diam-
eter and 1,500 light-years in its short.
Furthermore, since the Milky Way seemed to encircle
the sky and to be at least fairly bright on all sides, it
seemed reasonable to suppose that the Sun was some-
where near the center of the Galaxy.
Herschel’s observations and Olbers’ reasoning effectively
killed the notion of an infinite Universe for a century. The
labors of the nineteenth century astronomers to count and
record the stars in greater, numbers and with greater
accuracy merely succeeded in refining the details of Her-
schel’s overall picture.

Lens-shaped star system

Laborious star-counting by eye reached its climax with

the star-map called the Bonner Durchmusterung (“Bonn
Survey”), which began to be published in 1859 under the
 The Galaxy 61
supervision of the German astronomer Friedrich Wilhelm
August Argelander (1799-1875) who worked at the
University of Bonn. This map eventually recorded the posi-
tion of about half a million stars.
However, by the second half of the nineteenth century
_ the necessity of pinpointing each star by eye was passing.
Photography was developed and applied to astronomy. A
photograph of a starfield froze that section of the sky in
perpetuity and allowed star counts to be made in comfort
and at one’s convenience.
One who made great use of photographed star-fields
was the Dutch astronomer Jacobus Cornelius Kapteyn
(1851-1922). Like Herschel, he sampled the sky. He
went further in one respect. He attempted a systematic
count of the stars for each increasing magnitude.
If the number of stars was infinite, there should be a
steady increase in the total number present in each suc-
ceeding shell of space about us (if we revert to the picture
we drew in connection with Olbers’ paradox, see page
57), since each successive shell as we move outward is
larger than the one within and can hold more:stars. Since
_Stars generally appear dimmer with distance, there should
be a steady increase in the number of stars with decreas-
ing brightness.
- Kapteyn observed, however, that the rate of increase
was not steady but began to fall off in the higher magni-
tudes. This meant that the stars were beginning to thin out
in the very distant shells, and Kapteyn was able to make a ©
rough estimate’ as to the distance of those final shells in
which the stars finally petered out.
His results confirmed Herschel’s view of a lens-shaped
Galaxy with the Sun at or near the center. Kapteyn’s
figures for the dimensions of the Galaxy were higher than
Herschel’s, however. In 1906, he estimated the long diam-
eter of the Galaxy to be 23,000 light-years and the short
diameter to be 6,000 light-years. By 1920, he had further
raised the dimensions to 55,000 light-years and 11,000
light-years. This final set of dimensions involved a Galaxy
_ with 475 times the volume of Herschel’s.

The Moving Sun

The sidereal system pictured by Herschel represented

another blow at man’s estimate of his own importance.
In ancient times, man tended to accept himself very-
literally as the hub of the Universe. The Universe was not
 62 ‘The Universe
only geocentric, with the Earth—man’s home—the im-
movable center of everything; it was homocentric, with
man the measure of all things.
Once Copernicus had done his work, and the heliocen-
tric theory was put forth and slowly accepted, it became
more difficult to attach the proper importance to man.
After all, he inhabited only one planet of many, and man’s
planet, moreover, was not the largest nor the most spectac-
ular by far. Earth could not begin to compare with
Jupiter for size or with Saturn for beauty.
Nonetheless, to the astronomers of the seventeenth and
eighteenth century, the Sun seemed the immovable center
of the Universe, and the Sun still belonged to us. It was
the source of light and heat, the fount of life on Earth.
And yet, as the notion of the solid vault of the sky
faded out, it came to seem less and less likely that the Sun
could have this overwhelming importance. With stars
spread throughout a vast region and the Sun itself only a
star among stars, why should the Sun be the center of the
Universe any more than the Earth was?
Furthermore, if one accepted a Galaxy of the size
suggested by Herschel, one that was populated by hun-
dreds of millions of stars, how could one seriously
maintain that our Sun counted for very much among so
many spread out so far.
As the proper motions of more and more stars were
observed, there seemed no indication that the stars gener-
ally were moving in some grand revolution about the Sun.
The motions seemed to be random in nature, so it became
more and more tempting to suppose that all stars were
moving in this more or less random fashion (like individu-
al bees in a large swarm) and that those stars which
showed no proper motion were either too far away to
show one until they had been observed for many more
centuries, or else happened to be moving directly toward
us or away from us, so that a crosswise proper motion
could not be detected. :
If this were so, then it was reasonable to suppose that
the Sun was moving, too. Why should it alone be motion-
less in a Universe of moving stars? Herschel reasoned thus
in 1783 and set about trying to determine what the Sun’s
motion might be.
Suppose the Sun were surrounded by stars spread even-
ly through space. Those near the Sun would be seen as
separated by comparatively large distances, while those far
away would seem to be much closer together. We see the
 , Mt guage Pee rie ey ae a ae

re ‘The Galaxy 63
same phenomenon if we look at the regularly spaced trees
of an orchard, or men standing in carefully ordered ranks
and files. It is a common effect of perspective.
If a group of stars, then, were brought closer to the
Sun, without any alteration of their positions relative to
each other, they would seem to have become more widely
spread apart. Therefore, in the process of approaching,
they would seem to be moving apart. On the other hand,
if a group of stars receded, they would, by the same
token, seem to be moving together.
If the Sun were moving through the Galaxy, the stars
lying ahead of it in its direction of motion would seem, on
the whole, to be approaching the Sun and therefore to be
moving apart. (This effect would be masked in part by the °
fact that the stars are not standing still but have motions
of their own in all directions—but it would not be com-
pletely masked.) The stars lying behind the Sun in the
_direction opposite that in which it is moving would seem, .
on the whole, to be receding from the Sun and therefore
to be moving together. Finally, the stars lying at right

Motion of the Sun

 64 The Universe
angles to the direction of the Sun’s motion would have the
largest proper motions and these would tend, on the whole
to be in the direction opposite that taken by the Sun.
Not many proper motions had been determined in Her-
schel’s time, but by making use of those that were known,
Herschel was able to suggest on the basis of this reasoning
that the Sun was indeed moving. It was moving toward a
spot in the constellation Hercules.
This proved to be not too bad an estimate. Many more
proper motions have been determined in the century and a
half following Herschel’s work, but the “Apex” (the point
toward which the Sun seems to be moving) is now consid-
ered to be not far from the point Herschel determined.
According to the best observations now available, it seems
to be located in the constellation of Lyra, which is next to
Hercules. The Sun is moving toward the Apex (relative to
the closer stars) at a velocity of twelve miles per sec-
ond.

Star Clusters

The fact, then, that the Sun seems to be at or near the

center of the Herschel-Kapteyn model of the Galaxy is
not to be considered of great significance. It would seem to
be merely a fortuitous circumstance; had mankind made
its astronomic observations during another age, far in the
past or far in the future, it might have found itself located
toward one end or the other of the Galaxy.
It is not much of a sop to human vanity to think that
the Sun (and therefore the Earth and man himself) is at
the center of things only as a matter of accident. Yet even
this much grew shaky and uncertain as Kapteyn polished
the final details of his model in the first two decades of the
twentieth century.
The Herschel-Kapteyn model-was in trouble as the
result of evidence that arose in connection not so much
with stars individually as with groups of stars.
Even to the naked eye, such groups seem to exist. The
best known group is the “Pleiades,” a small cluster of
moderately bright stars in the constellation of Taurus, the
Bull. There are nine stars in the cluster that are bright
enough to be made out by the unaided eye, but some are
too closely spaced to be made out separately. The average
eye can see six or seven. (The cluster is sometimes called
the “Seven Sisters.”’)
When Galileo turned his telescope on the Pleiades in
 The Galaxy 65
1610; he found he could easily count 36 stars in the
group, and modern photography. shows at least 250, with
the total count probably close to 750.
The Pleiades represent a true association of stars and
not merely an accidental view of a number of stars at
- varying distances in nearly the same line of sight. Bessel
demonstrated this in 1840 when he showed that the vari-
ous members of the cluster had proper motions of a5)
seconds of arc per century in the same direction. If they
were independent stars, it would be entirely too much to
ask of coincidence that they all be moving in the same
direction and at the same velocity.
Astronomers have estimated that the average distance
between stars in the Pleiades cluster is only one-third that
of the average distance between stars in our own neigh-
borhood. The whole group is now known to be some 400 ©
light-years from us and to be spread out over a region of
space some 70 light-years in diameter.
Although the Pleiades are the most beautiful cluster
visible to the naked eye, they are only the feeblest token
of the spectacles made visible by the telescope.
Without knowing it, the French astronomer Charles.
Messier (1730-1817) had a glimpse of these greater
glories while searching for considerably lesser objects.
Messier was a comet-hunter and, in his lifetime, discov-
ered quite a few. However, he grew tired of reacting to
fuzzy objects in the sky which were permanently fixed and
were not comets. In 1781, he made a careful map of.
about forty such objects in order that he, and other ~
comet-hunters, might know their location and learn to
ignore them. Eventually, Messier and others increased the
number of objects on the list to a hundred.
Among them, for instance, was one fuzzy, starlike object
that had first been observed by Halley in 1714. Because it
was thirteenth on Messier’s list, it is sometimes called
M13. When William Herschel studied M13 several dec-
ades later, with a much better telescope than the one that
had been available to Messier, he realized he was not
looking at a mere blur of light but at a densely packed
spherical conglomeration of stars.
The Pleiades consisted of a group of comparatively
widely spaced stars and was therefore called an “open ~
cluster,” whereas the object, M13, was a closely spaced
globe of stars and was therefore called a “globular clus-
ter.” Because M13 is in the constellation of Hercules, it is
now frequently called the “Great Hercules Cluster.” A
 LA a. me Tttiee ‘PR ie i a” “Scars > 22 Re rs
aS
wy Dea matte Chara te, CE A eS Pe
. - ; . 4 ; * a

The Universe

, Gionalse cluster consists not of hundreds of stars but of

thousands. Some 30,000 stars have actually been counted—
in the Great Hercules Cluster, and the total must be well
over 100,000, perhaps even close to a million. Toward the
center of the cluster, stars must be distributed with an
average separation of— considerably less than 1 light-
year.
Nor is this the only globular ctsters Several others are
- included on Messier’s list, including M3 in the constella-
tion Canes Venatici (the Hunting Dogs) and M22 in the
‘constellation Sagittarius (the Archer). About a hundred’
such globular clusters are known.
Oddly enough, the globular clusters are not distributed _
evenly over the entire sky—something first remarked on in
the early nineteenth century by John Herschel (1792-
- 1871), son of the famous William Herschel, and himself
_an astronomer of note.
_ Almost all the globular clusters are located in one
_hemisphere of the sky and, indeed, a third of them are
crowded into the single constellation of Sagittarius, which
makes up only 2 percent of the sky. John Herschel be-
lieved this could not be accidental, but had to have some
significance.
_ What that significance might be eluded astronomers for
- the next century, partly because the actual position in
_ space of these globular clusters was not known. They were
far too distant to have a measurable parallax, and until
the twentieth century, no other method for determining
stellar distances was known.

Variable Stars
A proper method of handling the problem of globular
_ clusters was found in the early twentieth century in con-
nection with a certain variety of “variable star.” (A varia-
ble-star is one that periodically varies in brightness.)
The history of variable stars is surprisingly short, con-
sidering that a few stars bright enough to be seen by the
unaided eye vary appreciably in brightness. No comments
concerning such variations in brightness have come down
to us from the ancient astronomers. Indeed, the Greek
view, as expressed by Aristotle, was that everything in the
heavens was permanent and unchangeable. This weighty
- official view did not, however, alter the fact that variable
stars exist and that they attest to the fact that there is
change in the heavens.
 : The Galaxy 67
The most notable variable among the naked-eye stars is
Beta Persei, second brightest star in the constellation of
Perseus. It dims and brightens quite perceptibly. Every
two days and twenty-one hours,: it rapidly loses a full
magnitude of brightness for a brief period, and then as
rapidly regains the loss.
Neither the Greeks nor the Arabs (the latter being the
great astronomers of the Early Middle Ages) made mention
of this although individuals may have noted it and been.
uneasy about it. The Greeks, in their fanciful description
of the stars of this constellation, pictured Beta Persei as
located in the head of Medusa, which Perseus is holding.
Medusa was a demon with living, hissing snakes for hair |
and a face so horrible that anyone who beheld it turned to
stone. Beta Persei was sometimes called the “Demon
Star” because of this, a sign that something might have
been considered to be wrong with it. Then, too, the
Arabic name for the star is Algol (“the ghoul”), which
again indicates something chilling.
In 1782, the English astronomer John Goodricke
(1764-1786) studied Algol in detail and suggested that a
dim star was circling it with its orbit nearly edge-on to the
Earth. Every time it circled Algol, it passed between that
bright star and the Earth and cut, off part of its light. The

slight-curves of eclipsing binaries

 The ‘Universe eta eee
suggestion was not taken seriously at the time, be addi-
tional evidence since then -has shown it to be almost
certainly correct. Algol is now accepted as the best-known
example of an “eclipsing binary.” Its light does not really
wax and wane; it is merely regularly blocked off.
Light variability due to eclipses has nothing to do with
_the internal structure or properties of a star. If the com-
panion of Algol revolved about Algol in another plane and
did not intercept its light on the way to the Earth, Algol
would not be considered a variable.
- Quite different is the case of the star Omicron Ceti,
located in the constellation of Cetus (the Whale). This
star was first carefully observed in 1596 by the German
astronomer David Fabricius (1564-1617). At its -
brightest, this star may attain a magnitude-of 2 and at its
dimmest it becomes too dim to be seen by the naked eye.
It received the name of “Mira” (“wonderful”) as a result.
_ Its period of variation is about eleven months—that is, its
periods of peak brightnesses come about eleven months
apart. This is rather long for a variable star, and Mira is
therefore said to belong to the class of “long-period varia-
bles.”
Mira, unlike Algol, is truly variable. It actually grows
_ dimmer and brighter, astronomers have concluded. It is
_ therefore also classified as an “intrinsic variable.” :
. Another example of an intrinsic variable is Delta
Cephei, the fourth brightest star in the constellation Cephe-
us. It differs from Mira considerably. In the first place,
' Delta Cephei’s perod of variation is short, 5. 37 days, and
in the second, it is regular.
Other variable stars like Delta Cephei have been discov-
. ered, each with short and regular periods of variation. The
periods range from two to forty-five days, with periods in |
the neighborhood of a week being very common. The
manner in which the brightness increases‘and decreases is
distinctive, too, and all these stars are grouped together as
“Cepheid variables” or “Cepheids,” after the name of the
first to be studied.
Although the Cepheids were interesting stellar curiosi-
ties, they did not seem at first-to have any great signifi-
cance, That view changed sharply in 1912, when the
American astronomer Henrietta Swan Leavitt (1868-
1921) began to locate and systematically study hundreds
of Cepheid variables in the Small Magellanic Cloud.
 RP EA aieOey ON ey F :
"The Galaxy — 69
The Small Magellanic Cloud is one of two areas of
luminosity (the other is the Large Magellanic Cloud),
which look like isolated patches of the Milky Way. They
are. located so far south as to be invisible to observers in
the North Temperate Zone. They were first described in
1521 by the chronicler accompanying Magellan’s voyage
of circumnavigation of the’ globe—whence their names.
The Magellanic Clouds were not studied in detail till
_ 1834 when John Herschel observed them from the astro-
nomic observatory at the Cape of Good Hope. Like the
Milky Way, they represented assemblages of large num-
bers of dim stars—dim, presumably, because of great
distance.
In fact, the Clouds are so far from us that variation in
distance between stars on the near side and on the far side
can be expected to be relatively unimportant. As an analo-
gy, we can say that all the people in Chicago are about
the same distance from Times Square in New York; the
people in eastern Chicago are a little closer than those in
western Chicago, but the difference is too small to be
significant in comparison with the total distance.
This means that all the stars and, in particular, all the
Cepheids in the Small Magellanic Cloud may be taken as
being about the same distance from us. If one Cepheid in
the Cloud seems to-be brighter than another, it can only
be because it is actually brighter—more luminous—than
the other. No artificial difference in brightness need be
expected to be imposed on them because one is much
closer than another. ’
In her studies of the Cepheids of the Small Magellanic
Cloud, Miss Leavitt noted that the brighter the Cepheid —
variable, the longer its period. A Cepheid in the Cloud
with a magnitude of 15.5 had a period of two days; one
with a magnitude of 14.8 had a period of five days; one
with a magnitude of 12.0 had a period of 100 days.
Apparently, there was some regular relationship between
luminosity and period.
This relationship ought to be true of the Cepheids of
our own neighborhood, too, as well as for those in the
Magellanic Clouds. (Scientists generally find it profitable
to assume that a relationship that holds in one place or
. under one set of conditions also holds in another place or
under another set of conditions—at least until there is
evidence to the contrary.) Why, then, was it not evi-
dent?
ee AO ’ The Universe Oe
The trouble is that the period is related to-the luminosi-
ty of the Cepheid, and in our own neighborhood the
luminosity may be masked as a result of distance. A very
luminous Cepheid with a long period may be so far away
as to appear faint, whereas a much less luminous Cepheid
with a short period may be quite close to us and appear
bright. In such a case, bright stars can appear to have
short periods and dim stars long ones. Indeed, the confu-
sion of distances makes it seem that there is no connection
whatever between brightness and period. As a matter of
fact, there is not. The connection is between luminosity
and period, and from the apparent brightness of a star,
one cannot tell the actual pata aes unless one also knows
the distance of a star.
Unfortunately, in 1912 the distance of not one single
Cepheid was known. Only the measurement of parallax
could be used to determine distances and that had its
limitations. The farther the.star, the smaller the parallax,
_ and the more difficult it is to measure. Even today, the
* parallactic method is impractical for distances —greater
than 100 light-years, and there is no Cepheid variable as
close to us as that. he closest is at least 300 light-years
away.
Miss Leavitt noted the luminosity-period relationship in
the Small. Magellanic Cloud not because she knew the _
distance of the Cloud but because within the Cloud dis-
tances did not-matter. Brightness within the Cloud was_
__ proportional to luminosity, and therefore the luminosity-
_ period relationship showed up as an easily. observed
_ brightness-period relationship.
Once this relationship had been discovered, however, it
could be applied to our neighborhood and used as a
yardstick for distances far beyond those that could be -
plumbed by parallax.
Suppose, for instance, that two Cepheids were observed
to have equal periods but that one seemed brighter than
the other. Their luminosity would have to be the same
since their periods were the same, so the difference in
apparent brightness would be entirely the effect of dis-
tance. It would be easy to calculate how much more
distant one Cepheid would have to be than the other to
account for the difference in brightness.
If two Cepheids had different periods, the difference in
luminosity could be calculated. The difference in apparent
 CP hd te Sree
he 2k : rs
6, 2 aoe

oer OeThe Cstesy . 7

magnitude could be measured directly, and from both
pieces of information the relative distances could be calcu-
lated.
Such determinations of relative distances merely gave a
plot of the Galaxy to scale (as Kepler’s laws had once
done for the Solar system) and did not give actual dis-
tances. Even this much had its value, however.
_ In the years after Miss Leavitt’s discovery, the Ameri-
can astronomer Harlow Shapley (1885- . ).made use
of the Cepheid scale to study the globular clusters. Each
cluster contained some Cepheid variables. By measuring
the periods of these, Shapley could determine their rela-
tive luminosity. By comparing this with their apparent —
brightness, Shapley could determine their relative dis- .
tances—and the relative distancés of the globular clusters~
of which they formed a part.
When he did this, he found that ce globular clusters
seemed to be distributed in a spherical arrangement. They
marked out a large ball with its center in the direction of
the constellation Sagittarius. Astronomers on Earth viewed
this ball of globular clusters from the outside—and from
far outside, too, so that the entire structure seemed to
occupy a relatively small portion of the sky in and about
Sagittarius.
Why did the globular clusters arrange themselves so? To
Shapley it seemed logical to suppose that the clusters were
grouped about the massive center of the Galaxy, as the
planets are grouped about the Sun, the massive center of
the Solar system. If so, the Galactic center is far from the
Sun, and we are at the outskirts of the Galaxy zai than
at its center.
If we only knew how far the sphere of Sapte clusters
_ actually was from us, we could then determine how far
the Galactic center might be, how far out in the outskirts
we were, and so on. If the distance of even a single
Cepheid were actually known, the distance of all the rest
_. could be calculated. The actual dimensions of the Galaxy
(and possibly of the Universe, if the Galaxy were all there
were to the Universe) could be solved.
But how can the distance of the Cepheids be deter-
mined when not one is close enough to have its parallax
measured?
To explain Shapley’s method for circumventing this
purely fortuitous bad break, I will have to engage in a
CHAPTER 5

The Size of the Galaxy

The Doppler Effect

Those of us who lived in the days when trains were

more common than they are now know that the whistle of
an approaching train is higher pitched than the whistle of
a train standing still, relative to us.1 Similarly, if a train
is receding from us, the pitch of the whistle is lower than
it would be if it were standing still, relative to us. If we
were waiting at a station and a train, sounding its whistle,
approached us, passed us without stopping, and then has-
tened away, the pitch of the whistle would drop suddenly
as it passed.
In 1842, this phenomenon was explained accurately by
an Austrian. physicist, Christian Johann Doppler Sas
1853).
To begin with, sound is a.series of compressions and
rarefactions of air. The distance from one region of com-
pression to.the next is equal to the wavelength of the
sound. The longer the wavelength, the deeper the pitch of
the sound we hear; the shorter the wavelength, the higher
the pitch.
Suppose a train whistle, sounding at a constant pitch, is
stationary with respect to you. A region of compression is

+ A train does not have to be absolutely motionless to be standing

still, relative to us. If the train were moving and we were on it,
moving with it, the train would be standing still, relative to us. As
long as it is not speeding up, slowing down, or negotiating a curve,
“but is moving at a constant velocity, however high, we can still walk
about in it as if it were standing still. And it would indeed be stand-
ing still, relative to us, even if it were not standing still, relative to
the surrounding countryside. :
ae
 74 The Universe ee
' produced and spreads outward, followed by another such
region, then by still another, and so on. The regions of
compression are separated by some fixed distance. —
If, however, the train and its whistle were moving toward
you, the second region of compression is emitted a little
closer to you than was the first. The train has moved toward
the speeding first region of compression, so that the second
region is closer to the first than it would have been if the
train were standing still. The same thing happens in the case
of the third region and the fourth. As long as the train keeps
approaching you, it gains slightly on the sound waves and
the regions of compression are consistently closer together -
than they would be if the train were standing still. The
wavelength of the sound is therefore shorter, and a higher
pitch is produced than would be if the train were standing
still. ;
Precisely the reverse takes place when the train is
receding from you. The second region of compression is
produced farther from you and from the first region; the
third region is produced still farther away, and so on. The
wavelength of the sound produced by a receding whistle is
therefore long and its pitch deeper than it would be if the
train were motionless,
The faster a train speeds toward you, the more closely
spaced are the sound waves and the higher the pitch; the
faster it speeds away from you, the more widely spaced are -
the sound waves and the lower the pitch. Knowing the
normal pitch of the whistle and the pitch one hears, it
would be possible to determine whether the train. were
approaching or receding and at what velocity, without
- requiring any other evidence.
This change of pitch with motion is called the “Doppler
effect” in honor of the physicist.

The Spectrum

In theory, the same effect should be noted in the case of

any wave form radiating outward from a source. Notably,
it would be detected in the case of light, as Doppler
himself pointed out. ‘
Light, like sound, is a wave form (although not of the
same type). Light, too, possesses wavelengths, and differ-
ences in these wavelengths can also be detected by the
senses—as a difference in color. The longest visible wave-
lengths are seen as red. As wavelengths grow shorter, the
colors change to orange, yellow, green, blue, and violet, in
 RRAfacy ee e
The Size of the Galaxy 75
that order. The colors are not differentiated sharply, of
course, but gradually shift from one into the other. We see.
_ this effect in nature in the rainbow.
Isaac Newton was the first to study a man-made rain-
bow in detail. In 1666, Newton allowed a beam of sun- |
light to enter a darkened room through a hole in a
window-blind and then pass through a triangular piece of
~ glass or “prism.” The light beam was bent, or “refracted,”
on passing through the prism and fell on the wall opposite
as a broadened spot of successive colors, similar in ap-
pearance to the rainbow. Newton called the band of
- colors a “spectrum.”
In this way, Newton showed that sunlight was not a
pure unmixed entity, for if it had been it would all have
been refracted through the prism in the same way and
would have struck the wall as an unchanged beam of
white light. Instead, it seemed actually to be a mixture of
a large variety of kinds of light, each type being refracted
through a slightly different angle and each being inter-
preted by our sense of vision as a different color. The
mixture of all the different varieties, in the proportions
_. that occur in sunlight, is sensed by us as white light.
We now recognize that the varieties of light in sunlight
are distinguished from each other by wavelength. The ~
extent to which a ray of light is refracted by glass depends
on its wavelength; the shorter the wavelength, the greater
the refraction. Orange light is refracted more than red
light, yellow light is refracted to a still greater extent, and
so on. Violet light is refracted most of all. In the final
spectrum, the red light is at the end that is least réfracted,
the end that is closest to the direction of travel of the
- original unrefracted light beam. Violet light is at the other
end)":
Within any one of the colors—orange, for example—
the longest, most nearly red, wavelengths are toward the
red end of the spectrum, while the shortest, most nearly
yellow, wavelengths are toward the violet end of the
spectrum. —~
Suppose, now, that we try to apply the Doppler effect
- to light.
The Sun, on the whole, is neither approaching the Earth
nor receding from it. Therefore, it produces what seems to
be a balanced spectrum that contains the mixture our
sense of sight interprets as white. However, if the Sun
were approaching us, might we not suppose that the
wavelengths of light would be squeezed together and that
 + 716 The Universe
every wavelength of light reaching us would be shorter
than it normally would be. There would be a shift of the
entire spectrum toward the short-wavelength end.- Every
bit of the red band would shift toward the orange, every
bit of the orange would shift correspondingly toward the
- yellow, and so on. Because the entire shift is toward the
violet end of the spectrum, it is referred to as a “violet
shift.” 3
Under such conditions, one. might expect the mixture of
light in a spectrum no longer to produce a clear white.
-There would be a deficiency at the red end, an excess at
the violet end, and the color of the Sun (if it were
approaching us) might be expected to take on a bluish
tinge. The more rapidly it approached us, the bluer its
light would become by this reasoning.
The same argument, in reverse, could be used to fore-
cast what would happen if the Sun were receding from us.
_ This time the crests of successive light waves would be
- pulled apart. Wavelengths would become longer than nor-
mal, and the entire spectrum would be shifted toward the
red end of the spectrum—the “red shift.” If the Sun were
receding from us, we might reason, its light would take on
an orange tinge: The more rapidly it receded from us, the
more deeply orange its light.
And yet, although this reasoning seems airtight, it is
betrayed by the facts. The trouble is that the light we see
in a spectrum is not all there isto the spectrum. :
In 1800, William Herschel studied the spectrum pro-
duced by sunlight (the “Solar spectrum”). He noted the
heating effect produced on a thermometer exposed to
_ different portions of the spectrum, measuring in this way
the total energy content of each portion. It would have
been natural to expect the temperature rise to become less
marked and then vanish completely as one approached the
end of the spectrum. This was not so as far as the red end
was concerned. Indeed, the temperature rise was greater
some distance beyond the red end of the spectrum than
anywhere within it.
Herschel suggested that sunlight included wavelengths
of light that were longer than any that could be sensed by
our eye. Such wavelengths would be refracted still less
than those of red light and would be located beyond the
red end of, the spectrum. This would-be “infrared radia-
tion” (“below the red”).
Such radiation would be real and would differ from
ordinary light only in the size of its wavelength and in the
 ea EN Bet ee ah RWee eI CNS ON te oe ha ge
5 Beyer Re SR Ea RD eT a
. ‘The Size of the Giles eRe
fact that the human eye could not sense it. Thus, by 1800,
one could no longer speak merely of “light” to define the
entity by which we see; one had to speak of “visible light”
instead. Invisible light was no longer a contradiction in
‘terms, but an actual fact (although today infrared light is
easily detected by appropriate instruments even though it
remains invisible to the eye).
Nor was the violet end of the spectrum a true end. In
1801, the German physicist Johann Wilhelm Ritter (1776-
1810) studied the ability of light to bring about certain
chemical reactions. Light could, for instance, bring about
a breakdown of the white chemical, silver chloride, liber-
ating tiny particles of metallic silver. Small particles of
metal usually appear black, and the silver chloride was
therefore increasingly blackened as it was exposed to
light. Ritter found that different portions of the Solar
spectrum were not equally effective in bringing about this
change. The shorter the wavelength the more quickly the
silver chloride was blackened. Inspired by Herschel’s’ dis-
covery the year before, Ritter tested the region beyond
the violet end of the spectrum, where no light at all could
be seen. Sure enough, the silver chloride was darkened
more rapidly there os it was anywhere within the visible
spectrum.
~ The conclusion was that there were iavelan otis of
sunlight shorter than any that could be sensed by the eye.
Such wavelengths would be refracted still more than those
of violet light and would be located beyond the violet end
of the spectrum. This would be “ultraviolet radiation”
(“beyond the violet”).
We must therefore visualize the spectra produced by the
Sun (and by other stars) as consisting not only of the
red-to-violet colors we can see, but of invisible regions of
light beyond the red and violet. If a star is approaching
us, so that the wavelengths of its light grow shorter and its
spectrum undergoes a violet shift; then light does not “pile

Beyond the visible spectrum

 78 The Universe
' up” at the violet end. It “spills out” the end instead, mov-
ing into the ultraviolet region and invisibility. Nor does the
shift leave a vacant region at the red end. Infrared radiation
turns visible red with the wavelength-shortening and moves
in to fill the gap.
The spectrum as a whole shifts, but the visible ported
remains unchanged; what it loses at one end if gains at the
other. The same argument holds if the star were receding
from us to produce a general red shift.
To be sure, a star may be approaching us so rapidly
that the entire infrared region is shifted far into the visible
or receding from us so rapidly that the entire ultraviolet
region is shifted far into the visible. In that way, there
would indeed be a visible shift in color toward the violet
or toward the red. However, the velocities of approach or
recession required to produce such overall shifts in color
are so high that it seemed inconceivable (in the nineteenth
century, at least) that such speeds would be encountered
among the various heavenly objects.
And yet, as it happens, there are indeed differences in
colors among the stars. Some stars, Antares for instance,
are distinctly reddish in color; others, like Vega, are dis-
tinctly bluish. Can this possibly signify that Antares is
receding very rapidly from us or that Vega is very rapidly
approaching us.
Unfortunately not. There are other causes that might
account for differences in color.
Thus, the Sun reddens at the time of sunset although it
is not then receding from us. What-is happening then is
that its light is passing through an unusually great thick-
ness of atmosphere, and the molecules and dust particles
in the atmosphere have a greater opportunity to reflect
and scatter the sunlight. The short wavelengths of light are
more efficiently scattered than are the long wavelengths.
Thus, the daylight sky is blue with scattered sunlight from
the short-wavelength end of the spectrum. The amount
scattered in this fashion when the Sun is high is not enough
to affect the color of the sunlight significantly. However, as
the Sun approaches, the horizon so much of this
short-wavelength region is scattered and removed from sun-
light by the greater thickness of dusty atmosphere, which
the light must pass through, that what is left is distinctly
red i in color.?

® The setting Sun is redder than the rising one, because at the end
of the day, the air is generally dustier than it is at the beginning.
 Beers oat: eee aro ae OS ee es

The Size of the Galaxy 19

Again, a glowing substance will change its color if its
temperature is altered. An iron ball, if gradually heated,
will eventually become hot enough to glow a deep red in
color. If the temperature continues to rise, it will glow a
brighter red, then orange, then whitish, then blue-white.
The higher the temperature, the greater the overall shift
of its light in the direction of the shorter wavelength;
. however, such a shift would not imply any. motion of the
ball toward us or away from us.
Finally, color may vary with chemical composition. If
an object shines by reflected light, this is obvious. Differ-
ent dyes and different pigments will reflect lights of differ-
ent colors. Even when a substance is itself the source of
light and not merely a passive reflector, color can depend
on chemistry. If ordinary table salt (a compound of the
metal, sodium) is heated until it glows, it will produce
light of a distinctly yellow tinge. Compounds of strontium,
when heated, will produce a red light, those of barium a
green light, those of potassiumaviolet light, and so on.
Does it follow from this that there is no way of telling
from starlight anything about the motion ofa star, despite
the Doppler effect? Not at all. In fact, the situation is even
better than you would expect; not only motion but many
other properties of stars can be determined by examining
_ the spectra they produce. A spectrum contains more than
a rainbow of colors—much more.

Spectral Lines

In 1814, the German optician Joseph von Fraunhofer

(1787-1826) revolutionized the study of spectra. He was
a manufacturer of prisms of fine glass and he tested their
quality by their ability to-form spectra. Following the
work of the English chemist William Hyde Wollaston -
(1766-1828), Fraunhofer passed a beam of sunlight
through a fine slit before passing it through the prism.
_ Each wavelength of light was refracted through a charac-
teristic angle and produced an image of the slit in its own
- particular color at a particular position on the screen on
_ which the light was thrown. The various images over-
lapped to produce a nearly continuous rainbowlike spec-
trum.
The spectrum was not completely continuous because
some wavelengths were missing in sunlight and these miss-
ing wavelengths showed up as dark lines in the spectrum;
slit-images that were not there, so to speak. Such dark
 ne Srey fe a eR SO ROR Pg Lh ee
80 ~The Universe a Ss
lines in the Solar spectrum had been observed by Wollas-
ton, but Fraunhofer’s excellent prisms made them much
clearer and made many more of them visible—hundreds
of them. Fraunhofer was the first to study the dark lines
in detail and to map their exact position in the spectrum.
They are therefore called “Fraunhofer lines” or, more
generally, “spectral lines.”
The spectral-line pattern of the Solar spectrum is quite
distinctive. Other self-luminous objects, such as the various
stars, also show dark lines in their spectra, but in patterns
that often differ markedly from that in the Solar spec-
trum. Nevertheless, certain spectral lines, particularly the
most prominent, which Fraunhofer had denoted by letters
of the alphabet from A to K, appeared like landmarks in
the spectra of most stars.
It is possible to measure the wavelength corresponding
to any spectral line by measuring the angle of refraction
associated with it; that.can be done by accurately measur-
ing its position on the screen against a reference scale. If,
for any reason, the wavelength corresponding to the spec-
tral line is shortened, the angle of refraction associated
with it is made greater ‘and the position of the line moves
toward the violet end of the spectrum. If the wavelength
is made longer, the line moves toward the red end of the
‘spectrum.
Soon after Doppler had explained the changing pitch of
sound with motion, the French physicist Armand Hippo-
_lyte Louis Fizeau (1819-1896) pointed out that to de-
tect the effect in light, one should not worry about overall
color, but should measure the exact position of the spec-
tral lines and note their shift.
To see why this should be so, imagine a long featureless
rod of which you can only see a small portion. If this rod
is shifted slightly, we see a somewhat different portion, but
it is still featureless; we cannot tell how much the rod has
shifted or in which direction, or, indeed, whether it has
moved at all. If, however, the part of the rod we saw
contained a marking, then we could easily detect any
movement of the rod by noting the shift in position of the
marking.
It is in this fashion that spectral lines make it possible to
detect any Doppler shift in light. They are the marking on
the spectrum. And it is because Fizeau pointed. this out
that we sometimes speak of the “Doppler-Fizeau effect” in
connection with light.
The effect. is small, however, and hard to determine.
EPCs emanate ee A BN ey Rat es
Pee The Size of the Galaxy 81
Sound travels relatively slowly—about 331 meters per
second—and a ‘train can easily move at one-tenth this
speed. Sound waves can then-be pushed considerably
closer together or pulled’ considerably farther apart. Light,
however, travels at 300,000,000 meters per second, or -
nearly a million times the speed of sound. Most stars
‘move (relative to ourselves) at less than one ten-
thousandth of this speed. The wavelengths of light from
such stars are only very slightly altered by their mo-
tions.
It was not until 1868 that the British astronomer
William Huggins (1824-1910) was able to detect a small
shift in the spectral lines of the bright star, Sirius, and to
show that it was moving away from the Sun.
It is important in this connection to realize that stellar
motions must be understood three-dimensionally. A star
might be moving in a direction that is exactly at right
angles to our sight or exactly in our line of sight.
' Either case isvery. unlikely. It is much more likely that
a star moves neither in the line of sight nor at right angles
to it, but somewhere in between. When this is the case, the
motion can be split into two components, one in the line
_ of sight and one at right angles to the line of sight. This is
done by representing the actual motion as the diagonal of
a rectangle. The two adjoining sides of the rectangle then
represent the components. The lengths of the diagonal and
two sides are proportional to the actual-motion and the
two component motions, respectively.
‘The component in the line of sight is called the “radial
velocity” because that is the motion toward us or away

Component motions

speed
transverse

@
observer
 . 82 . - ‘The Universe
_ from us along an imaginary radius that connects our eye
to the vast sphere of the sky. The component at right
angles to the line of sight is the “transverse velocity” (the
velocity “across” the line of sight). It is this transverse
velocity that shifts the star bodily across the sky and
_makes itself evident as the proper motion of the star.
The relative size of the two components depends on the
angle the star’s motion makes to the line of sight. Con-
versely, if the size of the radial velocity and the transverse
-velocity are both known, then the actual velocity of the star
can easily be deduced.
The two component velocities are determined in two
entirely different ways.
The radial velocity is reflected in the star’s spectrum by
_ means of the Doppler-Fizeau effect. By measuring the
extent of the shift of the spectral lines, the. radial velocity
_ (toward us if the shift is toward the violet, away from us
if it is toward the red) can be determined directly as
kilometers per second. This determination does not depend
on the distance of the star. No matter how far the star.
may be from us, the wavelengths of the spectral lines have
_ set values and a given shift in those values indicates a
specific radial velocity whether the star is 4 light-years
away or 4000. The only requirement is that the star be
bright enough to yield a spectrum in which the position of
_ the lines can be measured.
A star’s transverse velocity is not reflected in the spec-
trum at all, but only as a bodily shift across the vault of
‘the sky. It is therefore detected as proper motion and
measured as angles of arc. In order to change a proper
- motion (in seconds of arc) into transverse velocities (in
_ kilometers per second) one must know the distance of the
star. For instance, if Barnard’s star moves across the sky
at a rate of 10.3 seconds of arc’ per year and-is 6.1.
light-years away, we can calculate that it has a transverse
velocity of some 90 kilometers per second (or 56 miles
per second).

The Galactic Center

Now it is time to return to the problem of determining

the distance of the Cepheid variables.
Suppose the matter is considered statistically. In the
case of some stars, moving in directions close to the line of
sight, the radial velocity is greater than the transverse
SE ee ee eye ot

The Size of the Galaxy . 83

“velocity. In the case of other stars, moving close to right
angles to the line of sight, the transverse velocity is greater
than the radial velocity. On the average, however, these
two opposing tendencies tend to cancel, and the radial and
transverse velocities may be taken, in general, as equal.
If this is so, then the measured proper motion of a
Cepheid (in angles of arc per year) may be taken as
equivalent not only to the transverse velocity but to the
radial velocity as well, and the radial velocity can be
measured directly, by spectral means, as kilometers per
second. If you know a motion both as a particular angle
of arc per year and as a particular number of kilometers
per second, you can calculate the distance; for at one
distance, and one distance only, a velocity of so many
kilometers per second will produce a shift of a particular
angle of arc per year.
For any given Cepheid, a calculation of this sort may
give a wildly wrong answer for the distance, for in its
particular case, the radial velocity may be much larger or
much smaller than the transverse velocity. If a number of
Cepheids in a given cluster, all of the same period, are
taken, however, and the distances determined for each,
then the average distance is quite likely to be close to the
truth.
The Danish astronomer Ejnar Hertzsprung (1873-
) determined the distances to certain Cepheids in this
manner in 1913, but several years later, Shapley applied
this statistical method to the specific problem of determin-
ing the structure of the Galaxy. He determined the dis-
tances to Cepheids of varying periods. From these dis-
tances and the apparent brightness of the Cepheids, he-
could calculate the luminosities of the Cepheids.
If the luminosities, so determined, were plotted against
the periods, then one should get, at least roughly, a
smooth line if one allowed for a certain scattering, owing
to the intrinsic uncertainty of the method. At least, one
should get this if there were any validity at all to the
method. On the other hand, if this statistical method were
of no value at all, then the distances determined would be
all wrong, the luminosities in consequence would also be
all wrong, and there would be no reasonable correlation
between luminosity and period.
When Shapley plotted luminosities against period, how-
ever, he obtained a very good smooth line, and his results
could therefore be accepted as essentially correct. Since he
knew both the luminosity and period of representative
 ae Sy" as Ty
; 2 Re 7am ee Seo URED ae ee eT
84 The: ‘Universe ;
Cepheids, he could determine the distance ae any Cepheid.
The Cepheid yardstick had become absolute.
In this way, Shapley could determine the. actual dis-
tances of the various globular clusters and then go on to
calculate the distance of the center of the sphere over
.which they were distributed. The center of this sphere he
‘assumed to be the center of the Galaxy, and according to
his figures it was 50,000 light-years (15, 500parsecs) from a
the Sun.
By 1920, then, the position of man in fis’Unione had
again been altered, drastically, and once again in the
direction of increased humility. Copernicus. had shown
‘that the Earth was not the center of the Universe, but he
had been certain that the Sun was, as part of the order of
nature. Even Herschel and Kapteyn had considered the
Sun at the center of the Galaxy (and therefore of the
Universe), at least through accident if not through natural
order. Now Shapley showed, quite convincingly, that this
» was not so, that the Sun was far on the outskirts of the
Galaxy.
Mn place
cl of Ptolemy’s geocentric Universe and Coperni-
cus’ heliocentric Universe, we now had Shapley’s “eccen- .
tric Universe”; one in which the Sun was ,,away from the
center’—the literal meaning of “eccentric.”
Shapley’s eccentric Universe raised some ‘problems,
however. If the bulk of the Galaxy was to one side of the
Sun, off in the direction of the constellation Sagittarius,
why was not the band of the Milky Way vastly (instead of
only moderately) brighter in that direction than in the
_ Opposite direction, where only the tag end of the Galaxy
existed?
The answer to that question came with the realization
that there was more to the Universe than met the eye—
quite literally.
Not everything that glitters in the heavens beyond the’
_ Solar system is a star. There were also more diffusely
_ bright objects, some of which had been carefully mapped
by Messier, and some of which had been observed even
before Messier.
In 1694, for instance, the Dutch astronomer Christian
Huygens (1629-1695) entered-inhis diary. the - descrip-
tion of a bright, fuzzy region in the constellation of Orion.
Such a bright, fuzzy region, resembling ‘a luminous cloud,
came to be called a “nebula” (the Latin word for
“cloud”). The particular one described by Huygens is the
“Orion Nebula.”
Nee ghee. ex Pe eh Go A eee
oar
E

“The Size of the Galaxy 85

The Orion Nebula is a huge object. It is now known to
be about 1600 light-years away, and in order for it to
show the diameter we observe it to have, it must be 30
light-years in diameter. It appears to be a vast cloud of
dust particles, reflecting and scattering the light of a
number of hot stars set within it. Many other luminous
nebulae of this ae some exceedingly beautiful, are now-
known. ;
Astronomers becetis aware, however, that just as there
were regions of space that glowed with a soft, cloudy
luminosity, so there were other regions in which there was
a surprising lack of luminosity. Thus, William Herschel, in .
_ studying the Milky Way closely, noted regions where there
were very few stars, although they might be bounded by
other, regions that were simply bursting with vast numbers
of stars. Herschel took these dark areas at face value,
assuming that they represented regions that did not con-
tain stars and that the Earth was so situated that men
could see into the empty region. “Surely,” said Herschel,
“this is a hole in the heavens.”
As more and more of these regions were studied, it
began to seem more and more improbable that such
regions, unexplainably empty of stars, could exist in such
numbers and that all should just happen to be so situated _
that we could look into the “hole.” By 1919, E. E.-
Barnard had listed the positions of 182 such dark regions,
and by now the number of those observed has-increased to
-over 350.
It was borne in on Barnard and on the German astron-
omer Max Franz Joseph Cornelius Wolf (1863-1932)
that these dark regions were not merely holes. They did
not indicate the absence of matter but rather the presence ©
of matter, vast clouds of dust particles that absorbed and
blocked off the light of the stars that lay behind them,
much as the clouds in the Earth’s atmosphere absorbed
and‘blocked off the light of the Sun behind them.
In short, there were “dark nebulae” as well as bright
ones. The bright ones shone only because they contained
stars; the dark nebulae were dark because they did not
contain stars.
Famous dark nebulae include the “Horsehead Nebula”
in the constellation Orion, which stands out like a dark
horse’s head against the luminosity of a bright nebula all
about. (Actually, the shape of the.-Horsehead Nebula
seems to me to be more like the head and shoulders of
 fet Fame er SebaR et oa Se
86 The Universe re :
Walt Disney's “Big Bad Wolf.”) There is algo the “Coal
Sack,” a region of. intense darkness- near the Southern
Cross. :

If fie dark hebrilae are more or less evenly distributed

throughout the Galaxy, they would be most- plentiful ex--
actly where stars are most plentiful. And, indeed, most ef
the dark nebulae are located in the plane of the Milky
._ Way and, particularly, in the direction of Sagittarius,
where the Galactic center and the bulk of the Galaxy’s
_ structure are to be found. It is easy to see, then, why the
Galactic center and all that lies beyond should be ob-
-. scured, Their light is permanently blocked off by clouds of
dust.
Visible portion of the Galaxy

!
1
\
1
1
|

A annnnnn RRR EEE eee

visible portion j invisible portion
of Galaxy ;

: The band of the Milky Way seems roughly equal in

_brightness in all directions, not because the Solar system
happens to be near the center of. the Galaxy, but only
because most of the light’ of the Galaxy is obscured by
_ dark nebulae. What we see of the Milky Way is only a
‘portion of our own neighborhood, so to speak—only our
own end of the Galaxy.
Thus, Shapley’s conclusion that we were eccentrically
placed in the Galaxy held fast even though the nature of
the visible Milky Way made us seem to occupy a central
position.

Galactic Dimensions

Nevertheless, Shapley’s conclusion was not correct in

detail. For the first time, an astronomer had overestimated
the size of a’significant portion of the Universe, after
ee

“The Size of the Galaxy 87

thousands of years in which the size of the Universe had
been consistently underestimated.
Again, the difficulty arose in connection with what one .
could not see. Once we grant the existence of dark nebu-
lae it becomes reasonable to wonder whether any part of. -
space is perfectly transparent.
We can draw an analogy here with our atmosphere. We
all know that unusual concentrations of dust particles
(whether solid as in smokes, or liquid as in fogs, mists, or
clouds) can block off light of even something as bright as —
the Sun. If the dust particles are low-lying (notably as
fogs), vision is cut off at moderate distances; in extreme
cases, we cannot see more than a few feet, so efficiently
do the particles of the fog or smoke scatter light.
It would. certainly seem, in comparison, that when the
air is sparklingly clear and free of any visible trace of
cloud or mist, it is then perfectly transparent and in no
way interferes with light.
This, however, is an illusion. Even at its clearest, the
atmosphere is not perfectly transparent. It will always
scatter light, as is evidenced by the mere existence of a
blue sky—the blue being built up of scattered sunlight.
And even the clearest night air absorbs a significant frac-.
_ tion of the starlight falling on the Earth. .
We can, in short, expect perfect transparency of noth- |
ing short of perfect vacuum. Interstellar space (that is, the
space between the stars) is nearly a perfect vacuum—but
not quite. It is far more transparent, even in the midst of
a nebula, than is the Earth’s atmosphere, but it is still, not
perfectly transparent., An occasional particle of dust
floating here or there in the enormous emptiness of inter-
stellar space will intercept and deflect a ray of starlight.
Taken singly, this amounts to nothing, but in the stretch of
light-years between one star and the next, enough dust
particles may intervene to result in a cumulative scattering
large enough to detect.
This minute scattering effect can best be detected by the —
fact that short-wave light is more easily scattered than is
long-wave light. Such scattering subtracts the blue-violet
end of the spectrum and what light remains is increasingly
reddened. If, then, the distant stars can be shown to be
dimmer and. redder than they ought to be, and if this
effect increases steadily with distance, then the presence of
interstellar dust is strongly indicated. -
The first to show definitely that this effect existed was
 : The Universe
- the Swiss-American astronomer Robert Julius Trumpler
~ (1886-1956). He did this in connection: with his studies
of star-clusters. Such clusters have an average size and an
average brightness, and both should decrease in the same
_ manner—in proportion to the square of the distance. It
followed then that, if a globular cluster takes up a particu-
_ Jar area of the sky, it ane ought to have a particular bright-
ReSSs 2
In 1930, Trumpler showed, however, that the light of
the more distant globular clusters was dimmer than was to
be expected from their sizes. The more distant the cluster,
the more marked this departure from the expected bright-
ness. The area of the clusters was decreasing in accord-
~ ance with the square of the distance, but the brightness of
the clusters seemed to be decreasing in ‘accordance with
__ the square of the distance plus some additional dimming
_ effect. And the more’ distant the cluster, the redder it
_ seemed.
The easiest way of explaining this was to suppose that
incredibly thin wisps of dust in interstellar space had a
distinctly dimming and reddening effect over vast dis-
tances. Over those vast distances, however, the dust suc-
ceeded in dimming and reddening the farther clusters, just
as the dust in the atmosphere of the Earth dimmed and
_ reddened the setting Sun. In the plane of the Milky Way
_ where dust concentrations are highest, it is estimated that
half the energy of a light ray is scattered after a journey |
of 2000 light-years,. half of what is left is scattered in
another 2000 light-years,_and’so on. After 30,000 light-
years (the distance of the Galactic center from us) only
1/32,000 of the energy of light rays would be left, even if
_ they did not pass through any of the usual dust
concentrations represented by the dark nebulae. It is no
wonder that we cannot see the center of the Galaxy.
This dimming effect is important in connection with
distant Cepheids. From the period of a particular
Cepheid, its luminosity can be determined. If this luminos-
ity is compared with its observed brightness, its distance
_ can be determined, since it is easy to calculate how far off ©
a star must be in order for its luminosity to be reduced to
the more pinpoint of light that is actually observed. This
calculation assumes, however, that the reduction of bright-
ness is caused entirely by the distance factor. The presence
of interstellar’ dust would perceptibly dim the light of a
distant Cepheid by an additional amount, however. If it
were not for the presence of dust, such a Cepheid would
 ee erEAS
ee 2° ? rele ) a x

"TheSize of the Galaxy 89

appear distinctly brighter ‘and would then be judged —
closer. In other words, the existence of interstellar dust (if
not allowed for) tends to produce a falsely large estimate
of great distance.
Shapley’s estimate that the center of the Galaxy was
50,000 light-years away was based on the assumption that
the Cepheids lost brightness only through distance. Allow-
ing for the presence of interstellar dust as a second dim-
ming agency, the center of the Galaxy need be less than
30,000 light years from us.
By the early 1930’s, then, the dimensions of the Milky
Way, as we now accept them, were finally determined.
The Galaxy is a lens-shaped object about 80,000 to 100,-
~ 000 light-years across, with our own Solar system about
27,000 light-years from the center.
Furthermore, from the Cepheids observed in the Magel-
lanic Clouds, it was possible to determine their distances.
The Large Magellanic Cloud is about 155,000 light-years
from us and the Small Magellanic Cloud about 165,000
light-years away. Their position is shown in the figure on
‘page 90, and there is reason to think that there are tenuous ~
connections between the two Clouds so that they form a
single system.
The size of the Galaxy, is not merely a matter of the
space it takes up.. How many stars does it contain?
To answer that question, Jan Oort assumed that the
Milky Way was strongly concentrated toward its center.
The central region might contain 90 percent of ail the .
stars in the Galaxy. If this were so, then the stars in the
outskirts (our own Sun, for instance) would revolve about
the center as, within the Solar system, the Earth revolves
about the Sun.
Furthermore, stars closer to the Galactic center than
the Sun is would revolve about that center at a greater
velocity than the Sun: did, while stars farther from the
center would move at a lesser velocity. (This follows from
gravitational theory and is analogous to the manner in
which Mercury, for instance, which is closer to the Sun
than the Earth is, moves about: the Sun at.a greater
velocity than the Earth does, while Jupiter, farther from
the Sun than the Earth is, moves at a lesser velocity.)
This arrangement holds true only if the various bodies
moving about the center have orbits that are circular or
nearly so, as is the case for the planets in our Solar
system. If some stars have very eccentric orbits about the

a .
 Sn Oa' hie Py 1 te !
oe cr ee ; *. Ter? Se ee
- ‘ :
90 Be ‘The Unters +a hata ee
Galactic center, their motions srould not be so eaaily
analyzed.
‘If, however, a large number of Aves is taken into
consideration, the effects of orbital eccentricity will aver-
age out. In that case, it can be seen that the stars that lie
between us and the Galactic center will tend to move
faster than we do and will catch up to our position,
moving slightly. toward us, radially, as they do so. Once
they pass us, they continue to move farther and farther
ahead of us, moving slightly away from us, radially, as
they do so. On the other hand, stars farther from the
center than we are move more slowly. We catch up to
them and move closer to them radially, then pass them
and move away from them.

The Magellanic Clouds and the Milky Way

On the whole, then, there should be a certain regularity

both in transverse velocities and radial velocities among
the stars fairly close to us. And there is. It is the regularity
in radial velocities that made it possible to detect the
apparent motion of the Sun toward the apex (see page
64), a minor motion, relative to the closer stars only. In
1904, Kapteyn had drawn more general conclusions, de-
ciding that there were two streams of stars, one moving in
a particular direction, the other in a directly opposite
direction,
In 1925, Oort showed that Kapteyn’s streams consisted
 The Size of the Galaxy 91
-of the inner stars catching up to the Sun, and the other
stars lagging behind the Sun. He was able to determine the
. nature of the general rotation of the Galaxy and from
that calculate the direction and distance of the Galactic
center by a method independent of the position of the
globular clusters. He showed that the center was some .
30,000 light-years distant, in the direction of Sagittarius.
This agreed with the evidence of the globular clusters once
the presence of interstellar dust was taken into account,
and such agreement sufficed to bring Shapley’s eccentric
model of the Galaxy into general acceptance by astrono-
mers.

Galactic rotation

E
-———- >
6

Hr
ben a ao ae DD

een = s! — im
ori G
Cr canta
Bae he : 0

Beer 10 a:
ere >»

paces
I
!
1
Furthermore, Oort could show, from his study of the
relative motions of the stars, that the Sun was moving in a
fairly circular orbit about the Galactic center at a veloci-
ty; relative to that center, of 220 kilometers per second
(or about 140 miles per second), completing one TevOnE-
tion about the center in 230,000,000 years.
For gravitational attraction to drive the Sun eboue its

E
 2 = Fs

S20 “The Universe

orbit at that distance and that velocity, the Galactic center
must have a mass about ninety billion times that: of the
Sun. If we assume that 90 percent of the mass of the
Galaxy is concentrated in its center, then the entire mass
of the Galaxy is equal to about 100 billion times that of
the Sun. 3?
Statistical studies of the stars seem to indicate that the
Sun is an average star in mass, so that we might suspect
_ the Galaxy to contain about 100,000,000,000 stars, al-
though some estimates place the figure higher.
Once allowance is made for their immense distance, the
Magellanic Clouds turn out to spread over a region of
respectable size. The Large Magellanic Cloud may be
some 40,000 light-years in extreme diameter, or nearly
half the diameter of the Galaxy. However, the Magellanic
Clouds are much less densely populated with stars than the
Galaxy .is, and that is perhaps the more important fact.
The Large Magellanic Cloud contains no more than five
to ten billion stars, perhaps, and the Small Magellanic
Cloud only one to two billion stars. In terms of star
content, the two, taken together, are only one-tenth as
large as the Galaxy. They might also be pictured as
“satellites” of the Galaxy. -
Nevertheless, despite the smaller size of the Magellanic
Clouds, they contain some kinds of objects larger and
more spectacular than any found in the Galaxy—at least
in those parts of the Galaxy we can see. For instance, the
most luminous known star, S Doradus, is in the Large
Magellanic Cloud. S Doradus is not quite bright enough to
see with the unaided eye, being only an eighth-magnitude
star. To be even so bright, however, at the distance of the
Large Magellanic Cloud, is monumental. S Doradus is
thirty times as luminous as Rigel, the most luminous of
those stars comparatively near to us and is about 600,000
times as luminous as the Sun.
The Large Magellanic Cloud also contains the Taran-
tula Nebula, a bright cloud of dust like the Orion Nebula,
but 5,000 times larger. The Tarantula Nebula is far larger
and more spectacular than any similar object that can be
observed in the Galaxy.
We have, then, a clear picture of that portion of the
Universe containing our own gigantic cluster of stars. If
we imagine a sphere with its center at the center of the
Galaxy and a radius of about 200,000 light-years, it would
 laxy id‘the Magellanic Clouds,
Including.a total ofeee as much as 150,000, He: 000 ©
‘Stars. 3

ao

> *
'

>

‘
: ‘
ay,
r : rae
We f ‘ ce Pi
bar’ 4 I+ AYYS a
 Ot ES a See

CHAPTER ¢

- Other Galaxies

The Andromeda Nebula

Man’s vision of the size of the Universe had increased

- enormously in 2000 years. Let us recapitulate.
By 150 B.c., the Earth-Moon system had been accurate-
ly defined. The Moon’s orbit was seen to be half a million
miles across, and the diameter of the planetary orbits. was
suspected to be in the millions of miles.
By 1800 a.D., the scale of the Solar system had been
'. defined. Its diameter was not merely in the millions of
miles, but in the billions. The distance of the stars was still
unknown but was suspected to be in the trillions of miles
(that is, a couple of light-years) at least.
By 1850 a.D., the distance of the nearer stars had been
defined as not merely trillions of miles, but tens and
- hundreds of trillions of miles. The diameter of the Galaxy
was still unknown but was suspected to be in the thousands
of light-years.
/ By 1920 a.D., the diameter of the Galaxy had been
defined at not merely thousands of light-years but many
tens of thousands of light-years.
At each new stage, the size of the regions of the
Universe under investigation turned out to exceed the
most optimistic estimates of the past. Furthermore, at
each stage, there was the conservative opinion that the
object whose size had been. defined represented all, or
almost all the Universe, and until 1920, that view had
always turned out to be wrong. f
‘The Earth-Moon system had shrunk to insignificance in
the light of the size of the Solar system: The Solar system
had in turn shrunk to insignificance when the distance of
the nearby stars was determined. And the system of the
 "Other Galaxies es 95
nearby stars was insignificant in comparison with the
Galaxy as a whole.
Would this process continue or did the Galaxy and its —
_ Magellanic satellites represent an end at last? Had astron-
omers finally probed to the end of the Universe?
Even as late as 1920, it seemed quite possible that the
conservative view would finally triumph. The Galaxy and
the Magellanic Clouds seemed very likely to contain all
the matter in the Universe and beyond them, one could
maintain, lay nothing.
This time, there were strong theoretical arguments to
back the conservative view. Remember that Olbers’ para-
dox seemed to imply the existence of a finite Universe
(see page 56) and the fact that the stars seemed confined
to a finite lens-shaped Galaxy bore this out. If there
proved to be numerous enormous objects beyond the
Galaxy and its satellites, then Olbers’ paradox might well
present astronomers with an insoluble dilemma.
_. And yet astronomers could not relax completely with
their finite Universe 200,000 light-years across. There
were grounds for some suspicion that numerous large
- objects might exist far outside the Galaxy, and it proved
extraordinarily difficult to argue that suspicion out of
existence,
A particularly troublesome item was a cloudy patch of
light in the constellation Andromeda, an object that was
called the “Andromeda Nebula” because of its location
and appearance.
The Andromeda Nebula is visible to the naked eye as a
small object of the fourth magnitude that looks like a
faint, fuzzy star to the unaided eye. Some Arab astrono-
mers had noted it in their star maps, but the first to
describe it in modern times was the German astronomer
. Simon Marius (1570-1624) in 1612. In the next centu-
ry, Messier included it in his list of fuzzy objects that were
not comets. It was thirty-first on his list, so that the
Andromeda Nebula is often known as “M31.”
There was no reason, at first, for thinking that the
Andromeda Nebula was significantly different from other
_nebulae such as the Orion Nebula. The Andromeda Nebu- ©
la seemed a luminous cloud and no more than that.
Some eighteenth-century astronomers even envisaged a
place for such clouds in the scheme of things. What if
_ Stars developed out of distended rotating masses of gas?
Under the effect of their own gravity, such clouds would
begin to contract and condense, speeding their rotation as
 96 ~The Universe
they did so. As they rotated more and more quickly, they
would flatten into a lens shape and, eventually, eject a ring
_ of gas from the bulging equator. Later, as rotation contin-
ued to speed up, a second ring would separate, then a
third ring, and so on. Each ring would coalesce into a
small planetary body, and finally what was left of the
cloud would have condensed into a large glowing star that
would find itself at the center of a whole family of
planets.
Such a “theory would account for the fact that all the
planets of the Solar system were situated nearly in a single
plane, that all of them revolved about the Sun in the same
direction. Each planet, moreover, tended to have a system
of satellites that revolved about it in a single plane and in
the same direction, as though the planets, in the process of
contracting from Besegus rings, gave off smaller rings of
their own.
The first to suggest such an origin of the Solar system
was the German philosopher Immanuel Kant (1724
1804) in 1755. A half-century later, the French astrono-
mer Pierre Simon de Laplace (1749-1827) published a
similar theory (which he arrived at independently) as an
appendix to a popular book on astronomy.
It is interesting that Kant and Laplace had opposing
views on the Andromeda Nebula, views that kept astrono-
mers at loggerheads for a century and a half.
_ Laplace pointed to the Andromeda Nebula as possibly
representing a planetary system in the process of forma-
tion; indeed its structure is such that it seems to be in the
’ obvious process of rapid rotation. You can almost make
out (or convince yourself you are making out) a ring of
gas about to be given off. For this reason, Laplace’s
suggestion as to the method of formation of planetary
systems is known as the “nebular hypothesis.”
If Laplace were correct and if the Andromeda Nebula
were a volume of gas serving as precursor for a single
planetary system, it cannot be a very large object and, in
view of its apparent size in the telescope, it cannot be a
very distant object.
Laplace’s nebular hypothesis was popular among astron-
omers throughout the nineteenth century, and his view of
the Andromeda Nebula represented a majority opinion
through all that time. In 1907 a- parallax determination
was reported for the Andromeda Nebula, one that seemed
to show it to be at a distance of 19 light-years. come
that seemed to settle matters.
 Other Galaxies ~~ 97
Yet there was Kant’s opposing view. Despite the fact
that he, too, had originated a nebular hypothesis, he did
not fall prey to the temptation of accepting the Androme-
da Nebula as visible support of his theory. He suggested
instead that the Andromeda Nebula, and similar bodies,
might represent immensely large conglomerations of stars,
which appeared as small, fuzzy patches only because they
were immensely far away. He felt they might represent
“island universes,” each one a separate galaxy, so to speak.
However, this suggestion of Kant’s was not based on
any Observational data available to the astronomers of the —
time. It made very few converts and, if Kant’s speculation
was thought of at all, it was dismissed as a kind of science
fiction.
But Kant’s suggestion did not die. Every once in a
while some small piece of evidence would arise that would
not quite fit the orthodox Laplacian view. Chief among
these was the matter of spectroscopic data.
Stars, generally, produce light which, on passing through
a prism, broadens into an essentially continuous spectrum,
. broken by the presence of dark spectral lines. If, however,
gasses or vapors of relatively simple chemical composition
_ are heated until they glow, the light they emit, when passed
through a prism, produces. an “emission spectrum” consist-
ing of individual bright lines. (The exact position of .the
bright lines depends on the chemical composition of the gas
or vapor.)
Then, too, a continuous spectrum usually (but not al-
ways) implies white light, while an emission spectrum is
often the product of colored light, since some of the
“bright lines of one particular color or another might
dominate the entire glow.
Many bright nebulae do indeed show very delicate color _
effects (that do not show up in ordinary black-and-white
photographs). When Huggins studied the light of the
Orion Nebula, for instance, he found that an emission
spectrum was produced with a particularly dominating line
in the green. One could conclude from this that the Orion
Nebula and other objects like it contained masses of hot,
glowing gas.1
1 Gas is not as efficient as dust particles in scattering light. The fact
that nebulae do scatter light efficiently shows that they must be
made up of dust as well as gas. As far as we can tell at present,
dust makes up only 1 or 2 percent of the total mass of the average
nebula, but that is enough to explain the scattering ability.
 ee SAG, Pape een ee
98 _ The Universe
The light from the Andromeda Nebula was a ich
white, however, and in 1899 its spectrum was obtained
and shown to be continuous.
If the spectrum of the Andromeda Nebula had been
shown to consist of bright lines, then the matter would
have been settled. It would have been a mass of glowing
gas, of no greater significance to the general structure of
the universe than the Orion Nebula was. As it was, the
dispute continued. White light and a continuous spectrum
meant that the Andromeda Nebula might consist of a
mass of stars and be so far off that those stars could not
be made out separately. On the other hand, that conclu-
sion, was not inevitable, for gaseous nebulae might, under
- some. circumstances, Possess white light and continuous
spectra.
This was so because emission spectra were produced by
hot gases glowing with their own light. Suppose, though,
that a. mass of gas was cold and was serving merely as a
passive reflector of starlight. In that case, the spectrum of
the reflected starlight would be essentially the same as the
‘spectrum of the original. starlight itself (just as the spec-
trum of moonlight is like that of sunlight).
If the Andromeda: Nebula were merely reflecting star-
light, that would explain everything. Its spectrum would
be consistent with the theory that it was a not-very-large
patch of gas quite close to the Solar System.
But one catch remained. If the Andromeda Nebula
were merely reflecting starlight, where were the stars
whose light it was reflecting? One could easily see stars
within the: Orion Nebula, and it was the radiation from
these stars that heated the Orion Nebula into a heat great
enough to produce an emission spectrum. But where were
the stars in the Andromeda Nebula? None could be
found.
At least, no permanent stars could be found. Occasion-
ally, a starlike object was found to be associated tem-
_ porarily with the Andromeda Nebula. As this turned out
to be highly significant, let us pause in order to take up,
the matter of a eat starlike objects in some detail.

Novae

To any casual observer of the heavens, the starry config-

urations seem permanent and fixed. Indeed, the Greek
philosophers had differentiated between the sky and the
earth by this fact. On Earth, Aristotle suggested, there was
 - Other Galaxies 99
perpetual and continuing change, but the heavens were
absolutely changeless.
To be sure, there were occasional “shooting stars” |
which made it appear, to the uninitiated, that a star had
fallen from heaven. However, no matter how many shoot-.
ing stars appeared, no star was ever observed to be miss-
ing from its place as a result. Consequently, such shooting
stars were considered to be atmospheric phenomena by
the Greeks and therefore, like the shifting of clouds or the
falling of rain, to be part of the changing earth and not of
the changeless heavens. The very word “meteor” applied
to shooting stars is from a Greek term meaning “things in
the air.””?
The Greeks were correct in deciding that the flash of
light accompanying a shooting star was an atmospheric
phenomenon. The object causing that flash, though, was a
speeding body (a “meteoroid”) varying in size from less
than a pinpoint to a multi-ton object. Before entering the
earth’s atmosphere, a meteoroid is an independent body of
the Solar system. After entering the atmosphere, it heats
through friction to the point where it flashes brilliantly. If
small, it is consumed in the process; if large, a remnant
may survive to strike the Earth’s surface as a “meteor-
ite.”
Another class of temporary inhabitants of the sky were
the occasional comets, often sporting long, cloudy projec-
tions that might be considered as flowing tails or stream-
ing hair. The ancients viewed it as the latter, for “comet”
is from the Latin word for “hair.” Comets came and went.
‘erratically, so the Greek philosophers considered them to
be atmospheric phenomena also. Here, they were clearly
wrong, for the comets exist far beyond Earth’s atmos-
phere and are actually members of the Solar system, as
independent a set of members as the planets themselves.
Nevertheless, suppose we modify the Greek view and
say that change is a property of the Solar system, but that
the stars far beyond the Solar system are changeless. If we
do this, we eliminate not only meteors and comets, but
also such changes.as the phases of the Moon, the spots on
the Sun, and the complicated motions of the planets. Is
this restricted view of changelessness tenable?
To the naked eye, it would almost seem to be. To be
' # This explains, by the way, why “meteorology” is the study of the
atmosphere and the weather, and not of meteors, The ate study -
- is now termed “meteoritics.”
 MD
| er fer eAe Oe eeee hit (oe, om Oe Poke Ce beige t= Se oe ae ee
“ne NBS 7 ae

100 Y The Universe

sure the intensity of the light produced by some stars
varies (see page 66), but such cases are few and unspectac-
ular, not obvious to the casual eye. Some stars also have
significant proper motions, but this is even less noticeable,
and it would take many centuries to be sure of the.
existence of such motions without a telescope.
One type of spectacular change, however, could take
place in the heavens, and so clearly that the most casual
observer could:see it. I am referring to the actual appear-
ance of a completely new, and sometimes very bright, star
in the sky. Such stars were clearly stars and lacked all
trace of the fuzziness of comets. Furthermore, they were
not momentary flashes like meteors, but persisted for
weeks and months. : J
Not only were’ such new stars evidence of change
-among the stars by the mere fact that they appeared and
-. eventually disappeared, but also they changed brightness
radically during the course of their brief stay in the sky as
visible objects. Only the fact that such objects were. so
_ tarely encountered made it possible for the ancient astron-
omers to ignore their existence and to continue to accept
the assumption of the changelessness of the heavens.
_. There is evidence, in fact, of only one such new star
having appeared during the period of Greek astronomy,
and that evidence is none too strong. Hipparchus is sup-
posed to have recorded such a new star in 134 B.c. We do
not have his word for this, for virtually none of his works
has survived. The Roman ‘encyclopedist Pliny _ (23-79 .
A.D.), writing two centuries later, reported it, saying that it
_ was this new star that inspired Hipparchus to prepare the
first star map, in order that future new stars might be more
easily detected. ‘
Perhaps the most spectacular new star in historic times
was not observed in Europe at all, for it appeared-in the
constellation Taurus in June 1054, at a time when Euro-
pean astronomy was virtually nonexistent. That we know
of it at all is thanks to the observations of-Chinese and
Japanese astronomers who recorded the appearance of
__ what they called a “guest star” at this time. It persisted
for two years and grew so fiercely brilliant at its peak as
to outshine Venus and become easily visible by day. For
_almost a month it was the brightest object in the sky next-
to the Sun and the Moon.
Then, in November, 1572, another such object, almost
as bright;-appeared in the constellation Cassiopeia, out-
Pn ya ee

Other Galaxies S101

shining Venus, at its peak, by five or ten times. By then,
however, European astronomy was flourishing again, and
‘an astronomer of the first rank was in his impressionable
youth. This was the Danish astronomer Tycho Brahe, who
observed the new star carefully and then, in 1573, pub-
~ jished a small book about it. A. short version of the Latin
title of the book is De Nova Stella (“Concerning the New
Star”). Ever since, a star that suddenly appears where
none was observed before has been called a “nova”
(‘‘new”).

One of the important points'made in connection with

the nova of 1572 by Brahe was that it lacked a measura-
- ble parallax. This meant it had to be many times as
distant as the Moon and could not be an atmospheric
_ phenomenon and therefore part of the changeable Earth.
(Brahe made the same observation in 1577 for a comet
and showed that comets, too, were not atmospheric
phenomena.)
Then, in 1604, still another nova aapeted: this ‘time in
the constellation Ophiuchus.. It was observed by Kepler
and Galileo. While distinctly less bright than Brahe’s
nova, that of 1604 was still a remarkable phenomenon
and, at its peak, rivaled the planet Jupiter in brilliance.
Oddly enough, no superlatively bright novae have
graced the sky in the three and a half centuries since
1604. This is rather a pity, for the telescope was invented a
few years after 1604, and astronomy entered a new era in
which such spectacular novae could have been studied
much more profitably than before.
Nevertheless the telescopic revolution in astronomy at
once affected the views concerning these novae. In the
first place, it was quickly seen that the stars visible to the
naked eye were not all the stars there were by any means.
A nova, therefore, need not be a truly new star, despite its
name. It might merely be a dim star—too dim to be seen
by the naked eye, ordinarily—which, for some reason,
_brightened sufficiently. to become visible. As astronomers
began to discover more and more variable stars, such
changes in brightness came to seem neither phenomenal
nor even unusual in themselves. What was unusual about
novae was not the fact that their brightness changed, but
the extent to which it changed. Novae could be classified
as a type.of variable star, but a particular type called
“cataclysmic variables.” Their changes in _ brightness
seemed not merely the result of some more or less quiet

i -
 tae thee glBae eS eS GE, ig ARTs7 lien ne ee
(102. ~—-s The _Universe
_ periodic process, but rather the consequence of some vast
cataclysm—somewhat like the difference between a period-
ic geyser spout and an erratic and unpredictable -volcanic
eruption.
Then, too, whereas in pretelescopic days, only those
sudden brightenings which reached unusual peaks could
readily be observed, the telescope made it possible to
observe much less drastic events. .
Since novae were associated with such brightness, dim
ones were not searched for, and for two and a half centuries
no novae were reported. Then, in 1848, the English astron-
omer John Russell Hind (1823-1895) happened to observe
a star in Ophiuchus that suddenly brightened. At its bright-
est, it only reached the fifth magnitude so that it was never
anything more than a dim star to the naked eye, and in
pretelescope days it might easily have gone unnoticed. -
Nevertheless, it was a nova.
Thereafter, novae of all brightnesses were searched for
- and discovered in surprising numbers. One of them, ap-
pearing in the constellation of Aquila in 1918 (“Nova
Aquilae”), shone, briefly, as brightly as Sirius. None, how-
ever, approached the planetary brightness of the novae of —
1054, 1572, and 1604.
It is now estimated that some two dozen novae appear
each year, here and there in the Galaxy, although relative-
ly few of them are so situated as to be visible from the
Earth.

The Andromeda Galaxy

' The matter of the novae entered the problem of the

» Andromeda Nebula when, in 1885, one appeared in the
central portions of the nebula. For the first time, a promi-
nent star was seen in connection with the Andromeda
Nebula.
There were two possibilities here. The star might exist
between the Andromeda Nebula and ourselves and be
seen in the nebula only because that object was in the line
of sight. In that case the star and the nebula would have
no true connection. The second possibility was that the
Andromeda Nebula was made up of stars too dim to. be
seen and that one of them had flared up into a nova and
had become visible in a telescope.
If the latter were the case, it might be possible to deter-
mine the distance of the Andromeda Nebula if one as-
eee AS et ec
: : aes
Other Galaxies 403°
sumed: that novae always reached about the same peak of
luminosity. In that case, variations in apparent brightness
would be caused entirely by a difference in distance. If the
distance of any nova could be determined, the distance of
all the rest could then be calculated. The opportunity
came with a nova that appeared in the constellation Perse-
us (“Nova Persei”) in 1901. It was an unusually close
nova and its distance was estimated by parallax to be
about 100 light-years.
The nova that had appeared in the Andromeda Nebula,
referred to now as “S Andromedae,” reached only the.
seventh magnitude at its peak (so that it would never have
been visible without a telescope) as compared with a
magnitude of 0.2 reached by Nova Persei. If the two
novae had indeed attained the same luminosity, S An-
dromedae would have to be some sixteen times as distant
as Nova Persei to account for the difference in brightness.
It was argued in 1911, then, that the distance of S An-
dromedae was 1600 light-years.
If S Andromedae were indeed part of the Andromeda
Nebula, that meant the nebula, too, was 1600 light-years ~
distant. If S Andromedae were merely in the line of sight
of the nebula, the latter would have to be beyond the
nova and even more than 1600 light-years from us. In
either case, the nebula was at least 800 times as far from
us as had been calculated from the apparent parallactic-
data obtained in 1907. If the nebula were 1600 light-years
distant, it had to be quite large to seem as large in our
telescopes as it does. It could scarcely represent a single
planetary system in the process of formation as Laplace
had supposed. Still, one could not yet accept the Kantian
view either. Even at 1600 light-years, the Andromeda
Nebula had to be merely a feature of the Galaxy.
This line of argument assumed, however, that S An-
dromedae and Nova Persei actually reached the same
luminosity. What if this assumption were not valid? What
if S Andromedae were actually much more luminous than
Nova Persei ever was? Or much less luminous? How could
one tell?
The American astronomer Heber Doust Curtis (1872-
1942) believed that the one-way of deciding this matter
was to search for more novae in the Andromeda Nebula.
What could not be judged in the case of one specimen
might become clear in the comparative study of many. He
therefore tracked down and studied a number of novae in
 104 es "The “Universe ae .
the Andromeda Nebula,and found himself able to make
two points.
First, the number of novae located in the nebula was 80
high that there was no possibility that they were not
‘associated with the nebula. To suppose that all those
novae just happened to spring up among stars located in
the line of sight between ourselves and the nebula was
ridiculous. Such a fortuitous concentration of novae was
completely unlikely. This further implied that the An-
dromeda Nebula was not merely a cloud of dust and gas
passively reflecting sunlight. It had to consist of numerous
-stars—a very large number indeed to have so many novae
(a very rare type of star) appear among them. That such
stars could not be made out even by large telescopes
argued that the nebula was at a great distance. Secondly,
all the novae observed in the Andromeda Nebula after
1885 were far dimmer than S Andromedae had been.
Curtis suggested in 1918 that these other novae should be
compared with Nova Persei, and that S Andromedae was
an exceptional, extraordinarily bright nova.
If the ordinary novae. in the Andromeda Nebula were
set equal in luminosity to Nova Persei, then the distance
that would account for the unusual dimness of the former
would have to be in hundreds of thousands of light-years,
_ at the very least. Such a distance would also account for
the fact that the nebula could not be resolved into stars.
_ At such a distance, individual stars were simply too faint
- to be made out—unless they brightened enormously, nova-
fashion.
But if the Andromeda Nebula were indeed at such a
_. distance, it must be far outside the limits of the Galaxy
-. and, to appear as large as it does, it must be a. huge
conglomeration of a vast number of stars. It was indeed
an island universe of the type Kant had once described.
Curtis’ conclusion was by no means accepted by other
astronomers, and even Shapley was opposed to him. -
Entering the lists, however, was the American astrono-
mer Edwin Powell Hubble (1889-1953). It seemed clear
to him that the argument from novae would always seem
inconclusive since not enough was known about them. If,
however, the Andromeda Nebula were actually an island
universe, then perhaps a new telescope—more powerful
than any available to nineteenth-century astronomers—
might settle the issue by revealing the individual stars in
the nebula. From the ordinary stars, far less mysterious.
 105 |
than the novae, it might be possible to draw firmer con- :
clusions concerning the nebula. 5
In 1917, a new telescope had been installed on Mt.
Wilson, just northeast of Pasadena. It had a mirror that
was an unprecedented 100 inches in diameter, making it
by far the most powerful telescope in the world (and it
was to remain the most powerful for a generation).
Hubble turned the Mt. Wilson telescope on the An-
dromeda Nebula and succeeded in making out individual
stars on the outskirts. That was the final settlement of one
problem: the nebula consisted of stars and not of gas and
dust.
By the end of 1923, Hubble was able to identify one of
the stars as a variable showing all the characteristics of a
Cepheid. He located other Cepheids soon after.:
This was exactly what he needed. Shapley had by then
worked out the Cepheid yardstick so that the period of
variation of the Cepheids in Andromeda could tell Hubble
at once the actual luminosity of those stars, provided one
could assume that the same laws governing Cepheids in
the Galaxy and the Magelianic Clouds also governed them
in the Andromeda Nebula.
Once the Juminosity of the Cepheids in the Andromeda
Nebula was determined, one could then calculate their ©
distance from their apparent brightness, therefore, the
‘distance of the nebula. Hubble calculated this distance to
be approximately 800,000 light-years.
‘By the mid-1920’s, then, the matter was settled, and it
‘has not been questioned since. The Andromeda Nebula is
‘not a member of the Galaxy but-is located far beyond its
bounds. It is a vast and independent conglomeration of
Stars, an island universe indeed. Kant was right; Laplace
was wrong.
Hubble therefore spoke of the Andromeda Nebula as
“one of a class of “extra-galactic nebulae,” to be distin-
guished from the ordinary “galactic nebulae” such as that
in Orion. Shapley, now converted to the new view, felt .
such terms to be inadequate. The Andromeda Nebula was
not to be compared with the Orion Nebula even by
terminology, but only with the Galaxy. The Andromeda
Nebula was another galaxy in its own right, and Shapley
_ suggested that all such bodies be termed “galaxies.”
Today, therefore, we speak of the “Andromeda
_ galaxy.” We distinguish our own galaxy either by giving it
a definite article and a capital, “the Galaxy,” as“I have
 106 © ———«*‘The ‘Universe
been doing in the last few Soe or by calling it “the
Milky wey galaxy.”

The Spiral Galaxies.

Nor was the Andromeda galaxy a sport or a unique

example of something beyond the Galaxy. It was one of a
large group, although, to be sure, it was by far the largest
in appearance, and, aside from the Magellanic Clouds, the
only one visible to the naked eye.?
Messier, in his 1781 catalogue, listed some dozens of
nebulae which, like the Andromeda, could neither be
resolved: into many faint stars nor found to contain a few
bright ones, and which, eventually, turned out to be galax-
ies. William Herschel, in his general sweeping of the
heavens, located no less than 2500 such nebulae, and his
son, John Hershel, in a similar investigation of the skies of
the southern hemisphere found an equal number there.
They were found all over the sky except in the plane of
the Milky Way, and there, as astronomers came to realize,
those that might be. present were obscured by the dust
clouds and star masses associated with that plane.
By the beginning of the twentieth century, some 13,000
nebulae of the Andromeda type were known, and there ~
was every indication that many more remained to be
found.
The Irish astronomer William Parsons, 3rd Earl of
Rosse (1800-1867), studied these nebulae more closely
than did anyone else in the nineteenth century. He had
built a 72-inch telescope on his estate, although the
weather was usually so poor that he got very little use out
of it. Nevertheless, he studied the nebulae and noted that
some of them seemed to have a distinctly spiral structure,
almost as though they were whirlpools of light set against
the black background of space.
The structures appear flat and, when seen edge-on,
seem to be merely elongated lens-shaped objects (like our
own Galaxy) with the spiral structure, if any, invisible. On
the other hand, some nebulae are seen squarely broadside
on so that the spiral structure is fully visible. A particular-
ly spectacular example is “M51,” more dramatically called
the “Whirlpool Nebula” (or, today, the “Whirlpool galaxy”
from its paebe.

8 The Anaron@eld galaxy is, indeed, notable for being the very
farthest object that can be made out by the naked eye.
 > Naser i raed ‘ e ay. *

Other Galaxies” SH) 107

_ As a result, astronomers began to speak of a new class
of objects, the “spiral nebulae” or, as they are now called,
the “spiral galaxies.” They consist of a central condensa-
tion, relatively small in some cases, much larger in others,
_ the “galactic nucleus.” Outside are the “spiral arms.”
There seems to be a distinct difference in properties
between the galactic nuclei and the spiral arms. The.
nuclei resemble huge globular clusters and, like the clus-.
ters, seem relatively free of dust clouds. The spiral arms,
on the other hand, are rich in dust clouds, which are often
clearly visible.
The dust is most prominent in the case of some spiral
galaxies that we happen to view edge-on. An example is
NGC 8914 in the constellation Andromeda. The dust
_ Clouds elong its equator make a dark, ragged line down its
length. This is also true of a beautiful. galaxy in the
constellation Virgo, in which the galactic nucleus is enor-
mous and the spiral arms rather compact. It is seen almost
. edge-on, and the dust in the arm forms a tight ellipse
around the rim. The effect is almost that of seeing Saturn
with dark rings. The dark rim reminds one of a decorated
hat brim; and the aie is familiarly called the “Sombrero
F galaxy. ”

: About 75 percent of the galaxies have a spiral struc-

_ ture. This includes the Andromeda galaxy certainly, al-
_ though we see that object so nearly edge-on that the spiral
arms are not as clearly seen as we would like. Still, the
appearance is striking enough to give the Andromeda
galaxy a swirling appearance and make Laplace’s sugges-
tion that it is a collection of rotating gas all too plausi-
D.. ble.
It’ is usually considered that our own Galaxy is also _
spiral in nature and, indeed, very much like the Androme-
da galaxy in appearance (although recent data seem to
indicate that the Galaxy possesses a less prominent nucle-
us). The Sun is located in one of the spiral arms of the
Galaxy, and it is for that reason that we are surrounded
by dust clouds that obscure the main body of the
Galaxy.
About 20 percent of the galaxies are spheroidal or
ellipsoidal and seem to be composed of galactic nuclei

4 Galaxies are often referred to by their number in the listing in

the New General Catalogue of Nebulae and Clusters, published from
___ 1888 to 1908 and usually referred to by the abbreviation, NGC.
 108 _ The Universe
ou without spiral arms. These are generally referred to
“elliptical galaxies.”
* The remaining 5 per cent are “irregular galaxies” with
no well-defined simple structure at all. The Magellanic
Clouds are often considered to be the best-known exam- —
ples, but this view was challenged in the -1950’s by the
French-American astronomer Gerard Henri de Vau-
couleurs (1918- ), who. maintained they were spiral
in nature. The Large Magellanic Cloud, he pointed out,
has a single arm extending outward for some tens of
thousands of light-years. The Small Magellanic Cloud is
seen virtually edge-on, so that its spiral structure, if
present, cannot be made out.
Modern astronomers can set no limit to the number of
galaxies. Better astronomic tools and more careful obser-
vations continually raise the number that can be seen, and
there seems, as yet, no sign of any thinning out in any
direction. Astronomers suspect that the total number of
galaxies in the parts of the Universe we can observe with
our best instruments may be as high as 100,000,000,000,
and there is, at present, no reason to think that the actual
number may not be indefinitely higher.
Furthermore, they Andromeda galaxy is the brightest
member of the class of undoubted spirals and is the
nearest large one. If it is 750,000 light-years away, then
the dimmest galaxies that can be made out must be many
hundreds of- millions of light-years away, possibly even
billions of light-years.
By 1925, man’s notion of the Universe had received
another colossal enlargement. Indeed, astronomers found
themselves facing the problem of infinity once again. A
century earlier, Olbers’ paradox had seemed to argue
against the possibility of an infinite Universe, and Her-
schel’s observation of a finite Galaxy had tended to bear
that out. For a century the notion of a finite Universe had
- reigned supreme. —_.
But now there was no sign of finiteness in the new and
larger Universe of the galaxies, and once again the astron-
omers would have to tackle the problem of Olbers’ para-
dox, with galaxies taking the place of stars.
This time, however, the problem of the extension of the
Universe in space—whether it was infinite or not—turned
out to be intimately connected with the allied question of
the extension of the Universe in time—whether it was
eternal or not.
 a. e. 3 +. eRe!
n
7‘Untilnow, ‘I have been considering only the aplega af 7a
* extension. in space. Before continuing further in that direc-
_ tion, it will be useful to turn to the problem of extension
in time.
 A7 “ Be aa in - ‘- a po” OG! i *
ee
7 ca —

CHAPTER Te

| The Age of the Earth

Angular Momentum

As long as astronomers considered the heavens to be

changeless, the implication was that they were eternal as
well. Certainly, a beginning and an ending are the most
drastic of all possible changes, and what could be observed
of the heavens in ancient times gave no evidence whatever
of any possible beginning or ending. (Men might speak of
the creation or destruction of the Universe through some
superhuman ageney and go on to describe the processes in
detail, but such descriptions arise from internal inspiration
and not from any actual astronomic evidence.)
By early modern times, however, it was recognized that
the heavens were not absolutely changeless; the novae
were the best evidence of that. The question therefore
arose as to whether the possibility of change implied the
ultimate change of a beginning and an ending and, if so,
when the Universe might have begun and when it might
end.
It was easiest to tackle the problem first in connection
with the Solar system, which by 1700 was already under-
stood in detail.
In 1687, Newton had established the theory of univer-
sal gravitation, according to which every body in the
Universe attracted every other body with a force that was
proportional to the product of their masses and inversely
proportional to the square of the distance between them.
In the Solar system, the Sun was so overwhelmingly
predominant in mass that it remained almost motionless,
while the much less massive planets responded to the force
of Solar gravitation by circling the Sun in elliptical orbits.
(Actually, the Sun also moves-in response to the gravita-
110
 RGSS ge wy s
; tna RIM

The Age ofthe Earth eas i.

tional pull of the planets. The center of gravity of the ©
Solar system about which the Sun and planets move is
near the center of the Sun but not quite at it. Indeed, it is
- sometimes so far away from the Sun’s center as to be a
little outside the Solar sphere. This, however, is a mere
detail in a larger. picture, and the planets can be viewed as
revolving about an essentially motionless Sun without too
great an imprecision.)
The Solar system cannot be looked on as changeless, in
the sense of being fixed and immobile, because all its com-
ponent parts (even the Sun itself) are constantly moving
with respect to the center of gravity of the system.
But if there is not a static equilibrium, there may, at
least, be a dynamic one. That is, although all the parts of
the system move, it may be that all the motions are
periodic, repeating themselves over and over, endlessly
and without significant change, and in that respect the
Solar system might be considered to be changeless.
Or is it safe to. assume that the motions are truly
‘periodic? To -be sure, the Earth would revolve about the
Sun in an absolutely periodic manner, never changing its
orbit, if the Earth and Sun were the only objects in the
- Universe. But they are not; there are other objects and
each of these affects the Earth-Sun system gravitationally.
“Neighboring planets affect the Earth’s motion through
their gravitational influence; so aor the Moon, so do even
the distant stars.
These minor effects on the Earth’s motion (“pertur-—
bations”) must be taken into account’in refined calcula-
tions of the Earth’s orbit. Perturbations also affect the
motions of the other planets.
In the short run, these perturbations do not seriously
affect the Solar system. Throughout man’s history, the day
and the year have remained essentially unchanged, and the |
motions of the planet have persisted with a grand stead-
iness, Still, the history of astronomic observations is, at |
best, only a few thousand_ years old, and this could be
considered a mere moment in the history of the Solar
System. What about the long run?
Theoretically, the law of universal gravitation could be
applied to predict the motions of every object in the Uni-
verse under the gravitational influence of every other
body. The machinery of the Solar system could then be
_ run (mathematically) forward and backward in time
through indefinitely long stretches—as is literally done, for
short stretches of time, in modern planetariums. In this
i,
.
 112 _ The Universe — ;
way, one could check as to whether there were any
systematic changes that might have pulled the bodies of
the Solar system together in the far past or that might
force them apart in the distant future.
Unfortunately, such a direct study is not practical. The
equations one must set up to account for the motions of
only three. bodies, gravitationally interlocked, are too com-
plicated for complete solution. What, then,'can be done for
a Solar system consisting of a dozen major bodies and-
uncounted numbers of minor ones? Short-cut approxima-
tions had therefore to be made and even they required the
full-time attention of first-class minds. E
The problem was tackled by the French astronomer |
Joseph Louis Lagrange (1736-1813) and was then fol-
_ lowed up'by Laplace. It was Laplace who finally solved
. the matter satisfactorily in his book Celestial Mechanics,
published in five volumes from 1799 to 1825. He showed
that while perturbations introduced small changes into
planetary orbits, these changes were periodic; that is, the
orbit would alter its properties in one direction, then back
in the other, and so on indefinitely. Over the long-run, the
average shape of the orbit would remain constant.
‘In other words, the Solar system was in dynamic equilib-
tium and could continue indefinitely into the future and
- might already have existed through an indefinite past.
(This is predicated on the assumption that there is no
introduction of overriding influences from without the
Solar system, that no star invades our immediate neigh-
borhood, that the stars as presently situated exert gravita-
tional influences too small to matter, and so on. This is a
pretty safe assumption even over long periods of time.)
Although there is nothing in the sheer gravitational
mechanics of the Solar system to prevent it from being
eternal, the concept of eternity remains a difficult one, no
more acceptable to the average mind than that of infinity.
_ For that reason, beginnings were sought by adding some-
thing else to gravity.
For instance, if the gravitational attraction of the Sun
were the only source of movement in the Solar System, the
various planets would respond by falling directly toward
and into the Sun. The fact that they revolved about the
Sun, as did the comets and asteroids, while the various
Satellites revolved similarly about their planets, meant that
each body possessed a motion more or less at right angles
to the pull of the Sun’s gravitation. This motion did not
 The Age of the Earth 113
and could not originate from the Sun’s gravitational at-
traction. How then did it come to be?
This can be put in another way. Any essentially circular
motion, whether that of an object rotating about its own
axis, or of an object revolving, as a whole, about a larger
body, involves the possession, by the moving body, of a
property called “angular momentum.” The quantity of
angular momentum possessed by a body depends on three
things: its mass, the speed of its circular motion, and the
distance of the body (or the various parts of it) from the
center about which it circles.
By the eighteenth century it was quite clear to physi-
cists, through observations of phenomena on Earth, that
angular momentum was neither created nor destroyed but
that it could be transferred, without loss or gain, from one
body to another. This is the “law of conservation of
angular momentum.” There was (and is today) no reason
to think that angular momentum is not conserved in the
Universe at large as it is on the Earth. In that case, any
theory that describes the beginning of the Universe or any
large part of it must not involve any violation of this
law. Shae
If angular momentum cannot be created, how does it
come to exist? One way out of the apparent dilemma
involves the recognition that angular momentum can exist
in two varieties according to the direction of spin. There
can be “clockwise” angular momentum and “counter-
clockwise” angular momentum, these varieties being
related to the conventional direction of turning of a
clock’s hands. f
If clockwise angular momentum is considered positive,
then counterclockwise angular momentum can be consid-
ered negative. Equal quantities of the two varieties can, in
adding together, cancel, and form a combined system with
no angular momentum, Similarly, a system without angu-
lar momentum can split into two systems, one of which
has a certain clockwise angular momentum and the other
an equal counterclockwise angular momentum. In this way
angular momentum seems to be destroyed and created
without, however, actually breaking the law of conservation
of angular momentum.
One might suppose, for instance, that at the beginning
the Universe did not contain angular momentum, but that
in the process of forming, some portions obtained one
variety and the rest the second variety.
If the Solar system is viewed from a position high above

=
 114 The Universe -
the Earth’s north pole, then the Sun and Earth and most
of the other bodies of the system will be viewed as rotating
about their axes in a counterclockwise direction. The
planets and satellites will almost all be seen to be revolv-
ing about their central bodies in a counterclockwise Uirec-
tion. This means that the Solar system does not have
equal quantities of the two kinds of angular momentum
and cannot be. viewed as a system essentially without
angular momentum. Rather, the Solar system has a great
deal of angular momentum, and any theory that explains
the origin of the system must take that into account.
Suppose, for instance, that the Solar system began as a
vast, thin cloud of dust and gas, in accordance with the
suggestions of Laplace when he advanced his nebular
hypothesis (see page 96). This cloud might already possess
a supply of angular momentum, a supply it had received
for its share when the Universe as a whole came into
being. Or else, if the cloud is supposed to have no angular
momentum to begin with, it might come under the influ-
ence of the very feeble gravitational attraction of some
comparatively nearby star. This would tend to pull more
strongly at the end of the cloud nearer itself than at the
opposite end. This would exert a “torque” on the cloud
and pull it into circular motion. The angular momentum
would have been supplied the cloud at the expense of the
quantity originally possessed by the attracting star, of
course, and the star’s own supply would be corresponding-
ly diminished. :
_ Whatever the source of the spin, the slowly rotating
cloud would be under the influence of the mutual gravita-
tional attraction of its constituent particles and would slow-
ly contract. As the cloud contracted, its various portions
' would be closer and closer to the center about which all
were revolving. As the overall distance to the center
decreased, this, taken by itself, would tend to destroy
some of the angular momentum, unless the decrease were —
balanced by an increase in the velocity of turning. (Angu-
lar momentum depends on both factors and on mass in
addition, but mass is not changed under conditions de-
scribed, so only distance and angular speed need be con-
sidered. The decrease of either automatically implies an
increase in the other.) ;
The law of conservation of angular momentum there-
fore makes it necessary for the vast rotating .spheroid of
gas to spin faster and faster as it contracts. The equator
bulges under the influence of a steadily increasing centrif-
 The Age of the Earth 115
ugal effect, turning the spheroid into a more and more
flattened ellipsoid. Finally, portions split off the equatorial
plane of the ellipsoid at. different intervals and condense
into planets.
Laplace’s nebular hypothesis seemed to explain a great
deal, and it bore the appearance of great plausibility.
It was very popular with astronomers throughout the
nineteenth century and with the general public as well.
Indeed, as the decades of the nineteenth century passed,
the nebular hypothesis seemed to fit in very well with
certain key discoveries in physics and to offer a mettiod
for determining the age of the Earth.

The Conservation of Energy

In the '1840’s, a new and even more powerful conserva-

tion law than that governing angular momentum came to
be established quite firmly. This was the “law of conserva-
tion of energy”; it was the product of the work of many
men, but it was first clearly enunciated by the German
physicist Hermann Ludwig Ferdinand von Helmholtz
(1821-1894) in 1847.
The law of conservation of energy states that energy
can be transferred from one place to another, but cannot
. be created out of nothing or destroyed.
The Solar system has an enormous store of energy as well
as of angular momentum, and the question must arise as to
where that energy comes from.

The nebular hypothesis

There is a vital difference, here, between the question of

the Solar system’s energy supply and its supply of angular
Se
1

116 - ‘The ‘Universe

momentum. As far as angular momentum is concerned,
_ one can say that the Solar system’s supply was brought
into existence when the Solar system was formed (as
described in the preceding section), and once that is
stated, one can relax. No significant quantity ofthe angu-
lar momentum of the Solar system is being lost. It can be
lost only by interaction. with the enormously distant stars
“or with the incredibly thin-wisps of matter between the
stars. Such processes would take away from, or add to,
the Solar system’s supply of angular momentum at so slow
a rate compared with the total supply present that the
whole process could be ignored even over long- stretches of
time. It seems perfectly safe to assume that the Solar
system possesses as much angular momentum now as it
did, say; millions of years ago, or as it will possess millions
of years from now.
_ Not so with energy. Energy exists in the Solar system in
a variety of forms, and one of these forms is that of the
radiation that constantly streams out ‘of the Sun. The
amount of energy represented-by this radiation is simply
colossal, and virtually all of it pours outward in all direc-
tions, at.an enormous rate, into the vast spaces beyond the
_ Solar system. Moreover, as nearly as we can tell, virtually
none of it ever returns. mis
This means that the total quantity of energy in the
Solar system must constantly be decreasing. It must some-
day decrease to a point so close to zero that the Solar
system, as we know it, may be considered at an end. If we
‘ look backward in time, the amount of energy in the Solar
system is greater and greater the farther back we go.
Unless we are content to postulate an infinite quantity of
energy to begin with (which no astronomer is), there
_ must be some definite point at which the Solar system
began with some original. store of energy much greater
than that which it possesses today.
Helmholtz himself was the first to worry about this
matter and to question the source of the energy that the
Sun so recklessly spilled out into ‘space in quantities so
magnificent that the utterly contemptible portion inter-
cepted by the tiny Earth at a distance of 93,000,000 miles
was sufficient to supply all the energy needs of man with
copious excess.
-The most common source of man-made energy in the
nineteenth century was that of burning coal. In this case,
heat and light were obtained at the expense of the chemi-
cal energy binding atoms together. When coal and oxygen
 ee ih eS

The Age of the Earth 117

combined, the carbon dioxide that formed required small-
er quantities of chemical energy to.keep its molecules in
being than was required by the original coal and oxygen.
The excess energy no longer required for atomic bonding
was then expelled as heat and light.
But the amount of energy that could be produced by
combining a given quantity of coal and oxygen was
known, and the amount of energy put out by the Sun each
second was also known. It is not difficult to show that if
the entire mass of the Sun (also a known quantity) were -
made up of coal and oxygen in correct proportions, the
resultant coal fire would keep the Sun going at its present
rate for only 1500 years. t
- Although there are some chemical reactions that yield
more energy per pound than a coal fire does, there is not
one known that would keep the Sun going throughout
historic times, let alone during the long eons of prehistoric
times. Helmholtz had to search for his energy source
elsewhere.
; One colossal source of energy was the gravitational field
itself. A meteorite striking the Earth’s atmosphere, while
moving at the high speed énforced by its response to the |
gravitational fields about it, converted the energy of mo-
tion into light and heat. Even a pinhead meteorite was
sufficient to produce a brilliance that could be seen for a
hundred miles or more.
Suppose, then, that meteors were constantly plunging
into the Sun. and that the energy of their gravitationally
induced motion were converted into radiation. It can be .
shown that if such meteorite collisions took place at the
rate necessary to keep the Sun going, the Sun would not
undergo any visible change in the course of historic times.
Indeed, such bombardment could continue for 300,000
years before the Sun would gain 1 percent in mass as a
result.
This sounds hopeful but additional thought spoils the
picture. The mass of the Sun would increase very slowly
as a result of the quantity of meteorite bombardment.
required to keep it going, but that increase would be
enough to introduce a perceptible strengthening of its
gravitational field. The Earth, more powerfully attracted
each year, would move more and more rapidly, so that
each year would be two seconds shorter than the year
before. This does not sound like much, but astronomers
could detect such a change in the year very easily and
 BR Ae eat he
us The Universe |
virtually at once. Since no such change is detected, the
- meteor infall theory must be abandoned.
Helmholtz turned, in 1853, to another alternative. What
if the Sun itself were contracting so that its own outer.
layers were, so to speak, falling toward the center. The
energy of. this gravitationally induced motion could be
converted into radiation just as that of meteors could be,
and without any overall change in the Sun’s mass either,
so that the Earth’s year could be left unchanged.
The question was: How much contraction would have
to be postulated in order to keep the Sun’s radiation going
at the necessary rate? The answer was: Very little.
It could be shown that in all the 6000 years of man’s
civilized history, the Sun’s diameter would have contracted
by only 560 miles which, in a total diameter of 864,000
- miles, can certainly be considered insignificant. The
shrinkage in diameter of the Sun over the 250 years from
the invention of the telescope to Helmholt’s time would be
_ only 23 miles, a quantity that would be undetectable even
_ with the most refined astronomical instruments of the time.
The source of the Sun’s energy was thus explained in
terms that seemed satisfactory at the time. In fact,
Helmholtz’s contraction hypothesis ‘could be combined
- with Laplace’s nebular hypothesis, and one could then
envisage energy as having been produced constantly all the
time Laplace’s original nebula was contracting. The Sun’s
- present-day contraction. would only be a final phase of the
general nebular contraction.
Moreover, if one assumes that energy was being pro-
duced by this contraction all along, at the same rate at
_ which it is now being produced, then one can calculate ;
when the original nebula had attained any particular de-
gree of condensation on its way toward the formation of
Hs comparatively small, hotly glowing Sun that exists
today.
For instance, 18, 000,000 years ago, the original nebula
would have contracted to a diameter of some 200,000,000
miles, and its still bloated sphere would fill all the space
. out to the present orbit of the Earth. It would have to be
18,000,000 years ago that the ring of matter (according
to Laplace’s view), which would eventually condense to
form the Earth itself, was liberated. The Earth could not,
in consequence, be older than 18,000,000 years.
By this line of argument, the planets closer to the Sun
“than Earth—Venus and Mercury—would have been
formed considerably less than 18,000,000 years ago, while
 Pore. 2 o- ¥, Lome « s aa

The Age of the Earth 119

the outer planets, Mars, Jupiter, and so on, would have
been formed earlier. The entire lifetime of the Solar
system, from the beginnings of the nebular contraction,
might be several hundred millions of years.
Nuclear Energy

Had Helmholtz presented his theory in 1803 rather than

in 1853, the time allotment of 18,000,000 years for the
Earth’s existence would have seemed satisfactorily long,
even excessively so. Indeed, as the nineteenth century
opened, most European scientists were still under the spell
of the literal language of the Bible and assumed that the
Earth had existed for only 6000 years or so. Eighteen
million years would have seemed a blasphemously large -
figure to most of them.
But the first half of the nineteenth century had seen an
important revolution in attitude. In 1785, the Scottish
geologist James Hutton (1726-1797) had published a
book entitled Theory of the Earth in which he studied the
slow changes that the Earth’s surface underwent—the
layering of sediment, the erosion of rocks, and so on. He
suggested the “uniformitarian principle” which held that
whatever changes were going on today had been going on
at essentially the same rate throughout the past. Ac-
cording to this principle, it would take enormous stretches
_ of time to produce all the thicknesses of sediments that
could be found, all the erosion that could be observed, all.
the buckling and other forced changes to which the
Earth’s surface had been subjected.
Hutton did not persuade his readers at the time, but
between 1830 and 1833, another Scottish geologist,
Charles Lyell (1797-1875), published The Principles of
Geology. In this book, Hutton’s work was summarized,
popularized, and backed by additional evidence. This
eventually turned the trick, and geologists began to inter-
pret the Earth’s history in terms of hundreds of millions of
years.
When Helmholtz emerged with his figure of 18,000,000
_ years as the extreme age of the Harth, geologists were
astonished. For a ring of dust and gas to appear 18,000,-:
000 years ago, slowly condense and undergo the changes
required to form a compact solid body, with an ocean and
atmosphere, and then proceed to undergo all the further
changes, after solidification, for which the Earth’s crust
gave evidence, seemed simply impossible.
 oi?
ey as
4200 “The Universe aa
Furthermore, biologists were coming to the conchisied
that life forms had been slowly changing over the course
of time. In 1859, the English naturalist Charles Robert
Darwin (1809-1882) published The Origin of Species in
_ which he argued that such changes had been. brought
- about by the pressures of natural selection, a process that
- was excessively slow and required eons of time to produce
' the changes observediin fossilized record of extinct forms
_ of life.
Darwin’s views won out over the Bible-centered preju-
dices of the time only with great difficulty, but more and
_more biologists came to accept them and they, too, found
_ they could not swallow Helmholtz’s figure. Yet there
- seemed no disputing Helmholtz’s logic, and no quarreling
with the law of conservation of energy. :
The last half of the nineteenth century witnessed a
standoff, then, on the question of the time of origin of the.
_ Solar system, and of the Earth in particular. Physicists
- supported a short lifetime, geologists and biologists sup-
_ ported a long one.
; The standoff was Rkond in the 1890’s, when the science
of physics underwent a revolution. In 1896, the French
physicist Antoine Henri Becquerel (1852-1908) discov-
ered that uranium compounds served as a continuing
source of high-energy radiations. (This phenomenon came
to be called “radioactivity.”) Apparently there were sourc-
es of energy much more intense than those involved in
_ chemical reactions or even in gravitational contraction.
By 1911, the New Zealand-born British physicist Ernest
Rutherford (1871-1937) had succeeded in demonstrat-
ing that the atom was not a featureless sphere but con-
_ sisted of a tiny “atomic nucleus” at the center, which
'. contained virtually all the mass of the atom and which
- was surrounded by ‘light particles called “electrons.”
_ Chemical reactions involved the forces holding electrons in
place around the nucleus, and this was the source of
energy for phenomena such as the burning of coal.
The atomic nucleus was itself composed of particles,
which were eventually discovered to be of two varieties,
“protons” and “neutrons.” They were held together in the
nucleus by forces many times stronger than those that
held the electrons to the nucleus or that held different
atoms or molecules together. There are “nuclear reactions”
that involve shifts in proton-neutron combinations and that
yield much greater intensities of energies than any chemical
reaction could. Radioactivity is a form of nuclear reaction.
er oe The.Age ‘of the Earth— 4a
One aspect. of nuclear reactions was brought rather
surprisingly to the fore in 1905 by. the German-Swiss .
physicist. Albert Einstein (1879-1955). He showed that
mass itself was a very concentrated form of energy, and
he presented the now well-known formula: e — mc?,
where e represents energy, m represents mass, anc c
represents the velocity of light in a vacuum.
If we remember that the value of c is very high (300,-
000,000 meters per second) and that the value of c? is
this tremendous quantity multiplied by itself to produce
something vastly more tremendous, we see that even a
small quantity of mass is equivalent to a large quantity of
energy. Thus, 1 gram of mass can be converted into
21,500,000,000. kilocalories, a quantity that could also be
obtained by the complete burning of 670,000 gallons of —
gasoline. ~
Release of energy is always at the expense of disappear-
ance of mass, but in ordinary chemical reactions, energy is
released in such low quantities that the mass-loss is insig-
nificant. As I have just said, 670,000 gallons of gasoline
must be burned to bring about the loss of 1 gram (1/27
of an ounce). Nuclear reactions produce energies of much
greater quantities, and here the loss of mass becomes large
enough to be significant.
. Suppose, for instance, that the Sun received its energy
not at the cost of gravitationally induced contraction, but
as the result of some nuclear reaction proceeding within
it. How much mass would it have to convert into energy
in order to radiate energy at its observed rate? This can’
easily be calculated; it turns out to be 4,600,000 tons of
mass per second, This mass would be permanently lost to
the Sun, for the energy into which it is converted would
be radiated out into interstellar space.
Is it possible for the Sun to support this steady drain of
mass at the rate of millions of tons per sécond? Yes, it
certainly is, for the loss is infinitesimally small compared
with the total vast mass of the Sun and trillions of
years would have to pass before the loss at such a rate
could consume even 1 percent of the mass of the Sun.
Nor would the loss of mass seriously affect the nature of
the Earth’s gravitational field. The loss proceeds at a rate
of only a thirty-millionth that of the gain produced by the
meteor-infall theory. The mass loss produced by the nu-
clear reaction theory would weaken the Sun’s gravitation-
al field only to the point at which the Earth’s year would
 wwEZZ - The Universe
increase in length by only one second in 15,000,000 Yee
Such an increase in length is insignificant.
- The nuclear reaction theory requires no perceptible
change in the volume or appearance of the Sun over an
extended period of time, and it might have existed in very
much its present form (and the Earth, too, therefore) not
. only for tens of millions of years, but for billions of years.
The geologists and biologists were vindicated, and Helm-
. holtz’s short-lifetime suggestion went by the boards.
Radioactivity itself offered a new method for determin-
ing the age of the Earth, and one that was more accurate
and reliable than anything previously known.
As uranium gives off its radiations, its atoms change
their nature, becoming other kinds of atoms which also
give off radiations and change nature again. Eventually,
the uranium is converted into lead, which is stable and
changes no. further.
The rate at which uranium changes in this manner
follows a simple rule, well known to chemists as a “first-
order reaction.” This means that if the rate of change is
determined over a short interval of time, it can be predict-
ed, quite accurately, over any longer interval. It could be
shown, for instance, that half of any quantity of uranium
- would break down and change to lead in 4,500,000,000
years. This tremendous time interval is called the “half-
life” of uranium-238 (the most common fornmr of the
uranium atom).
Suppose, now, that you consider a rock containing
uranium compounds. Inside it, the uranium is constantly
breaking down and turning into lead. If the rock remains
solid and unbroken, the lead atoms formed cannot pos-
‘sibly escape but must remain intermingled with the urani-
um. The uranium compounds may have been pure to begin |
with, but they become increasingly contaminated by lead.
Since the rate at which nuclear reactions proceed is not
affected by the puny changes in temperature and pressure
encountered on the Earth, we know that the exact quantity
of lead accompanying the uranium depends only on the
length of time the rock has remained solid (and on the
quantity of lead present originally) and not on any unpre-
dictable Sayicoumental changes to which it may have been
subjected.
This was pointed out as early as 1907 by the American
chemist Bertram Borden Boltwood (1870-1927);in the
years following, rocks were subjected to analysis for urani-
um and lead content, and techniques were worked out for
 "The Age of the Earth = —=—-123
| cpaverting these analyses into age measurements. Within a —
few years, rocks were discovered that, by the uranium-
lead method, must surely have been lying in an undis-
turbed solid state for periods in excess of a billion years.
In the last few decades, a variety of age-determining
methods based on one form of radioactive change or
another have yielded a reliable value of 4,700,000,000
years for the age of the Earth—a period 260 times as long
as that suggested by Helmholtz.
 Be
dal=o 2s

CHAPTER $

The Energy of the Sun

The Planetesimal Hypothesis

With the great multibillion-year age of the Earth well-

established by 1920, it was natural to ask what the age of
the Sun might be. If the nebular hypothesis were a true
picture of the development of the Solar system, it would
follow that the oldest planet was the outermost, the young-
est the innermost, and the Sun in its present form younger
- than any planet. If the age of the Earth is taken as 4.7
eons,! it would follow that the Sun!must be somewhat
younger than 4.7 eons, but not, perhaps, much younger
than that.
Unfortunately, no such easy conclusion could be
reached, for the nebular hypothesis, which had remained
generally popular throughout the nineteenth century, had
‘gone out of fashion at the turn of the century.
The problem over which the nebular hypothesis stum-
bled and fell was angular momentum. The nebular hy-
pothesis began with a vast quantity of dust and gas that
contained a supply of angular momentum. It visualized the
process of condensation with its concomitant steady in-
crease in the rate of rotation of the cloud and the eventu-
al splitting off of successive shells of dust and gas. It made
no attempt, however, to describe how the angular momen-
tum was to be divided between the shells that split off and
formed the planets and the main portion of the cloud that
continued to condense to form the Sun.
In 1900, the American geologist Thomas Chrowder

1 It is becoming increasingly common to make use of the word

“eon” (ordinarily used for any indefinitely long period of time) to
stand for a billion years.
124
RRM eA ee Penh eh oe yah
: _ The on of the Sun 125
Chamberlin (1843-1928) worked out the dynamics of a
spinning nebula very carefully. He showed that as the neb-
‘ula gave up a shell of equatorial matter and continued to
contract, virtually all the angular momentum would have
to remain with the main body of the nebula and very little
could be left with the shell. If such .a shell could then
coalesce into a planet (a very doubtful process, as it
turned out), the planet would have very littie angular
momentum, The final result would be a Solar system in
which the central Sun would contain almost all the angu-
lar momentum of the entire system and would therefore.
be turning on its axis very rapidly, with a period of about
half a day, in fact. The planets would contain so little
angular momentum that it was doubtful whether they
could stay in any reasonable orbit at all.
But this is not the picture of the Solar system as it
actually exists. In actual fact, the single planet Jupiter,
with only about 0.2 percent of the mass of the Solar
system, contains fully 60 percent of the total angular
momentum in the system. Despite the fact that Jupiter has
eleven times the diameter of the Earth, the period of
rotation of the larger planet is 10 hours, less than half the -
period of Earth’s -rotation. Add the other planets and
minor bodies of the system (all of which taken together
- contain less than 0.1 percent: of the mass of the system)
and the total planetary angular momentum is found to be
98 percent of the system-wide total. The Sun with over
99.8 percent of all the mass of the Solar system contains
only 2 percent of all its angular momentum and rotates on
its axis with majestic deliberation completing a turn only -
after 24.65 days.?
How could a nebula contract and in doing so, shift
almost all its angular momentum to the tiny shells of
matter it gave off? Chamberlin could find no plausible way
to account for this. He had to conclude that the angular
momentum was brought into the Solar system from out-
side.
In 1906, Chamberlin, together with the American ‘as-.
tronomer Forest Ray Moulton (1872-1952), suggested a
way out. Picture the Sun, to begin with, in much its
present form, but without planets. Perhaps it had con-
2 This is the period of rotation at the Sun’s equator. The Sun is
not a solid body and does not rotate all in one piece as does the
Earth. Points north and south of the equator rotate with a longer
period. At 50° north or south latitude, the period is about thirty-
one days.

— oe «© , a
 g ere

126 The Universe

densed from a nebula to begin with, but if so, it condensed
without liberating shells of matter; or, if shells were liber-
ated, they lacked sufficient angular momentum to remain
independent but gradually fell back into the main body or
drifted off into space. In any case, the Sun exists in lonely
splendor. ;
Picture a second star approaching the Sun. The tremen-
dous gravitational forces that result would produce huge tides
on both stars. Perhaps a gigantic gout of star-matter might
Tise out of both stars and form a temporary bridge between
them. As the stars passed each other, this bridge of matter
would be forced to swing rapidly round and would gain
angular momentum at the expense of the stars themselves.
Once the stars separated, each would carry off some
share of the matter-bridge, which would then coalesce into
planets, still retaining the angular momentum they had
gained. Before the approach the two stars would have
_ been fast-spinning with no planets; after the approach they »
would be slow-spinning with orbiting planets.
The objections to the nebular hypothesis seemed conclu-
sive, and the Chamberlin-Moulton theory seemed a neat
- substitute. In fact, it seemed particularly attractive since it
allowed what was almost a biological motif to enter as-
tronomy. It was as though planets were formed by a kind
of marriage between two stars, as though the Earth had
both a father and mother. This hypothesis was undisputed
- for nearly forty years. ;
Because Chamberlin and Moulton pictured the matter
pulled out of the Sun as quickly condensing into small
solid bodies or “planetesimals,” which in turn further
coalesced into planets, their suggestion came to be called
the “planetesimal hypothesis.”

’ The planetesimal hypothesis

In 1917, the English astronomers James Hopwood

Jeans (1877-1946) and Harold Jeffreys (1891- )
worked out the planetesimal hypothesis in still greater
detail and suggested that the bridge of matter pulled out
between the stars would be cigar-shaped. It would be from
 ee
Peers
Oeeeetn
ys
hee
ete be
te fe
The Energy of the Sun =— 1127
the fattest portions of the bridge in the middle that the
giant planets, Jupiter and Saturn, would form, while small
planets would form beyond Saturn and within the orbit of
Jupiter.

Constitution of the Sun

If the planetesimal hypothesis is accepted, it is no long-

er safe to assume that if 4.7 eons is the age of the Earth;
it is roughly the age of the Sun as well. How long might
not the Sun have existed in lonely splendor before the
x intruder blessed it with a family? Might not the planetary
system be a relatively late addition to a Sun whose exist-
ence might be measured in tens or even hundreds of
eons?
Such an extreme lifetime for the Sun became vaguely
conceivable once the interconversion of mass and energy
was understood. The Sun maintained its radiation at the
expense of mass, but who could say, at first, how large its
original mass might have been? If the Sun had originally
been twice its’ present mass and had been losing mass
constantly at its present rate, it would have existed for
1500 eons before attaining its present mass. And, of
course, it could then continue to exist another 1500 eons,
while radiating at its present rate, before disappearing
altogether.
However, it is eeiemiely unlikely that mass can be lost
at a steady rate until it is all gone. It was the experience
of physicists working with atomic nuclei that energy was
usually produced at the expense of mass when one set of
nuclei rearranged itself into another set of nuclei. Only a
small fraction of the total mass was, under such circum-
stances, converted into energy. If the Sun gained its ener-
gy from nuclear reactions proceeding within itself, it
could, at best, lose only a small fraction of its mass. Then,
when all its matter had been rearranged into the product
nucleus, nuclear reactions would stop. Little, if any, ener-
gy might be formed thereafter, even though great quanti-
ties of mass were still to be found in the Sun.
The quantity of. energy available to the Sun and, there-
fore, the length of time it might have existed, and might
continue to exist, depended on the nature of the nuclear
‘reactions going on within it. But how were scientists to
determine the nature of these reactions? Offhand, it might
seem impossible to solve such a problem unless one could
first determine the nature of the substances making up the
 128 The Universe
Sun’s structure and the conditions under which they ex-
isted, and then try to work out the type of nuclear
reactions such substances would undergo in such condi-
tions.
_ Surely, this is a formidable task. To begin with how
would one determine the constitution of the Sun from a
distance of 93,000,000 miles? In the early nineteenth
century, it might have seemed ridiculous even to dream of
such a thing. Indeed, the French philosopher Auguste
Comte (1798-1857) had considered what the absolute
limits of human knowledge might be and, as an example
of something that must forever remain unknown and
unknowable, he had listed, among other items, the ques-.
tion of the chemical constitution of the heavenly bodies.
Yet not everything about the Sun is 93,000,000 miles
away. Its radiation reaches across space and touches us.
As the nineteenth century progressed, scientists learned
’ how to squeeze more and more information out of such
radiation. (The question of radial velocity, for instance,
was answered by studying the radiation of stars.) Let us
return, then, to the spectrum and to spectral lines.
In 1859, the German physicist Gustav Robert Kirchhoff
(1824-1887) and his collaborator, the German chemist
Robert Wilhelm Bunsen (1811-1899), began a careful
study of the spectra produced by various vapors when
heated in the virtually colorless flame produced by a
“Bunsen burner” (a device, popularized by Bunsen, that
mixed air and gas to promote more efficient burning and a
hotter flame). The heated vapors produced an emission
spectrum, bright lines against a dark background. Further-
more, the nature of these bright lines depended on the
elements present in the vapor. Each element produced its
- own pattern of bright lines, and the same line in precisely
the same position was never produced by two different
elements. The emission spectrum served as a sort of
- fingerprint of the elements present in the glowing vapor,
and thus Kirchhoff and Bunsen founded the technique of -
“spectroscopy.”
The next year, in the course of their studies of spectra
produced by various minerals, Kirchhoff and Bunsen de-
tected lines that were not produced by any known ele-
ment. They suspected the presence of new and hitherto
undiscovered elements and were quickly able to verify the
fact by chemical analysis. The new elements were named
“cesium” and “rubidium” from Latin words meaning “sky
blue” and “red” respectively, signifying the colors of the
 The Energy of the Sun 129
lines that led to the discovery. Cesium and rubidium were
the first elements to be discovered spectroscopically, but
were by no means the last.
Kirchhoff and Bunsen did more than this. They worked
with light from a glowing solid (which produced white
light that formed a continuous spectrum) and passed that
light through cool vapor. They found that the vapor
absorbed certain wavelengths of light and that the spec-
trum that was formed after the light had passed through
the vapor was no longer completely continuous, but was
crossed by dark lines which marked the position of the
absorbed wavelengths. This was an “absorption spec-
trum,” and it seemed clear at once that the Solar spec-
trum was an example of this. The hot body of the Sun
produces white light in a continuous spectrum, and when
this light passed through the Sun’s atmosphere (which was
hot enough but which was nevertheless cooler than the
body itself), some wavelengths were absorbed; this was
the reason for the dark lines in the Solar spectrum.
Kirchhoff noted that the wavelengths of light absorbed
by a cool vapor were exactly the same as the wavelengths
of light emitted by that same vapor when hot and glow-
ing. Suppose, for instance, that vapor of the element
sodium was heated until it glowed. The light produced
would be a deep yellow; if that light were passed through
a slit and then a prism, a closely spaced pair of yellow
lines would appear and make up the total emission spec-
trum for sodium.
If, however, white light from a carbon arc were passed
through relatively cool sodium vapor, the continuous spec-
trum ordinarily produced by the arc light would be broken
by a pair of closely spaced dark lines in the yellow. The
dark lines produced by absorption by the cool sodium
vapor would be precisely in the position of the bright lines
produced by the glowing sodium vapor. The dark lines of
an absorption spectrum could serve as the identifying
mark of an element as well as the bright lines in an
emission spectrum could.
But then what about the Solar spectrum and the ab-
sorption lines present there? One of the most prominent
lines in that spectrum (one which had been tabbed “D” by |
Fraunhofer) was indeed in the position of the sodium
lines. Kirchhoff checked this by passing sunlight through
hot sodium vapor and finding that the D line grew deeper
and more prominent. Furthermore, by passing sunlight
through hotly glowing sodium vapor, he could supply the
 Sia ie see toes Pe EL RES ae enh! Dene ae ee ag.
2 ee
430 Picea The Universe
; ' +

sodium line and wipe out the dark D line in the Solar
spectrum.
If the lines produced in the laboratory were identical
with those produced in the Sun, it seemed reasonable to
suppose that those in the Sun were also produced by
sodium and that sodium vapor was present in the Solar
atmosphere. Similarly, the dark lines H and K were shown
to be produced by calcium and that must be present,
therefore, in the Solar atmosphere. In 1862, the Swedish
astronomer Anders Jonas Angstrom (1814-1874) showed
that hydrogen was present in the Sun. Comte’s dictum
proved completely wrong; it was indeed possible to work
out the chemical constitution of the Sun and, in fact, of
any heavenly object that gave off light of its own with
sufficient intensity to produce a detectable spectrum.
At first the Solar spectrum was used only as a means of
determining which elements were present in the Sun and
which were not. But the question “How much?” arose.
Spectral lines deepened and broadened with increasing
concentration of a particular element in the glowing or
absorbing vapor. It began to be possible to determine not
only whether an element were present but in what quanti-
ties it might be present.
Finally, in 1929 the American astronomer Henry Norris
Russell (1877-1957) carefully studied Solar spectra and
was able to show that the Sun was astonishingly rich in
hydrogen. He. decided that the hydrogen content of the
Sun made up fully three-fifths of its volume. This was
totally unexpected since hydrogen, while not exactly rare,
makes up but a small portion of the Earth’s crust, only
0.14 percent, in fact.
Yet later work showed that Russell was overconserva-
tive. The recent estimates of the American astronomer
Donald Howard Menzel (1901- ) show that 81.76
percent of the Sun’s volume is hydrogen and 18.17 per-
cent is helium. That leaves only 0.07 percent for all other
- atoms. ’
It seems safe to say, then, that the Sun is essentially a
glowing mixture of hydrogen and helium, in a ratio of
about 4 to 1 in terms of volume. (The element helium was
_ another one discovered spectroscopically—and in the Sun,
rather than on Earth. The English astronomer Joseph
Norman Lockyer [1836-1920] suggested that certain
unknown lines in the Solar spectrum might be produced
by an as-yet-undiscovered element which he named after
 ee ae The Energy of the Sun 131
Helios, the Greek god of the Sun. It was not until 1895 |
that helium was located on the Earth by the Scottish
chemist William Ramsay [1852-1916].)

Surface Temperature of the Sun

Knowledge concerning the constitution of the Sun dras-

tically reduced the number of nuclear reactions that might
serve as conceivable sources for the Sun’s vdst production
of energy. It was simply out of the question to suppose
that the main source could arise out of nuclear reactions
in which any substance other than hydrogen, or possibly
helium, might serve as the fuel. No other substance was
present in large enough quantities.
Suppose, then, we consider the atomic nuclei of hydro-
gen and helium. The nucleus of the most common type of
hydrogen atom consists of a single particle, a proton, and
this type of hydrogen atom is therefore called “hydrogen-
1.” The nucleus of the most common type of helium atom
is made up of four particles, two protons and two neu-
_ trons, and the atom is therefore referred to as “helium-
4.”

It is conceivable that four hydrogen nuclei might fuse

(“hydrogen fusion”) to form a single helium nucleus, a
process which we might represent as: 4H1 — He*. Without
going into the details of how this might come to be,
either directly or through a long series of reactions involy-
ing other atoms, let us ask only whether such hydrogen
fusion would suffice to supply the Sun with the necessary
energy. :
The mass of a hydrogen nucleus has been determined
with great accuracy in “atomic mass units.” In these units,
the mass of the hydrogen nucleus is 1.00797; four such
nuclei would have a mass of 4.03188. The mass of the
helium nucleus, however, is only 4.0026. If 4.03188 atom-
ic mass units of hydrogen are somehow fused into 4.0026
atomic mass units of helium, then 0.0293 atomic mass
units (0.73 percent of the whole) must be converted into
energy.
The Sun’s loss of 4,600,000 tons of mass each second
(see page 121) could, in that case, represent a mass loss
resulting from the conversion of hydrogen into helium.
Hydrogen would be the Sun’s nuclear fuel, helium its
nuclear “ash.” Since the mass loss as the result of the
conversion of hydrogen into helium is 0.73 percent of the
mass of the fusing hydrogen, the loss of 4,600,000 tons of
 132 ‘The Universe
mass each second means that each second 630,000,000
tons of hydrogen are being converted into helium.
This fact makes it possible to guess at what the age of
the Sun might be. The total mass of the Sun can be
calculated from the strength of its gravitational pull on the
Earth across a distance of 93,000,000 miles; this turns out
to be 2,200,000,000,000,000,000,000,000,000 tons. Each
second 630,000,000 tons of hydrogen are being consumed;
and if we assume that the Sun was pure hydrogen to start
with, that this hydrogen has been fusing at a constant rate
_ since the beginning, and the material of the Sun is always”
well mixed, we can calculate how many seconds it would
take before the quantity of hydrogen sinks from 100
percent to 81.76 percent. It turns out that this would take
20,000,000,000 years—that is, 20 eons. Furthermore, it
would take an additional 90 eons before the total hydro-
gen fuel that remains is consumed.
- It is, of course, unsafe to assume that the rate of fusion ~
will continue unchanged to the end of the fuel supply or
that it was always the same as it is now. Certainly, the
presence of different quantities of helium ash might be
expected to influence the rate and even, perhaps, the
nature of the reaction. Nevertheless, as the 1930’s
_ dawned, it seemed at least possible that a plausible scheme
might be drawn up that would account for a Sun with a
total lifetime of at least 100 eons. The Solar system
‘clearly had a vast stretch of past history, and an even
vaster stretch still lay ahead of it.
: Merely conjecturing that the Sun’s energy supply was
based on the fusion of hydrogen to helium was insuffi-
cient. It was also necessary to show that conditions on the
Sun were such that hydrogen would indeed fuse. There are
vast supplies of hydrogen here in the Earth’s ocean, for
instance, and. yet it does not undergo fusion. If it did, the
Earth would explode and vaporize, changing into a very
small and very temporary star. If the fusion proceeded in
a controlled fashion, however, man might be able to
supply his energy needs for millions of years to come.
However, conditions on Earth are not such that spontane-
ous hydrogen fusion is possible, nor can scientists—so
far—create the necessary conditions for controlled fusion.
The best they have been able to do is to force some
hydrogen to undergo uncontrolled fusion, thus producing
the “hydrogen bomb” of the 1950’s.
What, then, about conditions on the Sun?
All we can see of the Sun-is its surface, and that
 The Energy of the Sun 133 ©
surface, it is obvious, is hot. But just how hot? Again we
turn to its radiation. ;
Even an object that is quite cool, such as the human
body, is constantly radiating energy. The body’s warmth
can be felt from a small distance. But this radiation is
very long-wave in nature and is in the far infrared and
therefore completely undetectable by the eye.
If an object, such as a flatiron, is slowly heated, the
radiation it gives off becomes more copious and begins to .
spread into the range of shorter and shorter wavelengths.
Photographic film capable of responding to infrared radia-
tion can take a picture of a hot flatiron in a dark room by
the flatiron’s own “light” even though that light is still too
long-wave to be detectable by eye. Eventually, if the
flatiron were heated still further, some of the radiation
would appear in wavelengths so short as to be detectable —
to the eye. They would be made up of the longest visible
wavelengths to begin with, and so the object would glow a
deep red. Further heating would add more and more
radiation of shorter and shorter wavelength, and the color
would change correspondingly.
In 1893, the German physicist Wilhelm Wien (1864-
1928) studied the matter in detail. At any given tempera-
ture there was some radiation peak, some wavelength at
which more radiation was emitted than at any other. Wien
found that as the temperature increased, the position of
this peak shifted in the short-wave direction according to
a simple mathematical rule. If, then, the spectrum of any
glowing object is studied and if the radiation peak in that
spectrum can be determined, the temperature of the ob-
ject can be deduced. The nature of the spectral lines also
changes with temperature, and they, too, will serve as
evidence in this respect.
Working with the Solar spectrum, it could then be
shown that the temperature of the Sun’s surface is 6000°
C. (or 10,000° F.). The surface temperatures of other
- stars can also be determined in similar fashion, and some
turn out to be hotter than the Sun. The surface tempera-
ture of Sirius is 11,000° C., for instance, and that of
Alpha Crucis (the brightest star of the Southern Cross) is
21,000°.C.
In earthly terms, the Sun’s surface is very hot. It is hot
enough to melt and vaporize all known substances, Never-
theless, it is not hot enough—it is very far from hot
enough—to force hydrogen to undergo fusion into helium.
We can say with perfect safety that no such fusion takes
 Ne as a 3 mg eo er fea Bie, Ls ata

134 ; “The Universe ~ 3 ee

_ place anywhere on the Sun’s surface, yet it must take place
somewhere if the Sun’s energy output is to be explained.
That leaves us, then, with the question of what might be
going on in the Sun’s interior.

Internal Temperature of the Sun

Determining the properties of the Sun’s surface was,

indeed, a formidable achievement and one that, on the
face of it, might have seemed impossible. How much more
difficult might one suspect an attempted study of the Sun’s
interior to be.
Nevertheless, some conclusions concerning the Sun’s
interior are easily arrived at. Thus, the Sun’s surface is
continually losing heat to outer space, and at a great rate,
_too; yet the temperature of the surface remains constant.
Clearly, the surface must be gaining heat from the interior
as fast as it loses it to space, and it follows that the
interior must be hotter than the surface.
Since the Sun’s surface is already hot enough to vapor-
ize all known substances and since the interior of the Sun
is hotter still, it seems reasonable to suppose that the Sun
is gaseous in nature—just a globe of superhot gas. If this
is so, then astronomers are fortunate, for the properties of
_ gases are easier to understand and work with than are the
properties of liquids and solids.
During the 1920’s, the problem of the structure of the
Sun’s interior was tackled by the English astronomer Ar-
thur Stanley Eddington (1882-1944), working on the as-
sumption that stars were gaseous bodies.
He reasoned that if the Sun were affected only by its
own gravitational field, it would collapse if it were merely
a globe of gas. The fact that it did not collapse meant
some force was countering that gravity, a force exerted
from within outward. Such an outward-directed force
might be that produced by the expansive tendencies of
gases at high temperature.
Taking into account the mass of the Sun and the
strength of its gravitational field, Eddington in 1926 calcu-
lated the temperatures required to balance the gravitation-
al force at the different depths beneath the Sun’s surface
and ended up with amazing figures. The temperature at
the Sun’s center reached the colossal figure of 15,000,-
000°C. (Some recent calculations, indeed, would put this
figure as high as 21,000,000°C.)
Despite the astonishing nature of this finding, it was—
 _ Pee Noe oe oe

The Energy of the Sun 135

accepted by astronomers generally. For one thing, such
temperatures were necessary if hydrogen fusion were to
be possible. While the Sun’s surface was far too cool for
any hydrogen fusion reaction to proceed, the Sun’s interi-
or, by Eddington’s reckoning, was emphatically hot
enough for the purpose.
Then, too, Eddington’s reasoning helped explain certain
other phenomena. The Sun was in a state of delicate
balance between gravitational force pulling inward and the
temperature effect pushing outward. What if some other
star were not?
Suppose, instead, that a particular star were not quite
hot enough to oppose gravitational compression. Such a
star would collapse inward, and this collapse would indeed
(as Helmholtz had long ago argued) convert gravitation
energy into heat. The interior temperatures would begin to
Tise and expansive forces would intensify and eventually
reach the point where they would just balance gravitation-
al pressure. Inertia, however, would keep the star contract-
ing for a while past that balance point—but more and
more slowly. By the time the contraction was brought to a
complete halt, the temperature would be considerably
higher than was needed merely to counter the gravitation-
al pressure, and the star would begin to balloon outward.
As it expanded, the temperature would drop and quickly
fall to a point of balance again. But here, too, inertia
would keep the star ballooning past that point until it
slowed and began to contract again. This cycle would.
continue over and over indefinitely.
Such a star would pulsate about some equilibrium
point, much like a swinging pendulum or bobbing spring.
Naturally the brightness of such stars would change regu-
larly with the pulsations and in such a way (taking into
account both size and temperature) that the light varia-
tion would just match the behavior of Cepheid varia-
bles.
With the temperature and pressure of the Sun’s interior
decided on and accepted, it remained to work out the
manner in which hydrogen would fuse to helium under such
circumstances at just the rate that would account for the
quantity of radiation being emitted by the § 4. In 1939, the
German-American physicist Hans Albrecht Bethe (1906-
) was able to work out a series of nuclear reac-
tions that just filled the bill. The rate at which they would
proceed under the conditions of the Sun’s interior (as calcu-
lated from data, both theoretical and observed, gained in
 The Universe _
the physics laboratories of the Earth) fit the conditions -
very well.
The question of the source of the Sun’s energy, posed
by Helmholtz in the 1840’s, was thus finally settled by
Bethe nearly a hundred years later. And with that, a
potential 100-eon lifetime for the Sun was also estab-
lished.
The gathering evidence in favor of a superhot interior
of the Sun had, however, an unexpected side effect. It
_ Tuined the planetesimal theory of the origin of the Solar
system.
It was all very well to suppose that matter could be
_ pulled out of the Sun and that this would then condense
and form planets—as long as the Solar material was
assumed to be at a temperature of merely thousands of
degrees. Temperatures in the millions of degrees were
another matter entirely.
In 1939, the American astronomer Lyman Spitzer, Jr.
(1914- ) advanced what seemed valid arguments for
concluding that such superhot matter would never con-
dense to form planets.but would spread out rapidly to
form a gaseous nebula about the Sun, and would stay
so.
Astronomers therefore had to return to the task of
attempting to visualize the formation of the planets out of
relatively cool matter. They had, once again, to think of a
contracting nebula in the old Laplacian fashion. However,
much had been learned in the twentieth centry about how
such a nebula might be expected to behave and about the
electrical and magnetic forces to which it would be PDO:
in addition to the gravitational ones.
In 1943, the German astronomer Carl Friedrich von
Weizsiicker (1912- ) suggested that the nebula out of
_ which the Solar system was formed did not revolve as a
unit. Instead turbulent patterns would be set up in the
outer reaches, smaller whirlpools within the overall larger
ones. Where adjacent whirlpools met each other, there
would be collisions of particles and coalescence into larger
and larger particles and, eventually, the gathering of plan-
ets. In this way, Weizsacker attempted to account for all
that Laplace tried to account for and, in addition, to
explain the spacing of the planets, the distribution of
angular momentum, and so on.
Weizsacker’s theory was greeted enthusiastically, but
there was considerable controversy over many of the
details. The controversy continues, and a number of as-
tronomers have advanced their own versions, no one of
which has as yet met general acceptance. The English
astronomer Fred Hoyle (1915- ), however, has re-.
cently suggested a mechanism involving the Sun’s magnet-
ic field that has proven impressively popular.
All agree, however, in postulating the formation of the
entire Solar system, both Sun and planets, by a single
process. In other words, if Earth, in its present shape, is
4.7 eons old, then we may conclude that the entire Solar
system (including the Sun), in the form it now exists, is
4.7 eons old. t
CHAPTER Q :

Types of Stars
oneaicen of the Solar System

The process of hydrogen faction to helium, while serving

as the main answer to the question of the Sun’s energy
production, does not remove all problems. For one thing,
the Sun turns out to be unexpectedly poor in hydrogen
and rich in helium. If it has been existing only 5 eons or
so, it ought to have expended less hydrogen than it has
and formed less helium.
It could be, one might suppose, that the Sun was hotter
in the past and spent its fuel with a more liberal hand.
That seems a natural thought, in fact, since one might
expect the Sun to behave like a wood fire, its flame sinking
as its fuel is consumed and “burning” progressively more
slowly. In that case, its past is shorter than we might
think, but its future is correspondingly longer. The chief
trouble with this suggestion is that from what geologists
can guess or deduce concerning the Earth’s past history,
there has been no significant change in the Sun’s radiation
output in the last few eons.
A second possibility is that the Sun was expending
hydrogen before the formation of the Solar system, while
it was still an extended, turbulent nebula.
This is also unlikely. The nebula might, conceivably,
have existed for an indefinite number of eons before the
formation of the Solar system in its present form, but it
would not be losing energy at the expense of nuclear
reactions while it was a nebula. In an extended nebula, the
gravitational field is so diffuse that it brings about very
little temperature increase toward the center, not enough
to reach the ignition point that would start the hydrogen
fusion going. Such a nebula would slowly contract and
138
 A yi? 5 ter vs) va
ee ere TPS af -

‘Types
ofStars «39
only the gravitational energy of particles falling inward
would be available for energy production—after the old
fashion postulated by Helmholtz.
As the nebula contracted, the gravitational field would
grow more intense; the total energy involved would re-
main the same but it would be concentrated into a smaller
and smaller volume. As the pressures at the center of the ~
contracting nebula increased, so would the temperature,
until finally, the ignition point would be reached. At this
point, the collapsing nebula would fire up and become a
star. It is only then that nuclear reactions would take
place, and only at the Sun’s center, not in the outskirts
- where the planets were taking shape.
We are left then, not only with the problem of the
too-abundant helium, but with a quantity of elements still
more complicated than helium that also exist in the Sun
and in the planets. Where did the other elements come
from?
Let us consider the various elements for a moment.
Hydrogen, with its atomic nucleus made up of a single
particle, and helium, with an atomic nucleus made up of
four particles, are the two simplest elements. The remain-
ing elements are all more complicated. The most common
elements (next to hydrogen and helium) are carbon, ni-
trogen, oxygen, and neon, and their atomic nuclei are
made up of twelve, fourteen, sixteen, and twenty particles,
respectively.
One might suppose, of course, that while hydrogen
fused mainly to helium, there might be certain side reac-
tions in which helium atoms would fuse further to carbon,
say, or to oxygen. These further fusion reactions must be |
rare indeed since they have sufficed to produce only very
small quantities of the more complex atoms even over the
nearly S-eon history of the Sun. Oxygen, for instance,
makes up only 0.03 percent of the volume of the Sun.
Then, too, if this were so, the elements beyond helium
would exist only within the Sun as the result of their
formation from nuclear fusion reactions. How could so
much of the more complex atoms find their way into the
planets which were formed out of matter at the outskirts
of the nebula?
Earth, for instance, is formed almost exclusively out of
elements mere complicated than hydrogen and helium.
This is not quite as surprising as it might sound at first;
there are some excuses that can be made here.
Solid substances cohere by interatomic attractions and
 140 | The Universe
do not depend on gravitational force to hold together.
Gases and vapors, however, have only very small interatom-
ic attractions, and it takes gravitational force-to hold
them to the body of a planet. The motions of the atoms or
atom-groupings (the latter are called “molecules”) in
gases and vapors tend to counteract gravitational pull. If
_ the atoms or molecules move quickly enough, they leave
' the vicinity of a planet despite its gravitational pull. The
smaller the planet and the weaker its gravity, the more
easily do atoms and molecules drift away. Then, too, the
lighter the atom or molecule, the more rapidly it tends to
’ move, and the more likely it is to drift away.
Hydrogen atoms are the lightest of all. They tend to
double up to form hydrogen molecules. Although the
hydrogen molecule has twice the mass of the single hydro-
gen atom, it is still lighter than any other atom.
Helium occurs as single atoms. The helium atom is
twice as massive as the hydrogen molecule (and four
times as massive as the hydrogen atom), but it is lighter
than any atom or molecule other than those of hydrogen.
The Earth’s gravitational field is not strong enough to
hold either hydrogen or helium. In the case of hydrogen,
there are alleviating factors. Two hydrogen atoms can join
with a single oxygen atom to form a water molecule
(eight times as massive as a hydrogen molecule), and can
join with other atoms to form molecules of solids. As a
result, the Earth in the course of its formation, did man-
age to retain some hydrogen in combination with other
elements, but its gravitational field was never large enough
to hold hydrogen in its gaseous form. As a result, most of
the hydrogen that must have been in the vicinity of the
Earth as it formed was never captured, and that is one
reason why the Earth is as small as it is. As for helium, it
forms no compounds at all, so not even a respectable
remnant of helium was captured. Earth today possesses
only tiny traces of helium.
And yet quantities of other elements (chiefly oxygen,
silicon, and iron) remained to form a body the size of the
Earth, and other bodies the size of Mars, Venus, Mercury, ~
and the Moon.
A planet like Jupiter, much farther from the Sun than
we are, may always have been at a much lower tempera-
ture. The lower the temperature, the more slowly do
atoms move and the easier they are to hold. The matter
gathering to form Jupiter could hold hydrogen more easily
than the matter gathering to form Earth. As more hydro-
DAE
oe EN
141
gen was collected, the mass of Jupiter rose and so did the
intensity of its gravitational field. It became still easier to
collect further hydrogen, which, in turn, heightened the
gravitational field still more. It was this snowball effect
that allowed Jupiter to grow so large and, as spectroscopic
and other evidence tells us, to be so rich in hydrogen (as
are, indeed, the other cold, outer worlds).
Nevertheless, even Jupiter is not all hydrogen. The
atmosphere has a large admixture of helium, and there is
also evidence of the presence of.compounds containing car-
bon and nitrogen.
So there were surprising quantities of helium and more
complex elements throughout the Nebula out of which the
planets were formed. The alternatives are as follows:
1. The presence of the heavy elements is impossible
except within the Solar interior, and therefore the planets
must have originated from matter within the Sun. This
would argue against any nebular theory of origin and
force astronomers back to some version of the planetesi-
mai theory.
2. The presence of the heavy elements within the
extended nebula is possible, and they must have been
formed by some method other than nuclear reactions
within the Solar interior.
Most astronomers are reluctant to accept the first alter-
native if they can somehow work out the second in accept-
able detail. To see where the heavy elements might come
from, if the Sun is left out of consideration, let us turn our
eyes beyond the Solar system and look to the stars once
more.

Spectral Classes

The earliest differences noted among the stars were in

connection with their position and brightness. In a few
cases, there were color differences, too. Antares was red,
Capella yellow, Sirius white, and Vega bluish-white. Such
colors were noticeable to the naked eye only among a
small handful of very bright stars.
The first half of the nineteenth century added an impor-
tant difference in distance. Some stars were relatively close
by (only a hundred trillion miles or so), while others were
enormously farther off. This meant that one could calcu-
late the actual brightness or luminosity of those stars
whose distance was known, and important differences in
luminosity were noted.
 ath : CS Se
t - . “ 4 ;
142 _ The Universe
Once spectroscopy came into use in the second half of
the nineteenth century, it was natural to wonder whether
the different stars might produce different types of spec-
tra. The Italian astronomer Pietro Angelo Secchi (1818-
1878) studied the spectra that were available to him and
in 1867 suggested that they might be divided into four
classes. The Solar spectrum fell into his second class, one
that was characterized by the presence of numerous ab-
sorption lines of metals such as iron.
Later astronomers confirmed the existence of these
“spectral classes” and refined them, introducing more deli-
cate divisions. By 1900, the American astronomer Edward
Charles Pickering (1846-1919) was characterizing the
classes by letters of the alphabet. The Sun was in spectral
class G, for instance. It was eventually possible to classify
the spectra within each class by numbers from 0 to 9, and
the Solar spectrum was classified as G2.
_ The spectral classes were not widely different. Rather,
one faded into another to form a sort of continuum. It
seemed probable from that, that whatever difference in
properties gave rise to the different spectral classes, it was
a continuous one, and that the property that changed did
so smoothly.
The question is: What property change accounts for the
difference in spectra?
Kirchhoff and Bunsen had shown that each element pro-
duced its own characteristic spectrum. Therefore, if spec- ~
tra of two stars differed, should that not be because the
two stars consisted of different sets of elements? This was
not an attractive thought. While it was possible that one star
might possess elements another did not, that did not fit in
with the growing conception of all objects in the Universe
being made up of the same (rather limited in number)
elements.
Was it possible, instead, for spectra to undergo alter-
ations without any essential change in elemental makeup
of the body yielding the spectrum?
One method of producing this effect was to alter the
temperature. With changing temperature, the electrons
surrounding the atomic nuclei shift from one “energy
state” to another. As the temperature rises, an electron
will shift from a lower energy state to a higher one and
absorb a certain wavelength of light as it does so. It may
later shift back from the state of higher energy to one of
lower, and light of that same wavelength is then emitted.
Because electrons can shift from state to state in a num-
 NealePUNetTG
ERE as :Pe er ee OEMieeen eee apt PondSay By ie
,
Types of. Stars c 143
ber of ways, a particular type of atom will emit or absorb
a number of different wavelengths, forming a spectral
pattern of bright lines or dark ones, the patterns being the
same in either case.
Every element has atoms containing a distinctive num-
ber and arrangement of electrons. The electrons in each
different kind of atom have their own distinctive spectral
pattern therefore, not shared by any other kind of atom
with a different number and arrangement of electrons. It
is for this reason that spectral lines, either bright or dark,
can be used to identify elements.
The hydrogen atom contains but a single electron, and
its spectral pattern is comparatively simple since there is
only so much a single electron can.do. As atoms grow
more and more complicated, with more and more elec-
trons in the atom, the pattern grows more and more
complicated, too. Sométimes the spectral pattern does not
appear to be as complicated as one might expect because
most of. the lines are outside the visible range. The iron
atom, however, containing twenty-six electrons, produces
thousands of lines in the visible range. It is iron that is
responsible for the major portion of the complexity of the
Solar spectrum in the visible range.
If one continues to heat a substance and forces the
electrons of the atoms making it up to enter states of
higher and higher energy, there eventually comes a point
at which some of the electrons become sufficiently ener-
getic to break the hold of the central nucleus and leave
the atom. As temperatures increase, one electron after
another will leave the atom.
An atom with fewer electrons than its normal comple-
ment (or, for that matter, more than its normal comple-
ment) is called an “ion.” The loss of electrons is therefore
referred to as an “ionization.”
An ionized atom will produce a different spectral pat-
tern from that produced by a normal atom. With one or
more electrons gone, those that remain shift among energy
levels in a somewhat different manner. Furthermore, an
atom with one electron missing will not give the same
pattern as will that same atom with two electrons missing, Or
three.
Different kinds of atoms hold on to their electrons with
different strengths. A temperature that suffices to ionize a
sodium atom, let us say, is completely insufficient to ionize
an oxygen atom. Again, it always takes a higher tempera-
ture to drive off a second electron from an atom than it
 144 The Universe
did to drive off the first one, and a still higher temperature
is required to drive off the third electron, and so on.
In short, differences in spectra may reflect rot differ-
ences in elements, but differences in the ionization state of
the elements. This, in turn, will reflect a difference in
temperature.
Before this was understood, strange lines were some-
times located in spectra and attributed to unknown ele-
ments. This worked out well in the case of helium (see
page 130), but in no other instance. Thus, the element
“coronium” was reported in the Sun’s outer atmosphere—
a region called the “corona” and visible only during -a
total eclipse. Similarly, an element called “nebulium” was
reported as present in certain nebulae.
In 1927, however, the American astronomer Ira
Sprague Bowen (1898- ) showed that the lines at-
tributed to nebulium were actually produced by a mixture
of the long-known elements, oxygen and nitrogen, where
_ the atoms of each had lost a couple of electrons under
conditions that required very low densities. Then, in 1941,
the Swedish astronomer Bengt Edlen showed that coroni-
um was actually a mixture of iron and nickel atoms that
had lost about a dozen electrons apiece.
Once spectra came to be interpreted in the light of
ionization, it became possible to tell from the pattern of
the lines alone, the temperature of the surface of the star
being studied. The difference between the spectral classes
was taken to be that’ between stars at different tempera-
tures and to only a very minor extent that of stars made
up of different elements. Indeed, the constitution of the
vast majority of stars is remarkably uniform. Like the
Sun, most stars are made up largely of hydrogen and
helium.
If the spectral classes are arranged in order of decreas-
ing temperature, the letter designations read: O, B, A, F,
G, K, and M. There are four additional classes, rather
specialized ones, which are labelled R, N, S, and W (the
first three including cool stars, the last hot ones).

Giant Stars and Dwarf Stars

Once one obtains two classes of knowledge concerning

a number of different stars, such as their luminosity and
their surface temperature, the next logical step is to put
the two together. For instance, on the basis of experience
with glowing objects on Earth, we would expect that the
 Types
of Stars 145
cooler a star was, the less radiation it would emit, and the
dimmer and redder it would be. This, however, proved not
always to be so.
-For instance, if the temperature-interpretation of spec-
tral classes is accepted, then the coolest of the ordinary
stars are those of spectral class M. From the spectral lines
they exhibited and. from the location of their radiation
peak, it was estimated that a typical surface temperature
for this spectral class was 2500° C. as compared with
our Sun’s 6000° C. The M-class stars were, indeed,
uniformly reddish in color, but they were not—despite
expectations—uniformly dim. Many of them were, indeed,
dim stars even though a few, like Barnard’s star, were
quite close to us. Others, however, like Betelgeuse in
Orion, or Antares in Scorpio, were red in color but very
bright in appearance just the same. Nor was this because
they were particularly close. They were not only bright in
appearance, they were very luminous in actual fact. An-
tares, for instance, emits some ten thousand times as much
radiation as the Sun.
As early as 1905, E. Hertzsprung had speculated on this
matter and had decided that the only way in which a cool
star could be bright was to be enormously large. Its .

coolness would mean that its surface gave little light per
square mile as compared with the Sun but, on the other
hand, there would be vastly more square miles on a star
like Betelgeuse than on the Sun. The number of. square
miles would more than make up for the comparative
dimness of each one individually. For this reason, stars
like Betelgeuse and Antares came to be called “red
giants,” while stars like Barnard’s star were “red dwarfs.”
It was particularly interesting that there seemed to be no
“in-between” red stars that were neither giants nor
dwarfs.
This suggestion by Hertzsprung, based on theoretical
reasoning, was shown to be correct by actual observation.
The German-American physicist Albert Abraham Michel-
son (1852-1931) invented a device called the “interferom-
eter” in 1881. By detecting slight variations in the man-
ner in which light waves interfered with each other, this
device was capable of making astonishingly refined meas-
urements. It made it possible, for instance, to learn things
about stars that no telescope could reveal.
Even the nearest stars are so distant that the best
modern telescope does not make them seem large enough
to be more than a point of light. Nevertheless, the rays of
 The Universe oe
light that reach the telescope from the star do not all come
from the same point on the star. One ray may come from
the star’s eastern edge, another from the star’s western
edge. The two reach the telescope at a small angle to each
other, an angle too small to measure by ordinary methods
but sometimes large enough to allow the rays to collide, so
to speak, and interfere with each other. Michelson’s in-
strument made it possible to measure the extent of this
interference and to determine the angle, if it were not too
tiny. If the angle is known, and the distance of the star as
‘well, then the actual diameter can quickly be deter-
mined.
The results were astonishing. The diameter of Betel-
' geuse was measured in this fashion in 1920 and found to
be about 300,000,000 miles. If this is compared with the
Sun’s diameter of 865,000 miles, it can be seen that
Betelgeuse is nearly 350 times as wide as the Sun. It
therefore was 350 X 350 or about 120,000 times the
surface area of the Sun. No wonder it is much more
luminous than the Sun even though it is dimmer by the
square mile. As for its volume, it is approximately 40,000,-
000 times that of the Sun. If Betelgeuse were put in place
of the Sun, its mighty diameter would cause it to fill out
all space to och the orbit of Mars. It is a red giant
indeed.
Antares is Roles aonatles than Betelgeuse, but the
latter star is by no means a record-holder. One star of this‘
- sort, Epsilon Aurigae, is so cool that, despite a monstrous
size, it is completely invisible to us. It radiates almost
entirely in the infrared. We know of its presence only
because it has a bright companion which is periodically
eclipsed by it. From the duration of the eclipse and the
distance of the system, it was suggested, in 1937, that the
ae
ae
dark star is an “infrared giant” that is 2,300,000,000 miles
in diameter. If it were inserted into the Solar system in
place of the Sun, it would fill space out to the orbit of
‘Uranus.
Nor are the infrared giants as rare as they seemed at
first. The trouble is that if a star is so cool that it radiates
mostly in the infrared, it becomes very hard to detect. In
the first place, the TEarth’s atmosphere is not very
transparent in the infrared region, and secondly everything
on Earth itself is warm enough to emit considerable in-
frared radiation of its own, so that infrared radiation from
the sky tends to be lost in the glare, so to speak.
In 1965, however, astronomers at Mount Wilson Observ-
 a Sreane rer erty te re) a ee hee
a Types of Stars 147
atory made use of special techniques with which to scan
the sky for spots rich in infrared radiation, spots that
would indicate the presence of infrared giants. They found
hundreds of such objects, mostly concentrated in the plane
of the Milky Way, and thousands may be within range of
detection. Some of them, at least, may be even more
voluminous than Epsilon Aurigae. They are bright indeed
in the infrared region but extremely dim in the visible
Tange, and few are visible in even the largest telescopes.
Two stars which have been detected are judged, from
their colors, to have temperatures of 1200° K and 800°
K, respectively—the second just barely red hot.
Stars of other colors do not show quite the gap in size
among themselves that the cool, red stars do. Still, there

Sizes of the Stars

Spectro- Measured Calculated
Star scopic Diameter* Diameter
Type Sun = 1 Sun = 1
e Aurigae B_ _____ K5 2000 -
VV Cephei A ___ M2 1200
BCR sce M6e 400 q
a Orionis _-__ M2 300 — 400% 400. (6
¢ Aurigae A ___ KS 300 g
a Scorpii . Mi 300 320 12
e Aurigae A ____ FS 200
A Peas, MS 110 130
@ faut. ee KS 36 ST
V 380 Cygni A ___ B2 29 q
a Bootis —__.___ KO 23 26 3
a Aurigae —_ (GO) 16
Y Cygni A, B___ o9 59
B Persei A B8 Buk
B-Aurigae A AO 2.8
Procyon A F5 1.7
a Centauri E G4 1.2 3
se GO 1.0 1.0 82
W Ursae 3 Z|
Majoris A ___ F8 09 ZA
70 Ophiuchi A ___ K1 0.9 ‘a
70 Ophiuchi B —__ K5 0.7 a
Kriiger 60 A M4 0.5
Sirius B _ — AS 0.02
40Eridani B= A 0.02 [2%
van Maanen 2 ___ G 0.006 =e
Wolf 457 ______ ee (a) 0.003
a eee
* Values in italics are derived from Pease’s interferometric measurements
modified to allow for the best recent parallax measurements. Other figures
are derived from the light curves of eclipsing biniaries.
t Variable diameter.
 Ole ea aeaera ee Noe ager eee
148 The Universe To ee
are large‘ ‘yellow giants” (not as large or cool:as the ne
ones) and small “yellow dwarfs” (not as small or cool as
the red ones). Capella could be described as a yellow
giant, our Sun as a yellow dwarf.

The H-R Diagram

While Hertzsprung was discovering the red giants, H.

N. Russell was doing similar work. He prepared a graph
in 1913 (independently of Hertzsprung, who had done the
same thing some years before) in which the horizontal
axis was made up of the spectral classes in order of
descending temperature, starting with spectral class 0 on
_ the left and ending with spectral class M on the right. The
_ vertical axis represented the luminosity or absolute magni-
tude (see page 53). Each star has some absolute magni-
fude and some spectral class and could be represented by
a dot, therefore, in a particular position on the graph.
Such a graph is the “Hertzsprung-Russell diagram” or,
more commonly, the “H-R diagram.”

H-R diagram

In general, the hotter the star, the brighter it-is. On the

graph then, the farther left a star was in spectral class
(and therefore higher in temperature) the higher up it
was in absolute magnitude. As a result, most of the stars
 we: PAS oe
Reinet Fae tS
tee
a| . \ . v. 3

: Types of Stars | 149

plotted by Russell fell into a diagonal line stretching from
the upper left to the lower right. This forms the “main
sequence.” It is estimated now that over 99 percent of the_
stars we can observe fall somewhere on this main sequence.
The most conspicuous exceptions to this rule are, of
course, the red giants. They are of spectral class M and
are therefore to the right of the diagram. They are also of
high luminosity, however, and are clustered in the upper
right of the H-R diagram, well out of contact with the
main sequence.
When the H-R diagram was first prepared, notions
about nuclear reactions within stars were still quite vague,
and most astronomers were still thinking of stars in terms
not too far removed from the views of Laplace and
Helmholtz. The feeling generally was that stars systemati- |
cally and steadily contracted in the course of their life-
time. From this standpoint it seemed that the H-R di-
agram offered a clear and dramatic picture of “stellar
evolution”; the manner, that is, in which stars came into
being, passed through various stages, and finally ceased to
radiate.
Russell’s suggestion as to the significance of the ER
diagram can be summarized as follows:
A star begins as a vastly voluminous conglomeration of
cool gas which slowly contracts. It warms as it contracts
and at an early stage radiates chiefly in the infrared so
that it is an infrared giant like Epsilon Aurigae. It con-
tracts further and grows hot enough to glow a bright red,
like Betelgeuse and Antares. It continues to shrink and
heat up, becoming a yellow giant, smaller and hotter than
the red giant, and then a “blue-white star,” still smaller
and still hotter.
A blue-white star of spectral class O is only moderately
larger than the Sun, but much hotter, with surface tem-
peratures up to 30,000° C., five times that of the Sun. At
this point the radiation peak is in the blue-violet region of
the visible spectrum, and beyond that in the ultraviolet-—
hence the color of the star.
In traveling from the initial cool nebula to the blue-
white stage, the star has been moving leftward across the
top of the H-R diagram. At the blue-white stage it reaches
the upper lefthand edge of the main sequence.
Now the star is pictured as continuing to contract under
the influence of gravity, but for some reason it no longer
grows hotter. One early suggestion was that by the time
the blue-white stage was reached, the material at the
 Se Sy aReet Saas te Na tea
$50 oO. he Univerto
center of the star was compressed so tightly that it no
longer acted as a gas. With further contraction, more and
more of the center would be compressed beyond the
gaseous stage and that might, for some reason, progres-
sively cut down the production of heat.
Therefore, the blue-white star shrinks and cools,
growing rapidly dimmer for both reasons. It becomes a
yellow dwarf like our Sun, then a red dwarf, like Bar-
nard’s star, and finally blinks out altogether and becomes a
“black dwarf,” a burnt-out cinder of a star.
In shrinking from a blue-white star to the final stage of
black dwarf, the star slides down the main sequence from
the upper left to the lower right. We might refer to this as
the “slide-theory” of stellar evolution.
This scheme was a most attractive one, and there
seemed to be a great many plausible things about it. First,
the picture of steady shrinkage with heating at first and
then cooling seemed in line with what one would expect to
happen “naturally.” As gas was compressed in the labora-
tory, it grew hotter; as hot objects were allowed to stand,
they grew cooler.
_ Then, too, if a star is a red giant at some early stage in
its career and a red dwarf at some late stage, one should
expect to find red dwarfs not very much different in mass,
on the average, from red giants. In other words, a red
giant is not enormous because it contains a great deal of
matter, but only because what matter it does contain is _
spread out over vast space. This proved to be so. The red
giants are by no means as enormously massive as one
might expect from their sheer size but were, instead,
enormously thin, The matter of a star like Epsilon Au-
_ rigae would (through most of its volume) be considered
merely a vacuum if a portion of it could be transported,
unchanged, into the laboratory.
In fact, stellar masses are surprisingly uniform. Whereas
stars vary widely in volume, density, temperature, and
other properties, they do not vary much in mass. Most
stars have masses that range from % to 5 times that of
the Sun. ;
1 The masses of individual stars are determined most easily when
those stars are part of a binary-star system. If the distance of the
system from ourselves is known, then astronomers can work out the
actual distance of the individual stars of the system from each other,
and determine, also, their orbital speed. From this, by use of New-
ton’s law of universal gravitation, the mass of the stars can be
determined. Fortunately, there are so many binary stars that the
data on stellar masses are large enough to allow general conclusions. a
ne eect peer

CHAPTER LO

~ Stellar Evolution

The Mass-Luminosity Relation

This interesting picture of stellar evolution as a slide

down the main sequence did not, however, survive long. It
was dead within a decade.
By the slide theory, the Sun ought to be in a late stage
of its evolution, long past its greatest, hottest days. It had,
according to the slide-theory, already cooled from a blue-
white star to a yellow dwarf, with red dwarfhood and
final extinction perhaps not too far ahead (on a cosmic
time-scale). Yet as it came to be understood that hydro-
gen was the most likely stellar fuel and that hydrogen was
present in the Sun in overwhelming quantities, it became
clear that the Sun must have a long lifetime ahead of it
and must be a relatively young star in terms of that
lifetime. Any evolutionary scheme that makes the Sun into.
an old star cannot be right.
Then, too, the question of stellar masses grew increasing-
ly important. It is true that there are no enormous
differences in mass between large, bright stars and small,
dim ones, but it is also true that the luminous stars are
somewhat more massive than the dim stars. This moderate
difference, which is quite a consistent one, must be ex-
plained.
The slide-theory of stellar evolution could manage it.
One could argue that a large star had a greater fuel supply
and therefore lasted longer. A small star would run
through its fuel quickly and reach the red dwarf stage,
while a large star might still be at the blue-white stage.
This explanation broke down, however, as a result of
Eddington’s researches on the structure of stellar interiors
(see page 134),
151
 € 9 ret A a al p

“
152 , The Universe E
Eddington argued that temperature from within had to
balance the compressing effect of gravitational forces. The
more massive a star, the greater the gravitational com-
pression and therefore the higher the countering tempera-
ture, the more luminous it must be. This relationship, first
advanced in 1924, is called the “mass-luminosity rela-
tion.””?
This means that one cannot picture all stars as starting
at the extreme upper right hand of the H-R diagram and
traveling leftward (compressing and growing hotter as
_ they do so) to the upper left-hand portion of the main
sequence at Class O. This could only hold true for particu-
larly massive stars. Less massive stars would not need to
attain Class O temperatures to support their structures
against gravitational compression. They would develop
lower temperatures and would reach the main sequence at
levels below Class O, A star no more massive-than our
Sun would reach the main sequence at its present posi-
tion.
Red dwarfs would be of still smaller mass to begin with
and would reach positions on the main sequence even
lower than the Sun. Particularly small bodies, less than
1/100 the mass of the Sun, may be incapable of raising
the central temperature to the point of igniting hydrogen
fusion at all. Such bodies would condense to a cold, solid
structure and become a black dwarf, not as the dead
cinder of a once-glowing star, but as a body that was
never a star at all. If such a black dwarf happened to
form in the neighborhood of a luminous star, it would be
a planet. How many black dwarfs there may be in the
Universe, which are so far from luminous stars that they
- cannot be detected, cannot be predicted. Shapley suspects
there may be many.
The exact boundary between a glowing star and a cold
black dwarf is not likely to be a sharp one. Some of the
larger black dwarfs may work up enough nuclear fusion at
the center to be gently warm at the surface.
1 The mass-luminosity relation explained the connection between
luminosity and period among the Cepheid variables. The more mas-
sive a Cepheid, the more luminous it is and also the more ponder-
ously and slowly it pulsates in and out. (It is a common experience
that in all periodic phenomena,a larger object has a longer period
.than a small one.) Therefore, a more luminous Cepheid has a longer
period than a less luminous Cepheid. The new yardstick of the
Universe, first brandished by Shapley, turns out to be a logical
consequence of differences in stellar properties and not merely a
mysteriously lucky break, *
 Srna Big eer EP tePee Pee ee
ee ee eis: Stier: Evolution — - 153
Even Jupiter, only 1/1000 the mass of the Sun, may
not be entirely dead at the center. In 1965, measurements
of Jupiter’s infrared radiation, made at Catalina Station
near Tucson, showed it to be radiating perhaps 2.5 times
as much heat as it receives from the Sun. Its surface,
according to calculations at Catalina, should be —170°
C. if the Sun were Jupiter’s only source of heat. Instead,
the surface temperature is —145° C., twenty-five degrees
higher. The extra heat may be derived from the heat of
compression at Jupiter’s center, and Jupiter may thus,
after a fashion, be considered a particularly tiny and
particularly cold star.
Let us, however, return now to ordinary stars that are
hot enough to glow brightly.

Approach to the main sequence

Note: M is mass whore that

ofthe Sun is taken as 1.0

peivcliacampeicnie:

The initial stage of stellar evolution, in which a loose

aggregation of dust and gas compresses itself and travels
toward some part of the main sequence, produces energy
chiefly from the gravitational field, the process Heimholtz
had pictured to be taking place throughout a star’s life-
time. The gravitational source is not, however, a very
large one in individual stars, and this stage passes com-
paratively quickly. In no time at all (cosmically speak-
ing), the star has reached the main sequence, where the

b
a
 g 7 SACP oeoa Sie
154 The Universe
central temperature reaches the point of igniting hydrogen
fusion, which then serves as the main source of energy.
This supply of energy is a vast and steady one. In most
stars, hydrogen fusion suffices to supply energy at a virtu-
ally constant rate over long periods of time. While this is
so, the star does not move on the main sequence to any
large extent. Any considerable movement, up or down,
would represent an upset of the delicate equilibrium be-
tween gravitation and temperature.
If our Sun, for instance, were to heat up, for some
reason, to the point where its surface temperature was
300,000° C., so that it had suddenly moved to the upper
left end of the main sequence, the outward pressure would
become so enormously much higher than the inward grav-
itational compression that it would explode. Only a star of
relatively enormous massiveness could possess the gravita-
tional field required to hold its structure together under
the explosive force of the outward pressure produced by
such high temperature. The Sun is not massive enough for
this purpose and, in all likelihood, it never was and never
. will be. The thought therefore that it was once a Class O
star and slid down the main sequence through B, A, and F
to its present position as a class G star is untenable. Its
present position on the main sequence is the only position,
thanks to its mass, that the Sun can comfortably oc-
cupy.
Stars remain on the main sequence (more or less mo-
tionlessly, in whatever position their mass dictates) for so
large a percentage of their total lifetime that more than
99 percent of the stars we can see are on it. In other
words, there is less than one chance in a hundred that, in
observing a particular star, we will catch it during the
relatively short interval before it has reached the main
sequence or after it has left it.
Still, some stars remain on the main sequence longer
than do others. A large star has a greater fuel supply than
a small star, but it must maintain itself at a higher tem-
perature and therefore must consume its fuel at a more
rapid rate.
From the mass-luminosity relationship, it can be shown
that the rate of fuel consumption increases much more
rapidly with mass than the fuel supply does. The larger
and hotter a star, therefore, the shorter a time its fuel will
last, and the shorter a time it will remain on the main
sequence.
Our Sun, for instance, of spectral class G, will stay on
 mang ee he Stellar Evolution
the main sequence hee a total of some 13 eons. Of these, 5
eons have passed and about 8 eons remain, showing that
the Sun is still rather less than middle-aged. A class F
star, a little hotter and larger than the Sun, may have
more hydrogen to begin with, but consumes it at a suffi-
ciently faster rate to allow it to be on main sequence
considerably less than 8 eons. In general, the larger and
hotter a star, the briefer its stay on the main sequence. If
the Sun were a class A star and had been on the main
sequence as long as it has, it would be getting ready to
move off it now.
The hottest stars of all burn their relatively tremendous
supply of fuel at so rapid a rate that their stay on the
main sequence is measured not in eons but in mere tens
and hundreds of millions of years. The most luminous
star, S Doradus (see page 92), can stay on the main
sequence only two or three million years.
If we consider the period before arrival at the main
sequence as being short enough to ignore, we can sum-
marize Eddington’s view by making the general rule that
the brighter a star the shorter-lived it must be. (This is
just the reverse of the conclusion one might arrive at from
the slide-theory.) The red dwarfs are not near extinction
but may continue shining as they do, stingily doling out
their small supply of hydrogen, for many eons after the
Sun has reached its end. On the other hand, the huge
bright stars are not at the beginning of their cycles at all
but are expending their fuel so prodigally that they will
reach their ends while our Sun is still plodding along
exactly as it is doing now.
Eddington’s mass-luminosity relation produced a rather
surprising by-product that also had its effect on views of
stellar evolution. He had worked out the relationship on
the assumption that stars possessed the properties of gases
throughout their structure (see page 134). At first, he
accepted the general view of the early 1920's that only the
red giants were gaseous throughout and that the stars of
the main sequence, and particularly the dwarf stars, had
nongaseous cores. He expected, therefore, that: his conclu-
sions would not hold for stars of the main sequence.
To his surprise, though, whenever his conclusions could
be tested by observation, they were found to hold for all
stars, dwarfs as well as giants. He had to conclude that ail
stars, the dwarfs included, were gaseous throughout—a
conclusion still firmly accepted today.
This result further damaged the original slide-theory of
 x Se are ale ae = +)a 4
+S ORR xa

156, - "The Universe

evolution, for it made it difficult to explain why a star
should slide down the main sequence, contracting and yet
cooling down. The gas laws made it necessary to suppose
that contraction would be accompanied by heating, not
cooling.
By the mid-1920’s, then, the slide-theory of stellar evo-
lution was dead, and what we might call the modern
theory was established.

Interstellar Gas

So short are the lifetimes of the brightest stars that they

could not have existed in their present form when the
dinosaurs ranged the Earth. On a cosmic time-scale they
are ephemeral.
If stars could have formed a few tens of millions of
years ago, this is as much as to say that there are stars
that can be forming now. Stars may, in fact, be coming
into existence continually; perhaps not at the rate that
stars were formed in the dim past when the Galaxy
generally may have been in the process of formation, but
still at some significant rate. If this is so, is it not possible
that we may actually see stars in the process of formation
right now?
It is hard to tell whether this is so because the process is —
so slow in terms of human lifetimes (even if rapid in
terms of stellar lifetimes) that over the short period of
detailed observations, results are not clear-cut. Further-
more, stars in the formation-stage are not easily visible.
There are objects in certain nebulae that might be taken
to represent stars in the process of formation. In the
Rosette Nebula there are many dark globules that might
conceivably be matter condensing on the way toward the
main sequence. Other suspected sites of present day star
formation include the Orion Nebula and the Nebula NGC
6611 in the constellation Serpens.
But out of what can the new stars be formed?
Astronomers generally agree that stars are, in the be-
ginning, vast clouds of gas and dust; eons ago when the
Galaxy was being formed, this raw material of the stars
must have been plentiful. The Galaxy itself must have
been nothing more than a huge mass of swirling matter
out of which eddies separated and condensed to form
stars. But now, with a hundred billion or more stars
already condensed out of the primordial swirl, how much
raw material can be left?
 gs
Stellar Evolution He ee ASF .
I have sirens mentioned the existence of interstellar
dust, collecting in places in sufficient quantity to block off
starlight as dark nebulae, or reflecting starlight to become
Juminous nebulae. There is also interstellar dust more
generally spread through interstellar space and scattering
and dimming starlight generally (see page 87). The effect
is important, but it does not take much dust to turn the
trick, and that dust alone is not present in sufficient
quantities to serve as a generous store for star-
formation.
More important is the existence of interstellar gas.
Individual atoms or molecules of gases do not absorb or
scatter light efficiently and therefore do not make them-
selves evident as clearly as dust would, even though the
gas might be present in far greater quantity.
Individual gas atoms will, however, absorb specific wave-
lengths of light, just as the atoms in the Solar atmosphere
will. The concentration of the gas in interstellar space
must be so thin that light absorption over ordinary dis-
tances can only be vanishingly small and quite immeasura-
ble. Over hundreds and thousands of light-years, however,
cumulative absorption would pile up to measurable leveis.
It might be possible, therefore, that some lines in stellar
spectra could not result from the gases immediately sur-
rounding the stars, but from the very thin gas spread out
all the way between the stars and ourselves.
The first indication of this came about through spectro-
scopic studies of the binary stars. Some binaries revolve
about their center of gravity in a plane that is placed
edgewise or nearly edgewise to us. When both stars are
luminous, the eclipse of one by another is not very effect-
ive in changing the quantity of light reaching us, and it is
difficult to detect such binaries if the separate bodies are
too close to be seen individually by telescope.
However, as the bodies revolve in a plane edgewise to
us, one of the pair will be receding from us, while the
other is advancing toward us. After a while, one will move
behind the other and both will be moving transverse to the
line of sight, one to the right and one to the left. Later
still, the one that had earlier been receding will now be
advancing toward us, while the other, which had been
advancing is now receding. They cross transversely again
and then the process begins again.
When the components of the binary are moving in such
a way that one is approaching and the other receding, the
spectral lines belonging to the first will shift toward the
 sO at a Raat ? on Ny =
158 ae The Universe
violet and those belonging to the other will shift ogent
the red. When the components are both traveling trans-
versely, there will be no shift in either direction for either
- star. If the spectra of the two stars are of the same class
then, during the transverse-motion stage, the spectral lines
of the two will coincide. During the approach-and-recede
stage, however, the lines will become double as one set
shifts in one direction and the second in the other. In the
course of the rotation, the spectral lines will double twice
each period.
Indeed, it is possible through the behavior of the spec-
tral lines to recognize a star as a binary even when all
optical evidence of the fact is lacking. In 1889, the Ameri-
can astronomer Antonia C. Maury (1866-1952) noted
such a periodic doubling of lines in the case of Mizar, one
of the stars in the handle of the Big Dipper. This was the
first star to be recognized as a “spectroscopic binary.”
Many more examples have been discovered since.

Spectroscopic binaries

In 1904, the German astronomer Johannes Franz Hart-

mann (1865-1936) was studying such a spectroscopic
binary, Delta Orionis. He noted that during the periodic
doubling of lines, one of the lines did not double. Some-
thing was absorbing a specific wavelength of light and was
not sharing the motion of either component of the double
star system. It might have been a very massive third
component of the system, a component so massive that
the center of gravity of the entire system would be close
to its own center and it would scarcely appear to move.
 ~ Stellar<Rvokition arn Crave 159 © ee

Yet if nich a massive component were luminous it should

be seen; and if it were not, it should indicate its presence
by eclipses, as in the case of Algol.
It seemed to Hartmann much more likely that the
motionless absorption line was produced by the cumulative
effect of excessively thin gas in the space between Delta
Orionis and ourselves. Hartmann’s conclusion was not
immediately accepted, but additional evidence was report-
ed by other astronomers, in. particular the Russian-
American astronomer Otto Struve (1897-1963). Inter-
stellar gas is now accepted as a feature of the Galaxy, and
its total mass is perhaps 50 to 100 times that of the dust in
the Galaxy.
The motionless spectral line first observed by Hartmann
was one that was produced by calcium atoms, so it
seemed obvious that the interstellar gas included calcium.
Other atoms were also detected, but the composition of
the gas could not be determined merely from the spectral _
lines. A gas that happened to absorb certain wavelengths
in the visible light region very strongly—as calcium does—
might impress its mark on the spectrum even though it -
was present in only minor quantities. By the 1950’s it
became quite plain that the preponderant component of
the interstellar gas was the much less obtrusive (spectro-
scopically speaking) hydrogen.
It is estimated today that 90 percent of all the atoms in
the Universe are hydrogen, the simplest of all the atoms,
and that 9 percent are helium, the next simplest. All other
types of atoms make up the remaining 1 percent. In short,
the elemental makeup of the Sun is thought to be rather
typical of the elemental makeup of the Universe gener-
ally.
If the interstellar gas is mainly hydrogen and helium, of
what does the dust consist? Helium has virtually no tend-
ency to clump together, and hydrogen forms two-atom
molecules which also have little tendency to clump togeth-
er. Dust must form, therefore, with the aid of some of the
minor constituents of interstellar matter, but not a constit-
uent that is too minor, for the dust content of the Galaxy
is substantial.
One suggestion focusses on oxygen, the most common
of the minor elements. An oxygen atom can easily com-
bine with a hydrogen atom to form a “hydroxyl group,”
and in 1963 these were actually detected in the interstellar
matter. An oxygen atom can also combine with two hy-
drogen atoms to form a water molecule, and water mole-
 160 Ss The Universe chy 2.
cules have a stony tendency to clump together. The inter-
stellar dust might, therefore, be made up of ice crystals in
large part.
The interstellar gas and dust, though thinly spread out,
occupy an immense volume, and amount, in total, to a
large mass. Some estimates have made the interstellar
matter of the Galaxy equal in mass to all the stars, but
this is almost certainly an overestimate. The most recent
determinations would make the mass of interstellar gas
only 2 percent that of the stars, although the spiral arms
would be considerably more gassy than the Galactic nucle-
us. In the spiral arms the mass of interstellar material
' might be as high as 10 to 15 percent that of the stars.
Even at the lowest estimate, there would be enough
interstellar matter in a galaxy like ours to make up two or
three billion stars if it were all collected, so it is no
surprise that some stars may be forming even today out of
this largely hydrogenous interstellar matter, or that some
stars formed one to ten million years ago and shine today
with supernal brightness.
Some other galaxies may serve as even richer sources of
new star material than ours does. The Large Magellanic-
Cloud, for instance, has perhaps three times the concen-
tration of interstellar gas that the Galaxy does. ~
An explanation can now be offered for the unexpected
hydrogen deficit in the Sun and for its apparently too-
large supply of helium, as well as for the fact that a planet
such as the Earth is made up almost entirely of elements
more complicated than helium. Apparently the gaseous
material out of which the Solar system formed already
contained a considerable supply of helium and a small
quantity of more complex atoms as well.
The question then is: Where did the helium and the
more complex atoms in the interstellar gas come from?
We might suppose that the gas out of which the Galaxy
was formed simply contained a quantity of helium and
more complicated atoms to begin with. It is much more
tempting, however, to suppose that only hydrogen, the
simplest of all atoms, was present to begin with and that
all other atoms formed from hydrogen. Yet the only
processes by which hydrogen will fuse to form other
atoms, as far as we know, require the conditions present
in the cores of stars. In that case, how did the helium and
other atoms get back into the gas?
Let us keep that question in mind as we continue to
consider the possible course of stellar evolution.
The Moon. (Photograph from the Mount Wilson and Palomar
Observatories.)

Head of Halley’s Comet, May 8, 1910. (Photograph from the

Mount Wilson and Palomar Observatories.)
Saturn and Ring
System. (Photo-
graph from the
Mount Wilson and
Palomar Observa-
tories.)

The Milky Way in

Sagittarius. (Photo-
graph from the
Mount Wilson and
Palomar Observa-
tories.)
The Pleiades. (Photograph from the Mount Wilson and Palo-
mar Observatories.)

The Great Hercules Cluster. (Photograph from the Mount

Wilson and Palomar Observatories.)
 The Large Magel-
lanic Cloud. (Lick
Observatory Photo-
graph.)

Dark Nebulae. (Photograph from the Mount Wilson and

Palomar Observatories.)
Luminous Nebula.
(Lick Observatory
Photograph.)

The Horsehead Nebula. (Photograph from the Mount Wilson

and Palomar Observatories.)
The Andromeda Nebula. (Lick Observatory Photograph.)
Resolution of An-
dromeda Nebula.
(Photograph from
the Mount Wilson
and Palomar Ob-
servatories.)

Sombrero Galaxy.
(Photograph from
the Mount Wilson
and Palomar Ob-
servatories.)
Whirlpool Galaxy. (Photograph from the Mount Wilson and
Palomar Observatories.)
Spheroidal Galaxy.
(Photograph from
the Mount Wilson
and Palomar Ob-
servatories.)

Nebula About No-

va Persei. (Photo-
graph from the
Mount Wilson and
Palomar Observa-
tories.)
The Crab Nebula
—photographed
at various wave-
lengths. (Photo-
graph from the
Mount Wilson
and Palomar
Observatories.)
Planetary Nebulae. (Photograph from the Mount Wilson and
Palomar Observatories.)

Cluster of Galaxies. (Photograph from the Mount Wilson and

Palomar Observatories.)
Solar Flare. (Photo-
graph from _ the
Mount Wilson and
Palomar Observa-
tories.)

Sun’s Corona.
(Lick Observatory
Photograph.)
Polarized Light in
the Crab Nebula.
(Photograph from
the Mount Wilson
and Palomar Ob-
servatories.)
Gaseous Nebula IC 443. (Photograph from the Mount Wil-
son and Palomar Observatories.)
NGC 5128. (Photograph from the Mount Wilson and Palo-
mar Observatories.)
Cygnus A. (Pho-
tograph from
the Mount Wil-
son and Palomar
Observatories.)

Galaxy M87.
(Lick Observa-
tory Photograph.)
 gee eeeearn ei e r = ee

on Stellar Evolution 161

evade the Main Sequence ~

What happens toa starmaser it consumes so ‘much fuel

that it can no longer maintain the balance between gravi-
tation and temperature and can, therefore, no longer
remain on the main sequence?
The answer to this arises partly out of theory and partly
out of observation, Nuclear physicists can work out elabo-
rate theories of what might happen in the interior of stars
under certain conditions of temperature, pressure, and
chemical composition, and this can be checked against
what we observe in the heavens.
Astronomers can, for instance, observe those star-
clusters, such as the Pleiades, which are close enough to us
so that the individual members can be studied spectroscop-
ically. The stars in such a cluster are all at about the
same distance from us, so that their order of apparent
brightness is identical with their order of luminosity. That
gives us the vertical axis of the H-R diagram. If the
individual spectra are studied, we have the horizontal axis
also. We can, in short, make a small and special H-R
diagram of the cluster stars only.
' Furthermore, we can assume that all the stars in such a
cluster are the same chronological age. It seems reason-
able to suppose that a large volume of gas spreading out
over the entire volume of the cluster at one time in the
distant past condensed into the various stars making up
the cluster over so short a time period, cosmically speak-
ing, that all the stars of the cluster can be considered as
having been born at once. If this is not so and if the stars
of the cluster came into being independently, then it is:
hard to explain their close association today.
If chronological age were the only criterion of stellar
evolution, then all the stars in a cluster, being of the same
age, ought to be at the same stage of evolution. All would
be at some one point on the main sequence, or at some
one point before reaching it, or at some one point after
leaving it.
Chronological age is not, however, the only criterion.
Mass is another. The stars in a cluster vary over a range
of mass, and the more massive a star the more rapidly it
develops. A massive star born at a particular time would
have reached a later stage in evolution than a less massive
star born at the same period. In a cluster of stars, then,
the least massive stars would be at the earliest stage of the
evolutionary cycle, and successively more massive stars —
 RT Were ee a tiie | nee oes oe RP eR Ee
mip ‘ es ;

162 * The Universe

would be at later and later stages. If each star is tees at
its approximate point in the H-R diagram, the entire
course of evolution is sketched out, up to the point of
farthest evolution in the cluster. :
Trumpler (the man who demonstrated the existence of
interstellar dust) carried on such systematic studies of
clusters in 1925, and similar observations in great detail
have been made ever since. Combining these observations
with theory, astronomers now feel confident. they can
describe the evolutionary adventures of an individual
star.
1. A mass of gas and dust contracts and heats until it
reaches the main sequence. This will take perhaps 100,000
years for a large mass of gas which will end at the hot
upper end of the main sequence because the relatively
large gravitational field of a large mass will cause it to
contract relatively quickly. A star the mass of our Sun
will contract to main sequence in perhaps 2 million
years.
2. Stars will then stay on the main sequence anywhere
- from millions of years to tens of eons, depending on their
_ mass. While a star is on the main sequence, nuclear
reactions are proceeding in the high temperature core,
which gradually loses hydrogen and accumulates. helium.
' Thus, calculations reported in 1966 indicate that the Sun’s
core contains six helium atoms for every hydrogen atom,
while its outskirts see-the ratio just about reversed, seven
hydrogen atoms for every helium atom.
-_ 3. When the core reaches some critical content of little
hydrogen and much helium, the star begins to expand and,
consequently, cool. It then leaves the main sequence and,
on the H-R diagram, can be pictured as moving upward
and to the right. In some cases, it is ‘thought, the star
reaches .a Cepheid stage, pulsating regularly for some
millions of years. In other cases, the expansion continues
more or less smoothly until the star is enormous and its
matter, in the outer layers at least, very rarefied. It has
become a red giant which thus represents a late stage in
stellar evolution (rather than an early one as was sug-
gested in the slide-theory of stellar evolution).
Our Sun, someday, but not for many eons, will begin to
expand its way to the red giant stage, and if mankind has
not left the Earth by then (or put an end to itself long
before) that will be our finish.
Naturally, the larger and more massive a star, the more
tremendous a red giant it will balloon into. The red giant
BA ee ee oa See ae a eee r
Stellar Evolution ' 163
into which our Sun will someday bloat will not be a
particularly impressive specimen of the class. Red giants
such as Betelgeuse and Antares developed out of main
sequence stars considerably more massive than the Sun.
The most massive stars may have grown still larger and
cooler to form the infrared giants.
' 4, By the time the red giant stage has been fully
reached, the hydrogen in the core is entirely consumed.
The core has been constantly increasing in mass as more
and more helium collects in it. While the outer layers of
the star have been expanding, the core itself is contracting
and growing hotter. Eventually, a temperature of about
140,000,000° C. (nearly ten times the temperature of the
Sun’s interior at present) is reached, and that is hot
enough to ignite a nuclear reaction in which three helium
atomic nuclei combine to form the atomic nucleus of a
carbon atom. Such “helium-burning stars” begin to con-
tract and heat up again.

Stellar evolution

red giant turning point

 164 ‘The Universe
The development of helium fusion does not mean that
the star has gained a new lease on life comparable to the —
early stage of hydrogen fusion. Helium fusion does not
' supply as much energy as hydrogen fusion does.
Suppose we consider twelve hydrogen-1 nuclei fusing to
three helium-4 nuclei, and then the three helium-4 nuclei-
fusing into a single carbon-12 nucleus. The twelve hydro- . |
gen-1 nuclei have a total mass of 12.0956; the three helium-4
nuclei a total mass of 12.0078; and the carbon-12 nucleus a
mass of 12.000. The mass-loss (and, therefore, energy
release) in the hydrogen-fusion stage is 0.0878, while that in
the helium-fusion stage is only 0.0078. Thus, mass for mass,
helium fusion will yield only. 9 percent as much energy as
hydrogen fusion.
The star can make do, however, by fusing carbon atoms
still further to form more complicated nuclei, but this
cannot continue forever, A dead end is reached’ with iron.
The nuclei of the iron atoms are the most stable of all.
Once the iron nucleus is reached, no more energy can be
obtained out of nuclear reactions involving it. Whether the |
iron nucleus is built up to more complicated ones or ~
- broken down to less complicated ones, energy is not ‘re-
leased; energy must be supplied.
And, as it happens, little energy can be obtained even if
nuclei are followed all the way up to the dead end, once
hydrogen fusion is over and done with. Imagine 56 hydro-
gen-1 nuclei converted to 14 helium-4 nuclei, and. that
these 14 helium-4 nuclei are then converted into a single
iron-56 nucleus.. The respective masses are 56.4463,
56.0264, and 55.9349. The mass-loss in passing from hy-
drogen to helium is 0.4199, and from helium to iron is
0.0915. The energy gained from the conversion of helium
all the way to iron is only 22 percent that of the conver-
sion of hydrogen to helium, weight for weight. We might
say, then, that when a star has consumed its hydrogen, its
life as a nuclear reactor is four-fifths over.
_ After the red giant stage, as a star shrinks and heats up
steadily, a core forms within the core and another core
within it; and so on. Each core as it forms contains more
complicated atoms than the one before, until finally an
end comes with an innermost core rich in iron.
This process of shrinking and heating may be pictured
on the H-R diagram as a rapid passing leftward and
downward. The star reaches and crosses the main se-
quence heading for the lower left region of the diagram, a
region that would contain hot stars of low luminosity.
 a

- Stellar Evolution - 165

The lower left region is, naturally, just the opposite of
the upper right region where cool stars of great luminosity
(the red giants) are to be found. Just as cool stars can be
highly luminous if they are very large, so hot stars can be
unusually dim if they are very small.
In order to understand what ‘happens to stars when
their nuclear reactions approach an end, we must consider
the nature of very hot, very small stars—a group I have
not yet discussed.
 CHAPTER ] ]

| Stellar Explosions
White Dwarfs

_ + Knowledge concerning the hot but dim stars began with

Bessel’s discovery of the companionof Sirius in 1844 (see
page 51). Sirius and its companion moved in elliptical
orbits about their mutual center of gravity, with a period
of revolution of fifty years. Sirius is sometimes referred to.
as “Sirius A,” therefore, in deference to its existence as
an of a binary system, while its smaller companion is
- “Sirius B.”
_ Any object capable of Gwingiag a star like Sirius in an
orbit large enough to be visible from Earth must itself
produce a star-sized gravitational field. In fact, it could be
’ calculated that Sirius’ companion had to. have a mass fully
half that of Sirius A and must therefore be about as
' massive as our Sun. Yet Sirius B could not be seen, so that
Bessel assumed it to be merely the burnt-out cinder of a star.
That same year, Bessel reported a similar dark compan-
ion for Procyon so that one could.speak of “Procyon A”
_ and “Procyon B.” It seemed very likely that such- dim
stars might. be quite common. It was thought that only
‘their lack of light prevented them from being seen; and
that they could be detected only when they were part of a
multiple star system and their gravitational fields could
alter the motions of a visible star to such an extent that
the results could be made out from Earth.
To be sure, neither companion turned out, in the end,
to be completely dark. In 1862, Clark detected Sirius B as
a star of magnitude 7.1. This is not terribly dim, of
course, but the Sirius star systemi is only 8.8 light-years
away. To be at that distance and still appear so. dim, Sirius
B had to have a luminosity only a hundredth that of our
166
Pg Eee Pree ey yO SEE Fa

Stellar Explosions 167

Sun, or less. In 1895, the German-American astronomer
John Martin Schaeberle (1853-1924) detected Procyon B
and that turned out to be only of the eleventh magnitude.
Even allowing for the fact that Procyon B was a bit
farther from us than Sirius B, it could quickly be shown
that Procyon B was even dimmer than Sirius B. Such stars
might not be completely dark, but they were certainly
dwarf stars.
It was taken rather for granted at the turn of the
century that stars like Sirius -B and Procyon B were dying
stars that were dim primarily because the stellar fires were
flickering out. One might suppose that such stars might fit
neatly among the red dwarfs at the tail of the main
sequence.
However, even as the H-R diagram was being worked
out, it became quite apparent that Sirius B, for instance,
would not fit in just that place. To be at the tail end of the
main sequence, a star would have to be very cool, and,
therefore, a deep red in color. Sirius B, however, was not
ted. It shone with a clear white light. If it was a dwarf, it
was a “white dwarf.”
In 1914, the American astronomer Walter Sydney
Adams (1876-1956) succeeded in taking the spectrum of
Sirius B and found it to be a spectral class A star, exactly|
as Sirius A itself was. This meant that Sirius B had to
have a surface temperature as high as that of Sirius A
(10,000° C.) and higher than the mere 6000° C. surface
temperature of the Sun.
But if Sirius B was hotter than the care it should have a
surface brighter than the Sun, square mile for square mile.
The fact that Sirius B was so much less luminous than the
Sun could only mean that Sirius B possessed very few
square. miles of surface area. It was a star that was
white-hot but very small, just the kind of star you would
expect to find in the lower left region of the H-R diagram,
the region mentioned at the end of the previous chapter.
In fact, it would have to be quite small. To account for
its dimness, Sirius B would have to have a diameter of not
more than 17,000 miles and be-no larger in size than the
planet Uranus. It would be a white dwarf indeed.
And yet Sirius B would still have the. mass of the Sun.
That was determined from the gravitational effect of Sirius
B on Sirius A, and it could not be argued away. For a
star, then, to be as small as Uranus and as massive as the
Sun brought up serious questions of density—questions
that would have raised insuperable difficulties in the nine-
=zse Poem cnnceoeee ter ee Mos OM en Sie et ons yt ge
F,

168 - “The Universe

teenth century but, as it turned out, could be answered in
the twentieth.
Indeed, the problem of stellar densities arose not only
about unusual stars such as Sirius B, but about our own
Sun.
Once the distance of the Sun was known, it was easy to
, calculate its real diameter, and therefore its volume, from
its apparent diameter. Its volume turned out to be 1,300,-
000 times that of the Earth. Its mass, calculated from its
gravitational effect on the Earth, turned out to be 333,500
times the mass of the Earth.
Density is determined by dividing the mass of an object
by its volume. The Sun’s density should therefore be
333,500.+ 1,300,000 or just a little over one-fourth that of
the Earth. The density of the Earth is 5.5 grams per cubic
centimeter and that of the Sun is. therefore 1.41 grams per
cubic centimeter. 3
The density of water is 1.00 grams per cubic centimeé-
ter, so the density of the Sun is 1.41 times that of water.
Moreover, this figure represents an overall average. Cer-
tainly the density of the Sun’s outermost layers must be
considerably less than that, and the density of its interior
core (under the vast pressure of the weight of those
outermost layers) must be considerably more than 1.41 to —
compensate. :
There is an analogy here in the case of the Earth. The
overall density of the Earth, as I have said, is 5.5 grams
_ per cubic centimeter, but the outermost rocky crust has a
density of only. 2.6 grams per cubic centimeter, while the
density rises at the Earth’s center to 11.5 grams per cubic
centimeter. ‘
The change in density is much greater for the Sun than
for the Earth, since pressures are much higher at the
center of the huge Sun than at the center of the relatively
small Earth.
Once Eddington began to explore the internal structure
of the Sun and other stars, the figures for central densities
were found to be unbeliévably high. Calculations show
that in order for the Sun to remain in gravity-temperature
balance, the density at its center must rise to about 100
grams per cubic centimeter or more. This is five times as
dense as platinum (or its sister elements, iridium and
osmium), and these are the densest substances known on
Earth.
Furthermore, while stars that are larger and hotter than
the Sun are less dense, and giants like Epsilon Aurigae are
 SCT RC ah me
Stellar Explosions 169
very rarefied indeed, stars that are smaller and dimmer
than the Sun are more dense. A red dwarf, like the star
- known as-Kruger 60B, has a mass that is one-fifth that of
the Sun. Its volume, however, is only 1/125 of the Sun .
and its density is 1/5 + 1/125 or twenty-five times that of
the Sun. Its average density must therefore be 35 grams
per cubic centimeter, which is more than half again as
dense as platinum, and its density at the center must be
hundreds of times that of platinum. ;
In the first decades of the twentieth century, when these
figures were not known exactly but when they were ex-
pected to be quite high, it is not surprising that it was
assumed that the gas in stellar interiors was compressed to
the point where it no longer acted as a true gas. Theories
. concerning stellar energies and stellar. evolutions were
based on the assumption of nongaseous cores—and ‘proved
to be all wrong.
Eddington’s work in the 1920’s showed that all stars,
down to the red dwarfs, behaved as though they were
perfectly gaseous throughout, despite their densities, since
they obeyed the mass-luminosity relation which was postu-
lated on the assumption of all-gas stars.
But how can substances of such high density behave as
though they were thin gases? Indeed, how can substances
of such high density exist at all? If atoms were, as nine-
‘teenth-century chemists believed, hard little billiard balls,
unbreakable and incompressible, such high densities would
not be possible. In ordinary solids here on Earth, such
atoms are already in contact, and the density of the
platinum metals would represent very nearly the max-
imum possible.
From the 1890’s on, however, all notions about atomic
structure were revolutionized. It became clearer and
clearer that the atoms were. not featureless little billiard
balls, but were complex structures made up of “subatomic
particles” that were individually far tinier than the intact
atom. Whereas the atom as a whole had a diameter of the
order of a hundred-millionth of a centimeter, the diameter
of the subatomic particles was something like a ten-tril-
lionth of a centimeter. To put it in more easily grasped
terms, the subatomic particle was only 1/100,000 as wide
as the intact atom; it would take 100,000 subatomic part-
icles laid side by side to stretch across the diameter of a
single atom.
The volume of an atom was 100,000 X 100,000 X 100,-
000, or 1,000,000,000,000,000 times that of a single sub-
-
 . F a SS a ee
) ; ;
170 The Universe ; :
atomic particle. Since even the most complex atom contains
only alittle over 300 subatomic particles, it can clearly be
seen that the intact atom is largely empty space, held in its ©
wide-open structure by the electromagnetic forces that
kept a few electrons moving through wide spaces about
the tiny atomic nucleus at the center of the atom. If an
atom could be broken up into its individual subatomic
particles and compressed, the whole system could be made
to shrink to a tiny fraction of its former self. (As an
analogy, think first of the amount of space it would take
to store a dozen hatboxes, then imagine the hatboxes torn
into small pieces of cardboard and think of how much
smaller a space they could be packed into. The case of the
atom is far more extreme.)
At high temperatures, the atom is stripped of its outer-
most particles, the electrons; if the temperature reaches
values high enough (and at the center of stars it certainly
does), all the electrons are stripped away, leaving naked
atomic nuclei. Under the tremendous pressures of a star’s
interior, the electrons and nuclei can be compressed into a
-far smaller volume than the original atoms would have
taken up. As the volume decreases, the density correspond-
ingly increases to many times that of platinum.
Such crushed-together quantities of subatomic particles
_ are usually referred to as “degenerate matter.” There is no
question. but that there is a core of degenerate matter
within the Sun and, in fact, within all other stars. This
makes sense, too. In intact atoms, the outer electrons
completely shield the atomic nuclei and prevent them
from colliding directly and combining with each other. It
is only when the electrons are stripped away completely
that nuclear fusions can take place rapidly enough to
support stellar radiation.
Furthermore, although degenerate matter is squeezed
into immensely ,high densities, the individual subatomic
particles making it up are so small that the degenerate
matter is still very largely empty space. Degenerate mat-
ter, composed of separate subatomic particles, is as nearly
empty space, in fact, as are the much less dense ordinary
gases made up of the much more bulky intact atoms. It is
for this reason that degenerate matter may have incredibly
- high densities and still act like a gas.
Yet the densities of even the red dwarfs are as nothing
compared with those of the white dwarfs.. Sirius B, with
the volume of Uranus and the size of the Sun, must have
a density about 125,000 times that of the Sun or about
 2 StellarSeana : 171
8000 times that of platinum. A cubic centimeter of the
average material of Sirius B would weigh 200 kilograms.
A cubic inch of it would weigh a ton and a half. The
central regions of Sirius B must be much denser than this
average value, high as that is, and yet Sirius B, also, seems
. to behave as though it were completey gaseous.
Clearly, Sirius B must be made up almost exclusively of -
degenerate matter. This was hard to accept even in the
1920’s, but corroborative evidence was found. In 1915,
Einstein had worked out his general theory of relativity ©
which. predicted, among other things, that light would
experience a red shift if it traveled outward against a
gravitational force..Ordinary gravitational fields, his theo-
ty showed, would produce a red shift too small to meas-
ure, and at the time he advanced his theory, he did: not
realize that any sufficiently strong field might exist.
Eddington pointed out that if Sirius B were really as
dense as it seemed to be, it ought to- have a surface
gravity 2500 times that of the Sun. Under such conditions,
the “Einstein shift”? would be measurable. In 1925, W. S.
Adams checked the spectrum of Sirius B even more care-
fully, measured the position of various’ spectral lines after
allowing for the star’s radial motion and, sure enough, the
Einstein shift was found. This was an important point in
favor of the validity of the general theory of relativity. It
was also an important point in favor of the super-density
of Sirius B.
Sirius B is by no means unique. Other super-dense white
dwarfs are known. Procyon B is one of them, of.course,
With about 0.65 times the mass of the Sun. At least a
hundred other such objects have been discovered, many by
the Dutch-American astronomer Willem Jacob Luyten
(1889- ). In 1962, he discovered one with a diameter |
only half that of the Moon, the smallest one yet found.
A hundred white dwarfs does not:seem to be a very
impressive number against many billions of ordinary stars,
but remember their small size and dimness. They can only
be seen if they are quite close to us, whereas ordinary
stats can often be seen over immense gaps of space. The
fact that so many white dwarfs have been found despite
the handicap of dimness makes it plain that they must be _
very common objects indeed. Some have even estimated
that they make up 3 percent of all the stars in the Galaxy,
which would mean a total of some three billion white
dwarfs in the Galaxy. E
To return to the matter brought up at the end of the
 nyten ae oi ee

172 : ; The - Universe

previous chapter, is it toward the white dwarf stage that
the stars with nuclear fuel gone are heading, as they
approach the lower left region of the H-R diagram? Ap-
parently, yes—but not always by way of a smooth and
peaceful transition.

Supernovae

In the light of the new nuclear view of stars, it would

seem tempting to interpret the nova (see page 101) as a
kind of Cepheid gone wrong, so to speak. Where the
ordinary Cepheid pulsates in a controlled and measured
fashion, puffing up and settling back over and over, a
nova is a star which, after a long period of quiescence,
manages, for some reason, to build up a sudden head of
outward pressure—quite literally—explodes.
Its luminosity increases rapidly from 5000 to 100,000
times, as its surface area increases, especially since it is
_ suffering a runaway surge of radiation which keeps it from
cooling down as it expands. At its peak, such a nova
reaches an absolute magnitude of —8, at which time it is
about 200,000 times as bright as the Sun. .
This brightness peak only lasts a few days, however.
The force of the explosion puffs a portion of the star’s
matter out into space, and with it goes much of the
energy. What is left of the star begins to collapse again,
like a punctured balloon, and the star dims. It takes
several months, perhaps, to get back to its pre-nova
brightness and after that it continues much as it did
before.
. Such an explosion is catastrophic on an earthly scale. If
it happened to the Sun, Earth’s oceans would boil and its
life would probably be wiped out. On astellar scale, the
nova explosion is not too bad. The peak brightness is high
but, even at its height, a nova is not as brightas S
Doradus is all the time. As for the material lost in the
course of the nova explosion, this does not amount to
more than about 1/100,000 of the mass of the star—
scarcely enough to matter.
After a quiescent period, a nova is quite capable of
undergoing another explosion with intervals of anywhere
from 10 to 100 years. One star, T Pyxidis, has been
observed to have four such novalike peaks of. brightness
since. 1890. Whatever property they possess that causes
them to unbalance and blow their tops may, apparently,
 Cee RA ee PM ON
Stellar Explosions "173
persist on occasion, and cause it to happen again and
again, rather like a volcano that occasionally erupts.
Nor is the explosive nature of novae a matter of theory
only; the effects of the explosion can be observed. As the
nova approaches maximum (if astronomers are lucky
enough to be observing it on its way to the peak and do
not catch it—as all too often happens—only after the
fact), absorption lines in its spectrum showa strong violet-
shift, indicating the star is approaching us. Part of it
certainly is; for the exploding outer layer moves rapidly
away from the star and the portion between ourselves and
the star approaches us at a good, spanking clip.
The explosion, or its aftermath, is even visible optically
in some cases. After Nova Aquilae appeared in 1918, E.
E. Barnard noticed that it was surrounded by a nebulous —
sphere that had not béen there before. This sphere, pre-
sumably composed of exploding gases, continued to move
outward at a uniform rate, growing slowly more volumi-
nous and fainter until 1941 when it had become too faint
to detect. Other novae exhibited similar phenomena.
Unfortunately, astronomers have still not arrived at any
generally accepted explanation for the nova explosion.
One recent suggestion is that novae occur only among
stars that are part of a closely spaced binary system and
that it is the interaction between these stars that occasion- .
ally sets off the nova. Perhaps.
Not. all exploding stars are, however, merely novae.
This became evident in the mid-1920’s when the enormous
distance of the Andromeda galaxy first came to be appre-
ciated. If that distance were accepted, what of the nova, S
’ Andromedae, that had appeatys in it in 1885 (see page
103)?
When S Andromedae was first detected, it was of the
seventh magnitude, but there is a chance that it was not
’ detected until slightly after it had zoomed rapidly upward
to its peak, and that it had been visible briefly to the
naked eye as a star of slightly better than the sixth
magnitude.
This would still make it only a dim star, to be sure, and
- one that was just barely visible—but to be visible at all to
the naked eye at the distance of the Andromeda galaxy
bespeaks a brilliance so enormous as to stagger even the
hardened astronomer. § Andromedae at its peak outshone
-all the rest of the Andromeda galaxy. A single star was
brighter than the combined brilliance of billions of ordi-
nary stars.
 7 ee

174 te The Universe

Modern estimates, based on the latest values accepted

for the distance of the Andromeda galaxy, make it appear
that the absolute magnitude of S Andromedae at its peak
was —19. This meant that, for a few days anyway; it was
shining with a brilliance equal to a hundred thousand
ordinary novae, or nearly ten billion times that of our
~ Sun.
S Andromedae was not merely a nova; it was a “super-
_nova.”
Once this was realized, the search was on for other
examples of this new and spectacular class of objects.
- Whereas ordinary novae cannot be seen much beyond the |
nearest galaxies, the supernova, which is as bright as an
entire galaxy, can, of course, be seen as far as galaxies
can; that is, as far as our telescopes can reach.
Outstanding in the search for such galaxy-bright super-
novae was the Swiss-American astronomer Fritz Zwicky
(1898- ) who, in the years following 1936, located
a number of them in various galaxies. He estimates that
whereas. ordinary novae appear at the rate of twenty-five
per year in any given galaxy, supernovae appear at the
tate -of only three per thousand years in any given
galaxy.
_ By the end of the 1930’s penne by the German-
American astronomers Walter Baade (1893-1960) and
_ Rudolf Leo B. Minkowski (1895- ) had shown that
supernovae could be divided into two varieties, Types I and
_ Il, with, possibly, a Type III as well. The Type II super-
novae are the less luminous, only a couple of hundred times
brighter than an ordinary nova and probably the more nu-
merous, although fewer are seen simply because of their
lower luminosity. Type III supernovae are like the Type II,
but with a light-curve that fades off more gradually. Type I
supernovae are the true giants of the class, and S Androme-
dae was.a Type I supernova.
Clearly, supernovae, like ordinary novae, owe their in-
crease in brightness to an explosion. Since supernovae
reach greater peaks of brightness than do ordinary novae,
and stay brighter longer, the explosions accordingly must
be more catastrophic. It is estimated that whereas ordi-
nary novae lose 1/100,000 of their mass in the explosion,
a Type II supernova_loses from 1/100 to 1/10 of its mass
and a giant Type I supernova losses from 1/10 to 9/10 of
its mass.
Although a number of novae have been studied within
our Galaxy since the invention of the telescope, not one
 <2.
Se ae (A

Stellar Explosions — 1TS

_ object that can clearly be labeled a supernova has been de-
tected in our Galaxy in the three and a half centuries since
Galileo first pointed his magnifying tube toward the heavens.
However, looking back into history, there seem to have
been three supernovae-in our Galaxy in the course of the
last thousand years (right up to par, according to Zwicky’s
estimate). They were the “new stars” of 1054, 1572,-and
1604 (see page 170). There are also signs (according to
painstaking searches through Oriental records, reported in
1966) that a fourth supernova may have blazed out in
1006. ;
‘Of these, the feat was the most poorly observed but has
turned out to be the most interesting by far. It may have
been one of the brightest supernovae that ever formed and
possibly the closest in historic times. Moreover, it is
‘unique among supernovae in that it has left behind a re-
markable remnant of itself, in the form of what seems, in a _
small telescope, to be a cloudy patch.
In 1764, Messier observed this patch in the constellation
Taurus, and entered it in his list of nebulosities. Indeed, it —
was the first object in his list and is sometimes called
M1.
Eighty years later, in 1844, Lord Rosse observed it
more closely with his large telescope and was able to
make out its structure. It was like nothing~else in the
heavens, a mass of clearly turbulent gas with numerous
filaments of light within it. Because the numerous. ragged
filaments reminded Rosse of the legs of a crab, he named
the object the “Crab Nebula,” and that remains its name
to this day.
The Crab Nebula looks for all the world like a vast
explosion caught in mid-expansion. It takes no imagination
at all to see that; the conclusion forces itself on any
observer. The fact that it is located in just about the spot
where the Chinese astronomers of 1054 had placed their
supernova made it very tempting to consider the Crab
Nebula the remnant of that explosion, and by the mid-
1920’s, when astronomers came to understand the fact
that such supernovae existed, this view was generally ac-
cepted.
Photographs of the Crab Nebula taken over the last
generation or so. have shown the turbulent gases to be
moving outward by an amount that is tiny in terms of
seconds of arc per year, but measurable. Spectroscopic
evidence also shows that the portions of gas on the side
toward us are approaching us at a rate of about 1300
 - 176 The Universe >

kilometers per second. If the Crab Nebula is really an

exploding volume of gas, that gas ought to be moving
outward in all directions at roughly equal velocities: The
radial velocity in kilometers per second can therefore be
set equal to the transverse velocity in seconds of arc per
year. The distance at which such a velocity will make
itself evident as so many seconds of arc per year can be
calculated, and it turns out that the Crab Nebula is about
.4500 light-years from us and that the exploding sheli of
gases now has an extreme diameter of about 6 light-
ears.
If we take the rate at which the gas is expanding and
calculate that backward, we can show that the gas was at
its central starting point some 900 years ago, which is
exactly what would be’ expected if the Crab Nebula were
indeed the exploding supernova of 1054. By 1942, there
seemed no doubt about the matter, thanks chiefly to the
careful work of Oort.
No such obvious markers are to be found of Brahe’s —
supernova of 1572 or ‘of Kepler’s supernova of 1604 (the
latter perhaps a Type II supernova only), although certain
faint wisps of gas about 11,400 light-years from us (three
_ times the distance of the Crab Nebula) were identified in
1966 as possible remnants of Brahe’s supernova. On the
other hand, there is one class of astronomic object which
may be what is left of supernovas that blazed out in
Earth’s heavens before the dawn of written history or
even, perhaps before the dawn of man.
These are the so-called planetary nebulae: stars sur-
_ Tounded by extended spherical gaseous nebulae. We. see
through the greatest thickness of this gaseous halo at the
edges, and it is there that it becomes most prominent, as
in the Ring Nebula in the constellation Lyra. Such a ring
of gas is reminiscent of Laplace’s picture of shells of gas
being given off by a rotating nebula which, eventually, was
supposed to condense into planets, whence the name of
the objects. The Veil Nebula in Cygnus seems to be
broken fragments of such a shell expanded to relatively
enormous size. It may represent the remains of a superno-
va explosion that took place 100,000 or more years
ago.
There are about 500 planetary nebulae known, and
many thousands probably exist in the Galaxy. The nearest «
is NGC 7293 which is close enough to us to have a
measurable parallax. It is about 85 light-years away, and
the diameter of the ring of gas is about a third of a
 - Stellar Explosions 177°
light-year. There are various suggestions as to the nature
and cause of these gaseous shells, but it is at least. possible
that some, if not all, represent the material shot out many
thousands of years ago in a nova or supernova explosion
of the central star.

Dying Stars

The two types of objects discussed so far in this chap-

ter—white dwarfs and supernovae—seem to be closely
related.
To see how, let us return to stars at the final stage of
~ nuclear fuel consumption—stars that have been accumulat-
ing iron in their innermost core and have nowhere else to
go in the realm of nuclear reactions.
For such a star to continue to radiate, the one remain-
ing energy source, the gravitational field, must be tapped. -
Once again the star must contract, as it had done once in
the days long past, before nuclear reactions had been
ignited in its core. Only now, when the star is radiating
energy at a tremendous rate, the contraction must be
rapid indeed if enough energy is to be supplied.
Once the nature of degenerate matter was understood,
astronomers could see that this contraction could be ex- _
ceedingly rapid, and that what had previously been an
ordinary star could be converted in almost no time into a
_tiny white dwarf. The heat of compression would make it
white hot, but because of its small surface area, it would
be radiating far less energy, altogether, after compression
than before. The amount of energy it would be radiating
as a white dwarf was small enough to be fed over long
eons by an exceedingly slow rate of further compression.
The process which Helmholtz had once thought applied to
all stars generally, turned out to be applicable, indeed, to
white dwarfs.
Nor is a white dwarf deprived, by its enormous density,
of the possibility of further compression. Much capacity in
that direction remains. Sirius B might have a density
125,000 times that of the Sun, but the subatomic particles
that swarm about in its largely degenerate structure are
by no means in contact. Sirius B would have to shrink to a
diameter of a mere eight miles or so (!) before that would
be true. ;
As a white dwarf shrinks, it cools. Its surface tempera-
ture may be as high as 50,000° C.. at the moment of
formation, but Sirius B with its 10,000° C. surface tem-
 178. hs The |‘Universe
perature is stil a fairly young white dwarf, close to the
possibly maximum luminosity of 1/100 that of the Sun.
Older white dwarfs are correspondingly cooler, and one,
called “van Maanen 2,” which has been in the white dwarf
_ State for at least 4 eons, has a surface temperature of only
4000° C. It is distinctly reddish in color so that it is that
apparent contradiction in terms, a “red white-dwarf.” Still
even van Maanen 2 can continue to exist for many, many
eons, doling out its gravitational supply of energy, before
it flickers out. Such is the extended lifetime of a white
dwarf that it may well be that the Galaxy itself is not old
enough, yet, to have witnessed the final flicker of a single
such star.
' But if white dwarfs are dying stars, an interesting prob-
lem is raised by the Sirius system. Sirius A and Sirius B
- must have been formed simultaneously, and yet Sirius A is
in the prime of its life while Sirius B is in extreme old age.
Why should that be?
If both are of equal chronological age, it can only be
that Sirius B lived faster than Sirius A, and this would
~ imply by the mass-luminosity relation that Sirius B is more
massive than Sirius A, perhaps far more massive. Yet,
Sirius B, in its white dwarf state at least, has only half the
mass of Sirius A. What happened to the rest of the mass
that must once have belonged to it?
The one way we know of in which sizable quantities of
stellar mass can be lost is through a supernova explosion.
Can it be then that Sirius B, an eon or more ago, under-
went a supernova explosion? It is quite likely.
Such a conclusion is supported by the theoretical
work of the Indian-American astronomer Subrahmanyan >
Chandrasekhar _(1910- ). Even in a white dwarf,
there must be a balance between the gravitational force -
. compressing the star and the temperature effect forcing its
substance outward. The gravitational forces involved in a
white dwarf are many times more intense than those of an
' ordinary star, and the central temperature must rise to
balance it. The greater the mass of a white dwarf, the
stronger the gravitational force compressing it and the
smaller and denser it must be. At some critical point
(“Chandrasekhar’s limit”), no temperature would suffice
to keep the white dwarf’s structure from contracting to
some ultimate limit. In 1931, Chandrasekhar showed that
this critical limit was equal to about 1.4 times the mass of
the Sun. This suggestion is supported by the fact that of
 —.

Stellar Explosions _ 179

those white dwarfs whose mass has been determined all
are well below Chandrasekhar’s limit.
Once a star of average size, like our Sun, has exhausted
the nuclear fuel of its core, it would contract more or less
steadily. In the process it may undergo a mild explosion as
the heat of contraction ignites some of the nuclear fuel
remaining in its outer layers. There may even be several
such explosions in the course of time. At least some
ordinary novae may represent this stage of a star’s evolu-
tion. A white dwarf may even have formed in this fashion
under the watchful eyes of astronomers. One recurrent
nova, WZ Sagittae, exploded first in 1913 and again in
1946; it is now only 1/100 the luminosity of the Sun and
shows every sign of being a white dwarf.
The more massive the star, the more drastic the effect
of compression and the more tremendous the explosions
that result. Such explosions blow off larger and larger
fractions of the mass, reducing what is left to a safe value
below Chandrasekhar’s limit.1 Clearly then, a supernova is
the death throes of a massive star, and a Type I supernova
is the death throes of a particularly massive star.
But exactly what sparks off the final collapse of a
massive star? It happens so suddenly. At least two
mechanisms for this sudden catastrophe have been offered
and both may be valid, each in its own type of star.
One possibility is that the temperature of the iron-
choked, innermost core eventually rises so high that, at a
certain critical point, the iron atom is knocked apart by
extraordinarily intense radiation into helium fragments.
This is not a process that yields energy; it absorbs energy,
all the energy contained in the radiation that sparked it.
This happens to a certain extent at a wide range of
temperatures, but at the critical point, it happens at such a
rate that the energy absorbed can no longer be replaced
by the last dregs of nuclear processes going on in the core.
The temperature in the core drops slightly.
As the temperature drops, the star’s gravitational com-
pressive force, no longer completely countered, takes
over. The star shrinks, and the core heats up at the
expense of gravitational energy. The energy thus supplied
keeps the iron-to-helium process going, and the star must
shrink further at an ever-increasing rate. In short, all the

1 If the star for any reason fails to get rid of enough mass to bring
what is left below the limit, other possibilities offer themselves, which
will be discussed later in the book (see page 259).

>
 Moe Sarwan tO, he eee oeoe et eat ee TT NS eee ee
= 7 MY fe x be ~ oa hee 7
es Ne Sale ara ye t \ ae oe Sate

180 _ The Universe

energy given off by the core over the space of some
millions of years of helium fusion to iron must now be
paid back in a matter of hours (!) from the only source
- of energy left—the gravitational field. :
A second alternative was suggested in 1961 by the
Chinese-American astronomer Hong-Yee Chiu (1932- ).
To explain this second alternative, some background
must be introduced.
The ordinary radiation produced by a star’s core is
easily absorbed by matter. A fragment of radiation once
released is quickly absorbed, released once more, absorbed
_ again, and so on over and over. The radiation is passed
from hand to hand, so to speak, in any direction at
random, and it makes its way out toward the star’s surface
only very gradually. It is estimated that it takes an aver-
age of a million years for a given bit of energy to move
_ from the Sun’s core, where it is formed, to the Sun’s
surface where it is radiated away.. This makes the sub-
stance of the Sun an excellent insulator, so that its center
can be at 15,000,000° C., while its surface, less than haif
a million miles away, is only at 6000°C.
In addition to ordinary radiation, the nuclear reactions
in the cores of stars produce extremely tiny particles
Called “neutrinos.” Neutrinos travel at the speed of light,
as ordinary radiation does, but there is this important
distinction; neutrinos are only very rarely absorbed by
matter. Any neutrinos formed in the Sun’s center travel
outward in all directions without being affected in any way
by the matter making up the Sun. In three seconds, they
reach the Sun’s surface and then flash away into outer
space, carrying some of the Sun’s energy with them.
At the temperature of the Sun’s interior, the number of
neutrinos formed is minute in comparison to the ordinary
radiation emitted. If the neutrinos are considered as a
kind of energy-leak, so to speak, they form a virtually
imperceptible one in the Sun.
3 As a star ages, however, the temperature of its center
increases and the rate of neutrino-formation increases
more and more rapidly. The neutrino energy-leak becomes
more serious. When a critical temperature of 6,000,000,-
000° C. is reached (according to Chiu’s calculations), the
leak becomes so great that what nuclear reactions are
carried on in the core are no longer sufficient to supply
the energy required to keep the star from collapsing.
Whether the loss of energy at the core is caused by the
sudden initiation of an iron-to-helium process or of an
-
 Stellar Explosions 181
‘overwhelming neutrino energy-leak, the result, in either
case, is the sudden catastrophic collapse of the star. In the
process of collapse, the star’s outer layers compress. These
outer layers, however, still contain- nuclear fuel—even
hydrogen at the very surface. This hydrogen would not
undergo fusion at the surface temperature of even the
hottest stars, but with the added heat of compression, all
the fuel remaining in the star is ignited. In a short space
of time, energy that ordinarily might suffice for hundreds s

of thousands of years is radiated away.

If supernovas are the marks of the conversion of mas-
Sive stars to white dwarfs, then surely there should be a
white dwarf at the center of the Crab Nebula. As a
matter of fact, a tiny, bluish star—as hot as one would
expect a new white dwarf to be—does exist there, and it is
generally considered to be just that, a white dwarf. ,
There should also be white dwarfs at the center of at
least some of the planetary nebulae if these are remnants
of ‘supernova explosions, The central stars of planetary —
nebulae are uniformly blue-white stars as, again, one
would expect of relatively new white dwarfs. The central
star of the nearest planetary nebula, NGC 7293, is only of
the tenth magnitude despite its mere 85-light-year distance
and is clearly a white dwarf.
If there are indeed three supernova explosions per thou- —
sand years in a given galaxy, as Zwicky estimates, then in
the 5 eons that the Solar system has existed, there must
have been some 15,000,000 supernovas and the forma-
tion of that many white dwarfs. Add to these, supernovas
that may have taken place before the Solar system was
formed, and the white dwarfs formed of stars small
enough to avoid the supernova stage, and it is not at all
surprising that white dwarfs are as common as they seem
to be.

Second-Generation Stars
When a large star explodes as a supernova, a quantity
of matter pushes outward into space and, eventually, dis-
tributes itself among the thin gas already present there.
The exploded stellar matter is itself rich in all the
elements up to iron since all these were present in the star
at the time of explosion. Indeed, the exploded matter must
also ig te the elements beyond iron. These elements -
cannot be formed without an input of energy, and in an
ordinary star they would not be formed. In a supernova,
t ae ee, ee Ne oe
Poe
A ae
Oe tae ah‘ i ce oe oe fo, ee
St) oe eeee Ae
eaBee
182 The Universe
however, energy is produced at such an enormous rate’
that some is very likely to be put to work, so to speak,
building up atoms more complex than those of iron.
Atoms all the way to uranium, the most complex atom of
those occurring in quantity in the Earth’s crust, must be
built up.
It is even very likely that elements more complex than
uranium are formed. Such “trans-uranium elements” do
not exist naturally on Earth because they are quite un-
stable. To be sure, uranium is itself unstable and is con-
stantly breaking down, in stages, to lead. The uranium
breakdown is so slow, however, that even after 4.5 eons
(nearly the lifetime of the Earth) half the original supply
of uranium remains intact. The breakdown of the ele-
ments beyond uranium is much more rapid, and if any of
these elements existed at the time of the Earth’s forma-
tion, they are all gone now.
In supernovae, however, there is at least one piece of
evidence in favor of an at least transitory appearance of
the element “californium” (six places beyond uranium in
the usual listing of elements). Many supernovae seem to
lose brightness with what seems a half-life of fifty-five
‘days. It happens that one variety of californium,: “califor-
nium-254,” breaks down at just that rate. There may
possibly be a connection here.
~ In any case, long after a supernova has flashed to its
__ destruction and the resulting white dwarf has moved on in
its journey around the galactic center, the gaseous shell of
the explosion will linger behind through friction with the
_ thin gas already present in interstellar space and serve to
contaminate that gas in the region where the explosion
had once taken place. A region of space which orginally
contained a thin gas that was exclusively hydrogen may
now contain a smail portion of helium is even anal
portions of more complex atoms.
What nov, if a star is formed out of the matter of such
contaminated space? It would still be mostly hydrogen; it
would still have a long lifetime of many eons during which
it could radiate, with hydrogen fusion as its energy-source
(provided it were not too massive in‘the first place), but it
would contain larger helpings of complex atoms than one
might expect.
This seems to be the case with our Sun (and thus we
answer the problem raised on page 138). The Sun seems
to be a “second-generation star,”. one that formed in an
 ‘Stellar Explosions 183
area where once a previously existing star had died ex-
plosively.
If the Sun had orginally been 100 percent hydrogen, it
would have taken 20 eons of time for it to decline to its
present 81 percent of hydrogen. If it were only 87 percent
hydrogen to begin with, it would have taken only 5 or 6
eons to reach its present state, and this ey actually be
what happened therefore.
The planets forming ‘on the outskirts of the slowly
agglomerating gas cloud that formed the Solar system
would also end by being made up of material containing a
considerable admixture of more complicated atoms. In the
case of the Earth, which was too small and hot at the
beginning to hold on to hydrogen and helium, the more
_ complicated atoms make up almost the whole of the
structure. The Earth has an inner core, making up fully
one-third of its mass, which is almost entirely liquid iron.
This seems to be an indication of the quantities of iron
spewed into space by the exploding supernova with its
own iron-choked innermost core.
Indeed, Fred Hoyle suggested some years ‘ago that the
Sun was once, like Sirius, part of a double star system and
that the partner flashed into a supernova. In this way,
Hoyle tried to explain the constitution of the planets and
at the same time account for the fact that they possess so
much angular momentum, since they would still possess
much of the supply of the original exploded star. Howev-
er, if we accept this hypothesis we have to solve the
problem of the whereabouts of the white dwarf that ought
to have been formed in the course of the supernova.
Hoyle’s suggestion is novel and interesting, but it is not
considered at all likely by astronomers generally.
 ERAN erceting Ca tod Cot Nother ie es ee rr

CHAPTER 12

Galactic Evolution
The Question of Eternity _

As you see, then, our consideration of the. problems of °

_ stellar evolution has ended by suggesting that while the
Sun might be only 5 eons old, there must be a Galactic
history before that, for the Sun is built on the ruins of a
still older star. Indeed, careful consideration of the struc-
ture of certain globular clusters, of the quantities of hy-
drogen they have consumed as compared with that which
remains to be consumed, pives cluster Hieauacs of up to 25
eons.
But is even that a necessary maximum? Is there any
need to postulate any definite age for the Universe at all?
A star might have a definite age, but there could have
been stars, living and dying, before the present star in
endless succession. One might consider the analogy of
the human race which has existed for a period of
time much longer than the lifetime of any single indi-
vidual.
In a way, of course, we might argue that the energy of
the Universe (including matter, as one form of energy) -
has always existed and always will exist since, as far as .we
_ know, it is impossible to create energy out of nothing or
destroy it into nothing. This implies, we can conclude, that
the substance of the Universe—and' therefore the Universe
itself—is eternal.
That, however, is not what we really mean. We are
concerned with more than the mere substance of the
Universe. The question here is whether that substance has
always taken and will always take the form of the kind of
Universe we know—one with stars and planets, and capa-
ble of playing host to living things like ourselves, or
184.
 ~ Galactic ‘Evolution 185
whether this Universe-we-know had a definite beginning
and will have a definite end.
As- far as we can tell now, the Universe-we-know.
(which is what I shall mean when I speak of the “Uni-
verse” henceforth) exists in this fashion on the basis of an”
energy output developed by hydrogen fusion. Before hy-
drogen fusion began, the Universe would have to be
pictured (on the basis of the discussion in the previous
chapters) as no more than a vast mass of whirling gas,
perhaps dimly red-hot in spots. Once all the hydrogen has
been fused, the Universe would have to consist of nothing
but white dwarfs which have progressed for varying peri-
ods of time down the road to final darkness and extinc-
tion. It is only during the period when the process of
hydrogen fusion is progressing massively that we have the
Universe-we-know.
If we ask, then, whether the Universe might not be
eternal, we are, essentially, asking whether hydrogen fu- —
sion might not go on forever.
Now we can see why, on page 108, I raised the question
as to whether the Universe were infinite and then left the
matter where it was while I veered off to take up the
question of the age of the various heavenly bodies. For
now I am raising the similar question as to whether the
Universe is eternal, and it turns out that the two questions
are closely related in some ways.
For instance, if the quantity of hydrogen in the Uni-
verse in infinite then it could, clearly; continue to fuse to
form helium forever, provided the rate of fusion was
finite. In other words, an infinite Universe ought to be an
eternal Universe as well. On the other hand, a finite
Universe might inevitably be finite in time as well as in space.
For a finite Universe to be eternal despite its finiteness
there must be some process that can reverse the fusion of
hydrogen, restore the hydrogen and make it available for
fusion once more. Furthermore, this must not be done at
the irrevocable expense of other energy sources, such as-
gravitational fields.
It must be stated that atfirst glance this does not seem
possible. To be sure, according to firmly accepted scientific
doctrine, energy is conserved and can never be destroyed.
_ (This is sometimes called “the first law of thermodynam-
_ ics.”) However, although energy is always with us in
constant quantities, it is not always available for conver-
sion into useful work, and it is this availability for conver-
sion into work that is a fundamental requirement for the
 186 “The Universe
Universe-we-know. Indeed, a generalization, which is
- known as “the second law of thermodynamics” and which
seems to be as valid, universal, and important as the first
law, tells us that the amount of energy available for
conversion into work decreases constantly.
This means, to cite a very simple case, that water that
has flowed downhill cannot flow uphill again by itself; it
must be pumped uphill at the expense of energy from
some other source. Any system that has run down, wheth-
er it has done work in the process or not, can be restored
to its original state and allowed to run down- -again—but
_ only. at the expense of outside energy. Furthermore, it
always takes more energy to restore a rundown’ system
_ than the energy you would get by allowing it to run down
again, so that there is always a net loss in the process of -
restoration. The second law is, as far as we know, inviola-
ble.
We could conclude, then, that, little by little, more and
more of the energy of the Universe will be tied up,
irrevocably and irreversibly, into white dwarf remnants,
and that will be the end of the Universe from our stand-
point. Working backward, we might also conclude that
although there were stars before the Sun, there must have
been a time, say 25 eons ago, when all the energy of the
- Universe was in the form of thin wisps of swirling hydro-
gen, and that would be the beginning of the Universe
from our standpoint.
By this line of argument the Universe might have a
lifetime of perhaps 1000 eons; and of this enormous (but
finite) length of time, a fortieth has already passed.
Yet this conclusion rests on the assumption that the -
first and second laws of thermodynamics are really valid
everywhere in the Universe and not only in that small
portion of it which we have been able to inspect, and that
_ they are valid under all possible circumstances and not
only those we have been able to witness.
Keeping in mind that the assumption of the validity of
the laws of thermodynamics may be questioned, let us
continue to try to find out whether the Universe is infinite
and eternal, or finite and time-bound. To do so, let us turn
from individual stars to the galaxies.

Classes of Galaxies

Notions as to the evolution of stars were first developed

by a consideration of the different properties of different
 eee
Galactic Evolution 187
kinds of stars. The different properties were derived al-
most entirely from the spectra since the stars themselves
were too small to show structural detail.
Notions as to the evolution of galaxies can also be
developed by a consideration of the different properties of
different kinds of galaxies. Here, though, there is a differ-
ence; galaxies are much larger than stars. Even though
galaxies are far more distant from us than are the individ-
ual stars of our own neighborhood of space, these distant
galaxies appear in our telescopes as more than points.
They are-patches of light which have particular shapes,.
and several thousand of the nearer among them even |
show considerable detail.
Hubble noted the three main classes of galaxies: spiral,
ellipsoidal, and irregular (see page 107). He was’ able to do
better than this, however, and in 1925 published a detailed
classification that has been- used ever since. The ellipsoidal
galaxies, for instance, which lack spiral arms and look like
very distant and very large globular clusters, differ among ~
themselves in the degree of flattening. Some are virtually
spheres (and might be called “spheroidal galaxies”), while
some are rather flattened, some quite flattened, and so on.
Hubble symbolized all the ellipsoidal galaxies as E and
_ distinguished the degree of flattening by numbers. EO -
represented the spheroidal galaxies, and El through E7
represented increasing degrees of flattening. An E7 galaxy
is quite flattened, with its ends (if seen edgewise) sticking
out’‘point-fasbion almost as though it were on the verge of
possessing spiral arms.
As for the spiral galaxies, Hubble Fecpunteed two kinds.
There was first the ordinary spiral in which the arms are
directly connected with and wrapped about the ellipsoidal
nucleus as in the case of the Andromeda galaxy. Then
there were galaxies in which a straight bar of stars seems
to extend outward from the nucleus on either side. From
either end of this bar, there extend spiral arms. These are
the “barred spiral galaxies,” which seem to make up about
30 percent of all the spirals.
Hubble symbolized the ordinary spirals as S, and the
barred spirals as SB. He then differentiated among the ~
spiral galaxies of both classes according to. how tightly or
loosely the arms were wrapped about the nucleus, using
lower case letters for the purpose: a for the most tightly
wrapped arms, then b and c for looser structures. A spiral
galaxy could be referred to as an Sa, Sb, or Sc; or if
barred, SBa, SBb, or SBc.
 =188 ~~ -o.\The’ Universe
The Andromeda galaxy is~ classified Sb. Our own
Galaxy has usually been considered similar to the An-
dromeda galaxy and has therefore been classified as Sb
also. There may be a change here though. On the basis of
measurements made in 1965 of the brightness of stars
near our Galactic nucleus, it was suggested that the
Galactic nucleus is smaller than had previously been
' thought. It may be only 6500 light-years. across, half that
of the nucleus of the Andromeda galaxy. If the nucleus of
our Galaxy is indeed smaller than had been thought, and
the spiral arms correspondingly more prominent and wide-
ly spaced, then our Galaxy would resemble the Whirlpool
galaxy rather than Andromeda.: Our Galaxy, like the
Whirlpool, would then belong to the Sc classification.
Hubble arranged all these forms into a progressive
order as follows:
-ySa >Sb Sc
E0>E1>E2>E3>-E4~E5*E6?E7?S0
~ SBa>SBb>SBc

where SO represents a hypothetical form with characteris-

tics intermediate between ellipsoidal, spiral, and barred
spiral.
Hubble did not specifically state that this represented, in
his opinion, an evolutionary change, but apparently he felt
it did. Certainly the gradual progression of change from
spheroidal to loose spiral made.an evolutionary hypothesis
very tempting, and for a decade or so after 1925, the
views on galactic evolution resembled a kind of Laplacian
nebular hypothesis on a:much vaster scale.
Imagine, to begin with, a quantity of gas not merely
large enough to form a star, but large enough to form a
hundred billion stars. Within this tremendous amount of
gas, which we might call a protogalaxy, processes went on
which led to its condensation into billions of stars. The
protogalaxy would begin with a certain supply of angular
momentum and as it condensed, it flattened: It would
begin as a spheroidal galaxy and gradually flatten, passing
from E0 to E7.
As the galaxy as a whole continued to- contract and
grow more compact, the rotation would become more
rapid and so would the flattening progress until, just as in
Laplace’s planetary-system nebula, fragments would be
given off at the equator. Spiral arms would form, with or
without forming a bar (and neither Hubble nor anyone -
else has ever explained why the bar should exist at all).
 Galactic Evolution _ 189
_ Furthermore, as time went on and the galaxy continued to-
contract and increase its rotation, the arms would contin--
ue to move away from the nucleus, so that both spirals
- and barred spirals would pass through the a, b, and c
stages. The irregular galaxies might represent the final
stage of this evolutionary scheme.
Since elliptical galaxies seem to be particularly large
and irregular galaxies are usually small, we would have to
suppose that the larger a galaxy the more slowly it passed
through such a development. The very large galaxies
would linger in the elliptical stage, while the very small
galaxies would hasten through to the final irregular
stage. .
If this theme is correct, then the Andromeda galaxy -
and, even more so, our own Galaxy, are relatively old,
and far down the line of evolutionary development.

Stellar Populations

- By the 1940’s, however, new sets of views were arising.

In 1942, Baade had an unusual opportunity. The city of
Los Angeles was blacked out because of World War II,
and it was possible to get a clearer look at the Andromeda
galaxy than ever before, making use of the 100-inch Mt.
Wilson telescope.
Until then, only the stars in the spiral arms had been |
made out individually by Hubble and by those who fol-
lowed him. Now Baade was able to see and photograph
stars in the Andromeda nucleus.
An important difference showed up. The brightest stars ©
in the spiral arms were giant blue-white stars, very large
and hot, something like the brightest stars in our own.
neighborhood of our own Galaxy. The brightest stars of
the Andromeda nucleus were, however, reddish stars;
there were no blue-whites at all.
This seemed to fit in with general spectral information.
The spectrum of the Andromeda nucleus, and of those
other galactic nuclei that could be examined, as well as
the spectra of ellipsoidal galaxies generally, tended to be
of spectral-class K. The average starlight of those regions
tended to rise from surfaces dimmer and cooler than that
of the Sun. The general spectrum of the spiral arms of
Andromeda and of other galaxies tended to be of spectral
class F, The average starlight of those regions tended to
rise from surfaces brighter and hotter than that of the
Sun.~
 * 190 “The Universe
It was as though regions where stars were concentrated
closely together, as in globular clusters, galactic nuclei,
and ellipsoidal galaxies generally, were largely of one
type, which Baade called “Population II.”-On the other
hand, stars that were more loosely spread out, as notably
in the spiral arms of galaxies, were largely of another
type, which he called “Population I.”
The Population II stars seemed to be distributed in a
kind of spherical halo about the Galactic center, while the
Population I stars seemed to be distributed in a kind of
hollow disc along the central plane of the Galaxy. They
might be referred to as “halo stars” and the “disc stars,”
respectively.
On the whole, Population II stars tend to be sedate and
uniform, moderate to small in size and occupying regions
of space that are relatively free of dust and gas. Popula-
tion I stars, on the other hand, tend to include numbers of
rather spectacular members and to show a wide variety,
. including large stars, more brilliant and more fiercely hot
_. than anything among the Population II stars. Further-
more, Population I stars occupy regions of space that are
relatively rich in dust and gas.
This division into two populations (a division that has
since been made more complicated with Population I and
II both broken down into séveral subclasses) raised new
questions concerning galactic evolution. Gathering knowl-
edge concerning the nature of nuclear reactions within
stellar cores made it seem that large stars were shorter
lived than small stars so that Population I stars had to be
viewed as rather evanescent. Closer analysis showed that
the spiral arms themselves had to be short-lived, and no
completely satisfactory explanation has yet been advanced |
to account for the fact that so many galaxies seem to
sport spiral arms if the latter are as short-lived as they
ought to be,
In any case, the feeling grew that spiral arms had to be
temporary phenomena and that galaxies tended. to lose
them and become elliptical, rather than the reverse. Hub-
ble’s evolutionary scheme was therefore turned upside down.
Suppose that a protogalaxy formed stars in irregular
fashion at first, so that the earliest stage was that of an
irregular galaxy. At the center of the protogalaxy, where
the dust was most concentrated, stars would form most
rapidly and numerously. There would bea relatively small
supply of dust and gas per star, and the stars of the center
would be on the smail side and rather uniform in proper-
 | Galactic Evolution 191
ties. And when they were formed, the dust and gas would
be almost gone. Such stars would be of the Population II
type, very rich in hydrogen, very low in the more complex
atoms.
On the outskirts of the protogalaxy, however, the dust
and gas ‘would tend to be less evenly distributed. Perhaps
the general rotation of the protogalaxy whipped the dust
and gas into shreds, so that stars tended to form in lines
and regions that. turned out to be luminous spiral arms.
Because of the uneven distribution of dust-and gas, stars
would form numerously here and sparsely there; some-
would have an unusually small share of material to begin
with, some an unusually large share. There would thus be
a great variety in their masses, and the large members of
this Population I group would be bright and hot.
In the spiral arms, moreover, much dust and gas would
be left over, having been spread out too thinly in their
far-from-the-center location to serve as nuclei for stars.
Over eons of time they slowly condensed until such time
_as they could finally develop into stars. Furthermore, the
- bright, hot Population I stars would in the course of their
speedy evolution reach the death-stage and explode as
supernovae. The dust and gas in the spiral arms would |
come to be enriched with helium and more complex
atoms, so that Population I stars which developed later
would be second-generation stars, relatively poor in hydro-
- gen and rich in more complex atoms.
Thus, the first stage of galactic evolution after the
original irregular galaxy would be a spiral galaxy with the
arms loosely wrapped.
But the spiral arms are relatively short-lived. The dust
and gas are consumed in star formation, the brighter stars
die out, and the situation in the arms progresses steadily
from a Population I situation to a Population II situation.
Furthermore, as a galaxy revolves, the spiral arms trail
and tend to wind up, so to speak, coming to hug the >
. nucleus more and more closely, progressing from Sc to Sb
to Sa.
Eventually, the spiral arms melt into the nucleus, their
Population I character completely gone, and a flattened
elliptical galaxy is formed.
With time, the stars of such a galaxy, interacting gravi-
tationally with one another, spread their motions more
and more evenly through the whole, so that there is less
and less flattening until a spheroidal galaxy is formed.
Looked at in this fashion, it would seem that the An-
 192 The Universe
dromeda galaxy and our own are not in the later stages of
evolution, but in the earlier, and are, in fact, rather young
galaxies.
The reverse-direction scheme of galactic evolution has
- Some observations in its favor. It postulates a steady
diminution of dust and gas in galaxies (which seems rea-
sonable); irregular galaxies such as the Magellanic Clouds
do indeed seem to be more. debris-filled than our own
definitely spiral Galaxy, while spiral galaxies are in turn
dustier than the elliptical galaxies.
To account for the large size of so many spheroidal
galaxies, it might be suggested that large galaxies (like
large stars) evolve more quickly rather than more slowly.
The largest, under the influence of a particularly strong
gravitational field, would more quickly have condensed into
. stars, more quickly formed spiral arms in more dust-filled
outer regions, more quickly dragged the arms inward and
rounded themselves out to form the giant spheroidal gal-
axies we now have. Smaller galaxies would linger in the
spiral form, and still smaller galaxies would stretch out the
early irregular phase of the evolutionary process.
Then, again, the possibility is advanced that evolution in
the ordinary sense—with a galaxy of one kind turning into
- a galaxy of another kind—may ‘not take place at all. Once
a galaxy has taken on its star-filled shape, it is suggested,
that shape is set, and the difference between galaxies
depends entirely on differences in the original protogalax-
ies, chiefly in the matter of the quantity of angular mo-
mentum present.
Suppose a protogalaxy happens to have very little angu-
lar momentum. It rotates slowly and does not flatten
much, if at all. Little matter would be lost through the
centrifugal effect, so that such a protogalaxy would retain
a maximum size. Stars would form and make up a vast
spheroidal galaxy.
If the protogalaxy happened to have a larger supply of
angular momentum, it would turn more rapidly, flatten
somewhat, leak away some matter at the equator, and end
up ‘as a somewhat smaller, somewhat flatter ellipsoidal
galaxy.
- Changes would end once stars formed and, depending
on the quantity of angular momentum, different amounts
of flattening would take place before star-formation froze
the shape. In general, the more flattened the galaxy, the
smaller.
If the protogalaxy had a particularly large amount of
Formation of star populations
uu
te
ee etn NE erst eh ep
i> ee 2 :
¥b
ae es
los, =
as * 7
:a » "
LT ORES a
Or 5
he oF v—_
aur
= 2

t ‘

194 The Universe a

angular momentum, a new factor would enter. Stars
would begin forming but not at a rate fast enough to wipe
out the dust and gas before the rapid rotation succeeded
in flattening the protogalaxy considerably further. In
short, the gas and dust would flatten into a disc, leaving
behind a spherical or ellipsoidal halo of stars. The stars left
behind would be the Population II stars.
The disc of gas would be thrown outward into the
extremities and concentrate there, and within it would be
formed the Population I stars. Thus the formation of
spiral galaxies and of Population I stars would be entirely
a matter of the rate of rotation of the original proto-
galaxy.
Whatever scheme of galactic evolution is considered—
whether one progresses from protogalaxy to spheroidal to
spiral, from protogalaxy to spiral to speroidal, or from
protogalaxy to spiral and spheroidal independently—one is
left uncertain as to the age of the galaxies generally and,
therefore, as to the age of the Universe.
Did the entire Universe form from a system of proto-
galaxies that all (about 15 eons ago) began to develop
into galaxies? If so, then, the Universe as a whole is 15
eons old.
Or did some parts of the Universe begin galactic de-
velopment at some times, and other parts at other times?
Is that portion of the Universe with which we are most
familiar 15 eons old, while other portions are older? Are
there some vast regions, undetectable to us, that are still
in the protogalaxy stage and that, in effect, have not been
born yet?
Such questions do not get a ready answer from consid-
erations of galactic evolution alone. We must approach
the problem, therefore, from another direction.
i ha iieois eal i <: vee nT oon?

CHAPTER 1 3

The Receding Galaxies

The Galactic Red Shift

A key step in the study of the galaxies was taken in

1912 when the galaxies were not yet recognized as such
but were considered as spiral nebulae within our own
system of stars. In that year, the American astronomer
Vesto Melvin Slipher (1875- ) measured the radial
velocity of what was then called the Andromeda Nebula
and found it to be approaching us (its spectral lines
showed a violet shift) at a speed of 200 kilometers per
second.
This was interesting but not unusual. Radial velocity of
over 4 hundred kilometers per second was high but not
distressingly so. (In fact, we know nowadays that part of
the velocity is not to be attributed to a true approach of
the Andromeda. The rotation of our Galaxy happens to be
carrying the Solar system toward Andromeda, just as
millions of years from now it will be carrying us away
from it. If the effect of this rotation is allowed for and the
motion of Andromeda is measured relative to the center
of the Galaxy, it is still found to be approaching us, but
only at a velocity of about 50 kilometers per second.)
By 1917, things seemed somewhat more puzzling.
Slipher had gone on to measure the radial velocity of a
total of fifteen spiral nebulae. Offhand, on the basis of
sheer chance, one might have expected half of them to be
approaching, half receding. Instead, two_were approaching
and thirteen were receding. Although this is not the most
likely situation, it is still not impossible. (If you toss fifteen
coins, you might expect a roughly equal number of heads
and tails, but chance is chance—you might get two heads
and thirteen tails.)
; 195
 196 > The Universe
More disturbing was the fact that the radial velocities
were as great as they were. Those nebulae, which were
receding, were doing so at an average velocity of 640
- kilometers per second. While 200 kilometers per second
could be swallowed, 640 kilometers per second was quite
difficult to accept. It was much greater than the radial
velocities of the ordinary stars of the galaxies.
_ The more Slipher measured the radial velocities of the
spiral nebulae, the more extreme the situation grew. Each
new measurement he made showed a recession, never an
approach, and with increasingly astonishing velocities.
When Hubble demonstrated, in the mid-1920’s, that the
objects Slipher was observing were really galaxies that
were located far outside our own Milky Way, this, in a
way, eased the problem. That one particular ‘class of
objects making up our Galaxy should have such large
radial velocities of recession, while all other_ objects had
small velocities that were often velocities of approach,
made no sense at all. If, on the other hand, the objects
showing the unusual velocities were also at unusual dis-
tances, things were better. If two properties of an object
are unusual, there may be a connection and each can ace
explain the ‘other.
The work on the radial.velocities of the galaxies was
now taken up by another American astronomer, Milton
La Salle Humason (1891- ). With the greatest care he
began taking photographic exposures of days at a time so
that fainter and fainter galaxies could record their spec-
_ tra. He discovered velocities of recession among the faint
galaxies that made earlier determinations seem most con-
servative. In 1928, he tested the radial velocity of a
galaxy called NGC 7619 and got a value of 3800 kilome- _
ters per second, and by 1936 he was clocking velocities of
40,000 kilometers per second, better than one-eighth the
speed of light—and they were always recessions.
Such velocities were so great that astronomers fell to
questioning the nature of the red shift (and occasionally
the matter is still brought into question). Must a red shift
necessarily imply that the source of light is receding? Or
could there be some alternate explanation that would
make it unnecessary for us to accept such huge veloci-
ties?
Could it be, for instance, that the light of the distant
galaxies is simply reddened by incredibly long passages
through the thin gas of intergalactic space. Undoubtedly,
but such reddening of light is not at all the same thing as
 Ct RO a ne a =
/

The Receding Galaxies 197

a red shift. Short-wave light mania be removed from the
spectrum by such scattering and the color of the nebula as
a whole would be reddened, but the spectral lines would
not be shifted at all. And it is to the spectral lines
particularly, not to the color generally, that the term
ted shift is applied: —
Another suggestion sometimes made is that the light, on
its unprecedentedly long journey from the galaxies, some- ~
how loses energy in transit. When light loses energy, this
loss is displayed in terms of an increase in wavelength so
that one would expect, under such conditions to obtain a
true red shift. We would thus be victimized into thinking
the galaxies were receding at unprecedented velocities
simply because we were studying what might be called
“tired light.”
This explanation does not explain, however, because no
one has yet offered a mechanism that would account for
the postulated loss in energy. There is no known reason
why light should lose energy simply because it was travel-
ing through a vacuum for a long time. Furthermore, if it
were’ indeed losing energy in this fashion, no one could
offer a reasonable explanation as to what became of that
energy. Then, too, if light grew tired over long distances,
‘it ought to grow very slightly tired over shorter distances.
Judging by the red shift related to the galaxies, the red
shift that would be produced by objects closer than the
galaxies ought to be detectable, but it is not.
In short, the red shift can, so far, only be explained by
supposing the galaxies to be receding from us. No alter-
nate explanation can square with all the facts and make
sense. Until an alternate explanation does, astronomers
will have no choice but to continue to accept the precipi-
tous recession of the galaxies as a fact.
_ Hubble, working along with Humason, was naturally
interested in the recession of the galaxies. He had been
painstakingly making estimates of the distance of the
galaxies in several ways. He had made use of Cepheids in
the closest (see page 105). For those too far away to
reveal any Cepheids, he made use of any stars he could
see by assuming these were supergiants that were, say, as
bright as S Doradus. If the galaxies were too far away for
any stars to be made out at all, then he made the assump-
tion that, on the whole, galaxies were approximately equal
in total luminosity and that the dimmer a galaxy was, the
more distant it was. Naturally, one applied the inverse
square law. If one galaxy was one-fourth as bright as
 198 The Universe

another, it was twice as far; if it was one-ninth as bright,

it was three times as far, and so on. :
Using such criteria of distance, Hubble then made use
-of the velocity determinations of Slipher and Humason to
show that, on the whole, the velocity of recession of the
galaxies increased proportionately with the distance of
those galaxies from us. If one galaxy was twice as far
from us as another, it receded from us at twice the
velocity; if it was three times as far from us as another, it
receded from us at three times the velocity, and so on.
This is “Hubble’s law.”
The most astonishing feature of Hubble’s law that the
velocity of recession of a galaxy is proportional to its
distance from’ us can best be expressed by the simple
question, “Why us?”
What magic is there about ourselves that causes a
galaxy to hasten from us, so to speak? And how does the
_ galaxy “know” how far it is from us and guide its steps —
accordingly? ;
Fortunately, the. same explanation that accounts for the
relationship between speed of recession and distance in the
first place, also accounts at once why-it should be distance
from us. ;
The explanation was suggested by the new view of the
Universe presented by Einstein.

Relativity

The new view was contained in Einstein’s general theo-

ry of relativity, first put forth in 1915. In it, Einstein
worked out a set of “field equations” which described the
overall properties of the Universe. In order to do this, he
assumed that although the Universe showed condensations
of matter here and there (planets, stars, galaxies), it could
be treated with approximate accuracy if it were regarded
as though it were uniformly filled with matter; that is, as
- though the matter actually in the Universe were spread
out evenly. (This is analogous to the manner in which,
despite the fact that the surface of the Earth is manifestly
bumpy and irregular, ancient man made the assumption
that all these irregularities were not really of importance
and that, on the whole they could be looked on as though
they were spread out evenly so that the world could be
considered flat. We do the same sort of thing today except
that we consider the world to be a epheve )
 The Receding Galaxies 199
Einstein further assumed that the properties of the
Universe were, generally speaking, the same everywhere.
On the basis of this assumption, the possible geometries of
the Universe were strictly limited. To see why, let us use
the surface of the Earth as an analogy.
_. Wherever we are on Earth, we feel essentially the
same. The directions up and down are the same; the pull
of gravity is about the same; the horizon is always at the
same distance, and is equally so in all directions (assuming
that we disregard local irregularities in the surface and
consider all matter spread out evenly).
There are three kinds of surfaces the Earth could have
that would yield identical properties of this sort every-
where. It could be planar (that is, flat); it could be
spherical; or it could have a much less familiar shape
called “pseudospherical.” Early man assumed the surface
to be flat because that was simplest, but observation even-
tually made it necessary to choose the spherical shape
instead.1
The choice of the spherical surface of the Earth over ©
the others has an important geometric consequence. The
spherical surface is the only one of the three that is finite.
A straight line on a flat surface, or the equivalent of a
straight line on a pseudospherical surface, would go on
forever. The equivalent of a straight line on a spherical
surface, however, would close in on itself. That is, if you
begin at some point on the Equator and walk east, you
will eventually return to your starting point although you
have never changed direction. You can walk on forever
without coming to any “end of the Earth,” but you will be
repeating your path endlessly. The surface Bt a scr is
finite, but unbounded.
The same situation can be applied to the uadiveese as a
whole, except that in the Universe we are dealing with a
volume, rather than with a surface, and that makes the
matter harder to visualize.
Still, consider a ray of light traveling through the Uni-
verse. To us it seems that a ray of light traveling through
a perfect vacuum and encountering no interfering energy
fields must travel in an absolutely straight line forever,
receding from its source at a constant rate. This is equiva-
lent to saying that the Universe has the properties one
would describe by means of Euclidean geomtery. We
1 Of course, the Earth is not exactly spherical, and therefore prop-
erties such as gravitational pucety at the surface do vary slightly
from place to place.
 200 The Universe — Pe
could call it a “flat Universe” even though it is a volume
rather than a surface.
But is the Universe really Euclidean or is that an
illusion born of the fact that we see so small a portion of
it? A small portion of the Earth’s surface looks flat to us,
too, and only very delicate measurements tell us that it is
actually gently curved in every direction.
But if the. Universe is not Euclidean, what can it be? If
We assume its general properties are the same everywhere,
we have the same choices we had in connection with the
Earth’s surface, if we consider it in terms of a ray of light.
There are two alternatives to the flatness, two varieties of
a non-Euclidean Universe.
The ray of light can move in a grand circle, as though
it were moving along the surface of a sphere. The -geome-
try of the Universe would then correspond to a system
first described by the German mathematician Georg Fried-
rich Bernhard Riemann (1826-1866) in 1854. A Rie-
mannian Universe must not be looked on as a simple
spherical Universe; it is more complicated. It is one in
which. three-dimensional space itself curves in every direc-
tion with a constant curvature. The Universe is a four-
dimensional ‘analogue of a sphere, a “hypersphere”—
something that is very difficult to represent or imagine,
accustomed as we are to thinking in three-dimensional
terms.
The ray of light can also move as though it were follow-
ing a pseudospherical surface-in all directions. The geome-
try of the Universe would then correspond to a system
first described by the Russian mathematician Nikolai Ivan-
ovich Lobachevski? (1793-1856) in 1829. -
The Riemannian Universe differs from the Euclidean
and Lobachevskian Universes in being finite. A ray of
- light- traveling through a Riemannian Universe curves
back on itself. It can go on forever but only by endlessly
repeating its path, like the Earth’s equator. The Riemanni-
an Universe is therefore finite but unbounded:
How can one choose among these possibilities? If we
could make a ray of: light travel over a long enough
distance, we could perhaps see whether it was actually
straight, or whether it was deviating from a straight line in
either a* Riemannian or Lobachevskian sense. The Uni-
2 The satirist Tom Lehrer has written a very funny song entitled
“Lobachevski.” Those who have heard and enjoyed the song may
think I have made a mistake, but. Lehrer made use of the name of
a real—and great—mathematician.
 The Receding Galaxies 201
verse deviates from the Euclidean so slightly (assuming it
deviates at all) that a ray of light long enough for the
purpose would be difficult to handle. Worse than that, we
would be stymied in our search for straightness by the fact
that our criterion for straightness is light itself.
If we have a long measuring rod and want to know if it
_is straight or not, we can hold it up endwise to our eyes
and sight along it. If it is not straight, we see it dip below |
the line of sight or bulge above it or bend to one side, and
even small deviations from straightness are quickly detect-
ed. But-what we are doing in this case is to make the
assumption that rays of light are traveling in an absolutely
straight line. Our assumption of the straightness of light is
so absolute that when light is reflected or refracted, noth-
'ing can convince our sense of sight that the straightness
has been violated. We see ourselves behind the mirror; we
see a stick bend sharply where it enters the water.
The decision among the possible Universes must therefore
‘be made more indirectly. Einstein chose the Riemannian
Universe and,. by 1917, had worked out its consequences,
trying to find some that would deviate measurably from
similar consequences in a Euclidean or Lobachevskian
Universe. (This can be considered as marking the be-
ginning of the modern science of cosmology.) For instance,
he showed that in a Riemannian Universe light would
lose energy in traveling against a. gravitational field, and
the Einstein shift he predicted was found in the light
radiated by. Sirius B (see page 171). He also predicted
that light rays would bend on passing a massive object, so
_that stars would seem to shift position slightly if their
light passed near the Sun. The positions of stars near the
Sun were measured during a total eclipse in 1919 and
compared with the positions at times when the Sun was
nowhere near, and again Einstein was borne out. In all
tests made of Hinstein’s general theory of relativity, the
theory has been borne out. Not one observation going
clearly against it has been made, and it is generally ac-
cepted among astronomers that the Universe as a whole
foHows a Riemannian geometry although one that devi-
ates so slightly from the Euclidean that under ordinary
circumstances Euclidean geometry is perfectly satisfac-
tory.
Einstein further pictured a Riemannian Universe that
was static, one that did not undergo an overall change.
The individual components within it might move around,
but the overall density of matter, if all were smoothed out
 202 - The Universe
evenly, would remain the same. Since, in Einstein’s view,
the curvature of the Universe (the extent to which it was
Riemannian) depended on the density, a ray of light
would travel in a perfect circle if uninterfered with.
In 1917, however, the Dutch astronomer Willem de
Sitter (1872-1934), who had been among the first to
accept relativity, offered another design of the Universe
“that would also fit Einstein’s field equations. This was a
Universe that was empty and constantly expanding. This
meant that the curvature of space was becoming constant-
‘ly less marked; the Universe was Riemannian but was
perpetually approaching closer and closer to the Euclidean
(which it would reach when expansion had grown infi-
nite). A ray of light in de Sitter’s model of the Universe-
would not travel in a circle Latsina perpetually esas a
spiral.
Furthermore, suppose one were (in imagination) ‘to
insert two particles in de Sitter’s expanding Universe.
These two particles would separate at once and would
continue separating as the space between them expanded
and continued to expand.
If a large number of particles were scattered through
such a Universe, the general expansion of the Universe
- would increase the distance from any one of them to any
other. If the distance between a given particle and its
nearest neighbor was 1 light-year to begin with, that
distance would be 2 light-years after a while, 3 light-years
after another while, and so on.
Now suppose an observer were on one of those parti-
cles, looking at all the others. In one particular direction
there would be a particle 1 light-year away, another one:
beyond it which would be a total of 2 light-years away,
still another 3 light-years away, and so on. After a centu-
Fy, let us say, the distance between any two neighboring
particles has increased to 2 light-years. By that time, then,
_ the observer standing on his particle and looking in the
_ Same direction as before would see the nearest particle 2
light-years away, the next 4 light-years away, the next 6
light-years away, and so on.
If that is the case, then the nearest particle has moved
outward from 1 light-year to 2 light-years and has receded
at the rate of 1 light-year per century. The second parti-
cle has moved from 2 to 4 light-years and has receded at
the rate of 2 light-years per century. The third particle has
moved from 3 light-years to 6 and has receded -at 3
_ light-years per century.
 The Receding Galaxies 203
All the particles you see in a particular direction are
receding from you and at a rate that is proportional to
their distance. Furthermore this is true in no matter which
direction you look. Does that make your particle special?
Not at all. It would not matter which particle you were
on. The effect would be the same on any one of them.
-Every independent particle in such a universe is receding
from every other particle at a rate proportional to the
distance between the two particles being considered.
De Sitter’s expanding Universe seemed quite superior to
Einstein’s static Universe from a theoretical standpoint. In
fact, in 1930, Eddington was able to demonstrate that
even if Einstein’s static Universe could be assumed to exist -
it would be unstable, like a cone balanced on its point. If-
it began to expand ever so slightly for any reason, it
would continue to expand forever; and, for that matter, if
it began to contract, it would continue to contract indefi- _
nitely.

Clusters of Galaxies

Hubble’s law seems to show the existence, then, in

real-Universe terms, of de Sitter’s theoretical model of an
expanding Universe. The galaxies are receding from each ©
‘other not because they are moving individually, but be-
cause all of space is expanding. As a result of this expan- ~
sion, the velocity of recession of an individual galaxy is
proportional to its distance from us. Moreover, our
Galaxy seems the central point of the universal recession
only because we happen to be observing the Universe
from that point. Had we been in the Andromeda galaxy,
or in any other galaxy, the same phenomenon would have
been observed. The galaxy in which we were would seem
the central galaxy in every case.
Of course, one might say that if the Universe were
really expanding, then every galaxy without exception
ought to be receding from every other galaxy, and that
_while this is almost true it is not completely true, and
“almost true” wins no prizes. For instance, the Androme-
da galaxy is not receding from us; it is approaching us. It
is doing so quite slowly, to be sure, but it is approaching
and surely that upsets the entire notion of an expanding
Universe.
Not at all. In presenting a model of the Universe, one is
forced to make simplifying assumptions or else the com-
plexities of the model become too great to be analyzed or
 204 - The Universe —

described. In the de Sitter model, for instance, it is as-

sumed that the test particlesintroduced exert no force on
each other, but accompany the expansive movement of
the Universe without resistance.
But this is not so in reality. There are long-range forces
in the Universe that are capable of making themselves felt
over great distances as soon as matter is considered to
exist in it. These long distance forces are of two kinds.
There are electromagnetic fields and gravitational fields.
The electromagnetic field gives rise to forces of two kinds,
“an attractive one and a repulsive one, and the two effects
usually balance on a large scale. We can therefore ignore
the electromagnetic field in the Universe generally.
_ Not so in the case of the gravitational field. That gives
rise to only one kind of force, an attractive one. Any two
objects in a Universe, even in an expanding Universe, will
attract each other.® The closer two objects are, the strong-
er the gravitational attraction between them and the
more likely they are to cling together, so to speak, against
the separating influence of the expanding Universe.
The expanding Universe does not, for instance, separate
the components of the Solar system from each other, or
separate the stars within a Galaxy from each other. It will
not suffice to separate two or more galaxies which are
close enough together to be caught in the grip of a
sufficiently strong mutual gravitational field.
In short, the “independent particles” that recede from
~ each other in an expanding Universe ‘are, in our own real
Universe, not necessarily individual galaxies, but individu-
al groups or clusters of galaxies.
Such clusters of galaxies clearly exist. There are a
number of cases of two or more galaxies that visibly
interact, that are enclosed in common halos or are con-
nected by luminous threads. The cause of the interaction
may be gravitational or it may be electromagnetic, but the
interaction is there in either case and it is quite logical to
‘treat such galaxies as a unit, from the overall standpoint
of the Universe.
Our own Galaxy has the two Magellanic Clouds as-
obvious satellite galaxies that are firmly in our grip and
are not likely to be separated from us by the expanding
Universe. In the same way, the Andromeda galaxy (M31)
There are special cases where this is not so as, for instance, if
one object is a hollow sphere and the second object is within the
hollow. These special cases are not important, astronomically.
 BR re Sp

|The Receding Galaxies _ 205

has two small galaxies (M32 and M33) which are satel-
lites to it.
In fact, the Andromeda galaxy and ovr own Galaxy
may be considered as the two giant members of a group
made up of some two dozen members altogether. This
group is the “Local Group,” and no doubt it will maintain
its identity, at least over some long period, against the
expansive pull of the Universe.
The motions of the member galaxies of the Local
Group, relative to each other, do not reflect the general
expansion of the Universe, but, rather, local gravitational
forces. It is for this reason that the Andromeda galaxy
happens, in this particular eon of time, to be approaching
us.
There are hundreds of clusters of galaxies visible in the
sky. They are obviously clusters because of the close
propinquity of the individual members in space and of the
similar luminosity of-the larger members. Some of the
clusters are enormous. There is one cluster in the constel-
lation Coma Berenices made up of about 10,000 individu-
al galaxies.
Such clusters are very useful in estimating distances. If
galaxies are studied individually, it is not entirely safe to

- Local group W-L-MA

Se eo
Andromeda
e

=
®> Ursa Minor

¥,200,000 light-years
 206 The Universe .
suppose that the degree of dimness is determined only by
distance. There are giant galaxies and dwarf galaxies just
as there are giant stars and dwarf stars. The Andromeda
galaxy and its two satellite galaxies are at the same dis-
tance from us, yet Andromeda is much brighter than
either of the other two, simply because it is-a giant where
the others are dwarfs.
Of course, one could argue that, over the long run, such
differences in size average out and that, viewed as a whole
and from a generous statistical standpoint, dimness can ‘be
equated with distance. This is true, but the general equa-
tion of dimness with distance suffers from irregularities as
a result. ; é
In clusters, however, one can assume that the brightest
members are giant galaxies equivalent to the Andromeda
or to our own and that they thtrefore have a total
luminosity equivalent to an absolute magnitude of —19 or
—20. There is still the chance of differences and irregular-
ities from one cluster to the next, but these are much
smaller, astronomers believe, than would be the case if
one were dealing with individual galaxies.
Astronomers therefore feel more confident in judging
the distance of clusters of galaxies from dimness alone,
and then comparing that distance with the value of the
red shift to see how Hubble’s law holds up. (It does.) A
cluster of galaxies in the constellation Virgo is estimated
by this method to be nearly twenty times. as far away as
the Andromeda galaxy, ‘and other clusters are located at
nearly a thousand times the distance of the Andromeda.
 CHAPTER 14 E Sabpeio? 3

The Observable Universe

Olbers’ Paradox Again

The scale of the Universe is seen to be so large that it

dwarfs even the first consideration of the distances of
nearer galaxies such as Andromeda. When the distance of,
the Andromeda galaxy was first determined and expres-
sions such as “hundreds of thousands of light-years” came
into vogue, that was considered mind-stretching indeed.
Within a decade or so, however, it became obvious that
Andromeda was merely next door. It is even, as I have «
said, part of the Local Group, part of a system of which
we are also part:
One has to ask again where it ends. Time and again,
man has had to expand his vision to take in larger and -
larger groups. Small nonluminous objects group together
about a star to form a planetary system. Stars group
. together to form a simple multiple-star system, or larger
open clusters, or even larger globular clusters, or still
larger galaxies. Galaxies group together to form clusters
of galaxies. Can these group together still more extensively
to form clusters of ‘clusters of galaxies? De Vaucouleurs
suggests that this may be so, that there are signs that a
“supergalaxy” may exist in which the Local Group is but
a small item. If his analysis is correct, then we are tens of
millions of light-years from the center of such a super-
galaxy—and beyond it in every direction would be other
supergalaxies.
And would there not be clusters of supergalaxies and
clusters of clusters of supergalaxies and so on? What
would be the end? Would there have to be an end at ail?
May we not be facing an infinite Universe?
To be sure, if we accept Einstein’s theory of relativity
207
 OR he * ¢ \ A 4a ree t <3
‘ : sr rial * 1 <a

208 The Universe

then his Riemannian Universe must have a finite volume.
Even if it is expanding, that finite volume, although con-
stantly growing, remains finite.
And yet it is sometimes argued that even though the
Universe may have a finite volume, it might be able to
hold an infinite number of galaxies. If that is so, then the
system of clusters of clusters of clusters of clusters of
galaxies may continue onward into greater and greater
complications without end..
But if we are going to consider the possibility of an
infinite number of galaxies, do we not again run into
Olbers’ paradox (see page 56)? Will not the existence of
an infinite number of galaxies in every direction supply the
‘Earth with an infinite amount of light? And from the fact
that the Earth does not receive an infinite amount of light |
must we not argue that the number of galaxies must be
finite?

Supergalaxies

We must admit a finite number of galaxies, indeed, if

the Riemannian Universe were a static one, as Einstein
had first proposed. In such a static Universe, the argument
I will later advance in favor of the possibility of an infinite
number of galaxies would not hold. In such a sstatic .
Universe, both volume and the number of galaxies would
be finite and Olbers’ paradox would pose no problem. ;
But we seem to be living in an expanding Universe, and .
here the argument for an infinite number of galaxies can
hold. What now? How do we get out from under Olbers’
paradox.
 The Observable Universe 209°
' In an expanding Universe, in which the galaxies recede
steadily from each other, there is a new factor which must
be taken into account, one which would not be present if
the Universe were static and the galaxies maintained
themselves at distances from each other that did not (on
the average) change. That new factor is the red shift.
In an expanding Universe, the light from the galaxies is
weakened and enfeebled by the red shift. The further the
galaxy the stronger the red shift and the greater the
weakening of the total radiational energy reaching us. ©
If we allow for the steady weakening of the radiation
reaching us from greater and greater distances, then the
total amount of radiation that impinges on the Earth can
be shown to reach some finite value, and one that is not
excessively large, even though the number of galaxies was .
infinitely large. Olbers’ paradox would no longer stand in
the way of an infinite Universe (in terms of numbers of -
galaxies) if that Universe were expanding. ~
This may sound impossible. It may seem to you that if
every one of an infinite number of galaxies contributes a
little radiation, then, no matter how little that might be,
the sum total must be infinite. This is equivalent to saying,
that the sum of an infinite series of numbers, however
small the individual numbers might be, must be infinite.
This may sound logical, but it is wrong just the same, as
can be demonstrated very easily.
Consider the series of numbers: 1, 1/2, 1/4, 1/8, 1/16,
1/32, 1/64 . ... Each number is half the number before, .
and there are an infinite number of members of this
series. No matter how far the series is extended it can
always be extended farther by writing down half the last
number written, and then one more by writing down half
the additional number written, and so on indefinitely.
One might suppose, then, that the sum of such an
infinite series must be infinite—but try it. The first number
taken by itself is 1; the sum of the first two numbers is 112;
the sum of the first three is 134; the sum of the first four is
1%. If you continue onward, you will satisfy yourself soon
enough that although the sum continually increases. as
more and more members of the series are taken, that sum
can never reach 2. It approaches 2 more and more closely
but never quite reaches it. The sum of the series 1+1/2+
1/4+1/8+1/16+1/32+1/64 ... is said to approach
2. This is an example of a “converging series,” a series
with an infinite number of members but a finite sum.
In the original description of Olbers’ paradox, I ex-
 M0 - The Universe
~ plained that in an infinite Universe of the type envisaged
by Olbers each shell of space delivers an equal amount of
‘light to Earth. If the light of one shell is considered 1,
then the light of all shells is 1+1+1+1, and so on
forever. This is a “diverging series” and obviously has an
infinite sum. It is that infinite sum that is the nub of the
paradox.
If the red shift is taken into account, however, each
shell (working outward) delivers less light than the one
before. If a series is set up, it converges and yields a finite
sum. In Einstein’s Reimannian Universe, whether static or
expanding, we can therefore forget about Olber’s paradox.
It presents no problem.
But. now something else comes up. If we are going to
consider the red shift and the consequent progressive
weakening of the energy of radiation from more and more
distant galaxies, we must ask ourselves not how large the
Universe is, but how much of the Universe we can ob-
serve.
Man has taken it for cranted that if he improves his
instruments and refines his techniques, if he builds bigger
telescopes and better spectroscopes and more delicate
cameras, he will be able to see farther and farther out into
space—but is this so? If the radiation from distant galax-
ies becomes feebler and feebler, is there a point at which
the radiation becomes so feeble that no instrument, how-
ever close to perfection it may be, can detect it?
If this is so, then there is some limit, in principle, to the
size of the Universe we can observe; there is some- outer-
most boundary we cannot peer beyond. This boundary
--would delimit what we might term the “observable Uni-
’ -verse.”?
To find out if there is such an inviolable limit and
where it might be, let us return to Hubble’s law.

Hubble’s Constant

Hubble’s law states that the velocity of recession of a

galaxy is directly proportional to its distance from us. This
means that if its distance from us is multiplied by some
definite quantity (a quantity called “Hubble’s constant”),
we will get the velocity of recession.
Suppose we represent the distance as so many millions
of light-years and call that D. We can represent the
velocity of recession as so many miles per second and call
 TAP —.-* & Ty
pg) Son tae © “ i .
~

Whe: Oliecryable “Universe 211

that V. For the proportionality constant we can write k.
now we can express Hubble’s law as follows: .
V=kD
where kD is the mathematical way of saying that k must
be multiplied by D. ‘
Everything depends on the value of k, so let us rewrite
the equation above, making use of simple algebraic tech-
niques, and get:
=V/D
which tells us that k must be equal to the velocity of re-
cession of a galaxy (in miles per second) divided by its
distance (in millions of light-years). If we have good
reliable figures for both the distance and the velocity of
recession of a single distant galaxy or group of galaxies,
then we can solve for k. That value of k will, if Hubble is
correct, hold for all galaxies.
Consider the Virgo cluster, for instance. The red shift
of its components shows it to be receding from us at a
- velocity of 710 miles per second. By comparing the
brightness of the brightest galaxies of the cluster with that
of the Andromeda galaxy, the cluster turns out to be 16.5
times as far distant as the Andromeda. If the Andromeda
galaxy is 800,000 light-years away (according to the peri-
od-luminosity relationship of the Cepheids, see page 57);
the Virgo cluster must be 16.5 X 800,000 or 13,000,000
light-years away.
In order to find the value of k, then, we must divide
Virgo’s velocity of recession in miles per second by its
distance from us in millions of light-years. We find that
k=710/13, or about 55.
We can expect then that a galaxy that is 1,000,000
light-years away should be receding from us at a velocity
of 55 miles per second; one that is 2,000,000 light-years
away should be receding from us at a velocity of 110
miles per second; one that is 10,000,000 light-years away
should be receding from us at a velocity of 550 miles per
second, and so on.
Is there any limit to this, supposing Hubble’s law to
hold exactly at all distances? Mathematically, there is not.
A galaxy that is 1000 million light-years away should
recede at a velocity of 55,000 miles per second; one that
is 1,000,000 million light-years away should recede from
us at a velocity of 55,000,000 miles per second, and so
on.
Physically, however, there is a limit. Einstein’s “special
theory of relativity” (advanced in 1905, ten years before
 212 The Universe
- the more all-encompassing general theory) makes it neces-
sary to suppose that the maximum velocity that.can be
measured relative to one’s self is the velocity of light in a
vacuum. This is equal to 186,282 miles per second (or
299,776 kilometers per second).
; At a certain distance, a galaxy must be receding from
us at that velocity and, according to the viewpoint of
relativity, that represents an absolute, insurmountable lim- .
it. We can deal’ with nothing moving more rapidly than
that, and therefore nothing more distant from us than
that.
There might be all sorts of arguments concerning
whether or not there were galaxies farther away, but that
would all be irrelevant. It would not matter if there were
‘or not. The point is that once we reach a point so distant
from ourselves that.a galaxy recedes at the velocity of
light in a vacuum, light from that galaxy cannot reach us.
The Doppler-Fizeau principle stretches out each wave--
length infinitely and therefore reduces its energy to zero.
Nothing reaches us from a galaxy that distant. No light,
no radiation of any sort, no neutrinos, no gravitational
influence. Nothing. -
Even if it were possible to conceive of something be-
yond that limit of distance, it would be something that
would remain forever undetectable to us—not because of
the imperfection of our instruments but because of the
nature and design of the Universe. Consequently, we can
forget all about any talk of a Universe of infinite dimen-
sions. We must talk, instead, about an observable Universe
that is finite in diameter and volume.
It remains only to tell what the diameter of the observ-
able Universe is. To do this, let us begin with the mathe-
_ matical equation representing Hubble’s law, V=kD, and
solve for D (distance fromi us). This changés the equation
to: ;
D=V/k ;
What we want is the distance from us that represents a
velocity of recession equal to the speed of light. We set V
equal to the speed of light or 186,282 miles per second,
and let k equal 55. In that case D (in millions of light-
years) equals 186,282/55 or 3400. This means that the
limit-of the observable Universe seems to be 3400 million
light-years—or 3.4 billion light-years—in all directions
from us.
_To put it more concisely, the observable Universe is,
apparently, a sphere with ourselves at the center and a
pe
 eae
‘The Observable Universe 213
radius of 3.4 billion light-years and, consequently, a diam-
eter of 6.8 billion light-years. The limit of the observable
Universe is, in other words 4250 times as far from us as
’ the Andromeda galaxy is, according to this analysis.
These are enormous dimensions and they certainly. .
sound like a fitting climax to man’s long search for limits,
a search that began with his consideration of the horizon a
mere few miles away.
And yet—there were problems. There was something
wrong with the scale of the Universe as it was- worked out
by the 1940’s through the use of Hubble’s constant.

The Cepheid Yardstick Revised

The distances worked out for the far-off galaxies were

based on the comparison of their apparent brightness with
that of nearby ones whose distance was in turn deter-
mined by the Cepheid yardstick. And of the nearby ones,
the distance determination was most certain and reliable,
it seemed, for the Andromeda galaxy. If the distance of
the Andromeda galaxy was wrong, then all the distances
were wrong; the entire scale of the Universe was wrong.
And by 1950, the uncomfortable feeling was growing
that the determination of the distance of the Andromeda
galaxy was indeed in error. If Andromeda was at a dis-
tance of 800,000 light-years, as the Cepheid yardstick
seemed to indicate, certain peculiarities showed up. For
one thing, the Andromeda galaxy seemed to be consider-
ably smaller than our own Galaxy, perhaps onlya quarter
as large. There was no crime in this, taken alone, but all
the galaxies. whose size could be estimated seemed to be
considerably smaller than our Galaxy.
One might argue that some one particular galaxy had to
be larger than all the others, and we just happen to be
living in that one. And yet why should our Galaxy be so
much larger?
Whatever process formed the galaxies produced them in
a wide range of sizes. No one could argue with the fact
that the Galaxy was far larger than the Magellanic
Clouds, and that the Andromeda galaxy was far larger
than its satellites, M32 or M33. But there were numerous
members representing every portion of the range; no sin-
gle galaxy was unique in size, either at the large or the
small end of the scale—except our own. Our Galaxy stood
alone, far larger than the rest.
Furthermore, our Galaxy was the wrong shape to be so
 214 ‘ The Universe
large. Where different galaxies could be compared direct-
ly, it was always the elliptical galaxies—particularly the
spheroidal ones classified as EQ—that were the giants.
Why should the largest of all, our own Galaxy, be a
spiral? - 708
What was worse still was that our Galaxy was not only
larger all together, but that its component parts were
larger and brighter than the analogous component parts in
other galaxies such as the Andromeda.
For instance, the Andromeda galaxy has a halo of
globular clusters about its center, just as our own Galaxy
has (see page 71). The number of giobular clusters, their
appearance, and their distribution are all very similar in
- both cases. One could, however, begin with the apparent
‘brightness of the individual globular clusters of Androme-
da and, considering them to be at a-distance of 800,000
light-years, work out what their actual luminosity must be..
It turns out that the globular clusters of Andromeda are
- less than a quarter as bright, on the average, as our own
globular clusters are, and only about half as wide in
diameter. Even individual stars showed the same effect.
Ordinary novae, appearing in Andromeda, usually attained
considerably. less luminosity than novae in our own Galaxy
did, allowing for an 800,000-light-year distance. >
To suppose that our own Galaxy was not only a giant
among galaxies, but that it was made up of globular
clusters that were giants among globular clusters, ‘and of
stars that were giants among stars, was asking too much.
It looked almost as though we were looking at the An-
‘dromeda galaxy (and, therefore, at the other galaxies,
too) through a diminishing glass that was reducing every-
thing about it in size. Since everything about the An-
dromeda galaxy was determined on the basis of its distance,
the question had to arise as to whether that distance
might not be wrong. Since the distance, as accepted in
1950, depended, in turn, on the Cepheid yardstick, the
question had to arise as to whether there might not be
something wrong with the Cepheid yardstick.
Baade, in the early 1950's, addressed himself to this
question. He reasoned that the stars of the Magellanic
Clouds and of the globular clusters of our own Galaxy
were of Population II (see page 190), the generally small-
er and stabler of the two populations. It had been Popula-
tion II Cepheids, therefore, that had been used to set up
the period-luminosity law in the first place, and it had
 The Observable Universe 215

been those which had been used to determine the scale of

our Galaxy and the distance of the Magellanic Clouds.
However, the Cepheids that had been used to deter--
mine the distance of the Andromeda galaxy (and there-
fore, indirectly, of all the far-off galaxies) had been those
of the spiral arms of the Andromeda because the giant
blue-white members of the Population I stars in those
arms had been the most easily seen at Andromeda’s vast
distance. Could it be that the Population I Cepheids of the
spiral arms of Andromeda did not follow the same period-
luminosity law followed by the Population II Cepheids
that Leavitt and Shapley had worked with?
Certainly, there seemed to be considerable difference
between the two types of Cepheids. The Population II
Cepheids included a considerable number with particularly
short periods running from an hour and a half to a day,
whereas such periods were quite rare among the Popula-
tion I Cepheids, where periods of several days to several
weeks were much more common. Secondly the Population
If Cepheids were, on the whole, smaller and dimmer than
the Population I Cepheids. This second difference was
masked by the fact that the Population I Cepheids in our
own Galaxy, located in the dusty spiral arms, were
dimmed and reddened by the interstellar dust by an
amount that had not been properly allowed for.
_ The Population II Cepheids are, in fact, even in posses-
sion of a special name because of the distinctiveness of
their properties compared with other variables. They are
the “RR Lyrae stars,” named for RR Lyrae, the first (and
nearly the brightest) of these variables to be studied.
Because RR Lyrae stars are regularly found in globular
clusters, they are sometimes called “cluster-type varia-
bles.”
Baade carefully studied the Population II Cepheids and
the Population I Cepheids separately, and in September
1952, announced that the period-luminosity law as worked
out by Leavitt and Shapley, applied only to the Population
II variety. The distance of the Magellanic Clouds and the
dimensions, of our Galaxy were therefore correct. The
Population I Cepheids, however, followed a somewhat
different relationship and for a given periodicity were a
magnitude or two brighter than would have been expected
from the ordinary relationship used by Shapley.
Let us see what this means. Suppose we observe a
distant Cepheid with a period that yields us an absolute
magnitude of —1. This means that if it were 32.5 light-
 216 ~—: _. The Universe
years (10 parsecs) from us, it would appear to have a
magnitude of —1. To be reduced from ~—1 to its actual
magnitude of something like 20, it would have to be some
24,000 times more distant than 32.5 light-years—or 800,-
000 light-years away.
But suppose it turned out that, using Baade’s new period-
luminosity scale for Population I Cepheids, the particular
Cepheid under study had an absolute magnitude of —3
tather than —1. It would then be more than six times as -
bright as had been thought. In order to reduce such a
six-times-brighter star to a magnitude of about 20, it
would have to be placed correspondingly-farther off—S8,-
000 times more distant than 32.5 light-years, or nearly —
2,000,000 light-years away.

The two populations of Cepheids

0.1 03:05 J 3.5 °10 30 50 100

period in days

_ By using the revised Cepheid yardstick and adding some

additional refinements that now appear necessary, the An-
dromeda galaxy is today thought to be some 2,300,000
light-years away. All other galaxies beyond the Androme-
da must be moved correspondingly. :
This removed, at once, all the uncomfortable unique-
ness of our Galaxy. If Andromeda is 2,300,000 light-
oh (eee ea ee kee.

-'The Observable Universe 217

years away (rather than 800 :000) and still appears as
large and as bright as it does in a telescope, it might be
much larger and brighter in actual fact than had been
. supposed in the days when the shorter distance was ac-
‘cepted.
Nowadays, the Andromeda galaxy is accepted as being
somewhat Jarger than our Galaxy. The Andromeda con-_
tains perhaps 200,000,000,000 stars. Moreover, its globu-
lar clusters, which are also farther away than had been
thought, are now seen to be larger and brighter than they
had been considered—as large and as bright, in fact, as
our own globular clusters. The novae in the Andromeda
are also as large and as bright as those in our own
Galaxy: Furthermore, all other galaxies are now seen to
be larger and brighter than had been thought, and many
of the spirals rival our Galaxy in size, while some
spheroidal galaxies may be ten to thirty times as large.
Our Galaxy remains a giant galaxy, but it is no longer
unique, no longer a one-of-a-kind monster.
Since this new scale of distance has removed the most
serious peculiarities from the galactic scene, astronomers
are hopeful that they now have the scale. about right.
Certainly in the years that have passed since Baade’s
correction, nothing has happened to shake this faith. In
fact, since Baade’s death in 1960, astronomers such as the
“Russian-American Sergei Illarionovich Gaposchkin (1898-_
) have continued analyzing the photographs of An-
dromeda taken by Baade, using the 200-inch telescope,
and have confirmed his work completely.
The new scale of distance has not, of course, affected
_ the red-shift determinations. These determinations are in-
dependent of distance. The Virgo cluster of galaxies is
receding from us at a rate of 710 miles per second
whatever distance we determine for it. From the bright-
ness of its brighter members, as compared with the bright-
ness of the Andromeda galaxy, it is still 16.5 times as far.
away as the Andromeda.
But now that the accepted distance of Andromeda has
been ‘tripled, so must the accepted distance of the Virgo
cluster. It must now be considered at a distance of 2,300,-
000 X 16.5 light-years or something like 38,000,000 light-
years away, rather than merely 13,000,000.
To determine Hubble’s constant, we divided the velocity
of recession of a galaxy or-.cluster of galaxies by the
number of millions of light-years it is distant from us.
Instead of dividing 710 by 13, we must now divide it by
 Bi yy

218 The Universe

38, so that Hubble’s constant comes out to be 18.5_rather
than 55. If anything, this stilt probably errs on the side of
conservatism. Let us therefore set the value of Hubble’s
constant at 15.
To determine the distance at which a galaxy must be
receding at the speed of light, let us once again use the
equation: D=V/k, setting V equal to 186,282, and k, this
time, at 15. It turns out that D equals 12,500 and we can
therefore say that a galaxy at a distance of 12,500 million
light-years, or 12.5 billion light-years, can no longer be
detected. That is the limit of the observable Universe, or
the “Hubble radius.”
To put this another way, we can say that the diameter ‘*
of the sphere of the observable Universe (with ourselves
at the center) is 25 billion light-years—a diameter nearly
four times that thought correct as late as 1950. :
 CHAPTER 15

The Beginning of the

Universe

The Big Bang

The change in the scale-of distance of the Universe did

more than remove the anomaly of our Galaxy’s apparent
super-gianthood. It greatly lessened an even more serious
discrepancy.
In the second quarter of the twentieth century, astro-
physicists and geologists once again disagreed on the age _
of the Earth, as they had done nearly a century earlier in
Helmholtz’s day (see page 119).
Again the discrepancy arose over a phenomenon that
seemed to cause no trouble in the present and in the
future, but raised serious difficulties when extrapolated
into the past. In Helmholtz’s day, it had been the supposed
contraction of the Sun; in Hubble’s day, it was the expan-
sion of the Universe.
If one tries to look forward in time, then, and if one
accepts the fact that the galaxies will continue indefinitely ~
to recede from each other in the present manfier, no
insuperable difficulties arise. Every galaxy outside our
Local Group will continue to recede at a regularly in-
creasing rate, matching its regularly increasing distance.
The galaxies will grow dimmer and dimmer, both because
of their increasing distances and because of their increas-
ingly pronounced red shifts and consequent decreasing
light energies. Eventually all will approach the limit of the
observable Universe and be lost to us. The observable
Universe will then consist only of our Local Group.
This sounds like a lonely future, but we will lose only
objects not visible to the naked eye, objects of whose
219
sheet. Se
 2 08 Se
“tae
ae

220 _ The Universe

existence and true nature we only became aware of in the
last fifty years. The loss, therefore, is not a great one to
non-astronomers. Furthermore, it will not happen for a
long time—a hundred eons or more—and by that time,
events more immediately affecting us will have occurred.
Our Sun will have become a white dwarf and our plane-
tary system will be uninhabitable, even if we suppose it to
have survived the Sun’s red giant stage. All stars larger’
and brighter than. the red dwarfs will certainly be white
dwarfs, and all the galaxies will be in extreme old age.
New stars may have formed, but a hundred eons ffom
now the supply of dust and gas may have sunk to minimal
values and such new stars may be few indeed. Further-
more, the final new stars may be formed out of gas so
charged with complex atoms (which will have been spread |
through space by the hundreds of millions of supernovae
that will have exploded in the interval) that they will have
abnormally low hydrogen supplies and will be particularly
short-lived.
Yet if this future sounds grim, it does not seriously
. conflict with any accepted scientific beliefs and poses no
serious. dilemma for astronomers. The Universe cannot be
expected to respect human emotions. It can age and die
without regard for man’s regrets, and its large components
can continue to recede from each other in eternal expan-
sion even after the galaxies have flickered down to white
dwarf cinders.
But suppose we look backward in time. Suppose we run
the expansion of the Universe in reverse, as though it were
a movie film..In that case, we must picture the various
galaxies approaching each other at known rates, and this
time the process cannot be continued eternally. Eventual-
ly, the galaxies must meet. If Hubble’s law holds, so that
every galaxy moves inward at a velocity proportional to
its distance from any particular galaxy (such as-our own)
used as standard, then the Universe generally must be
looked on as contracting and all the galaxies must meet
simultaneously at a point.
At some particular time in the past, then, all the matter
and energy of the Universe must have existed in one large
lump. At that time in the past (“zero-time”) the Universe
could not possibly be as it appears today; the Universe-
we-know could only have existed since this zero-time and
this zero-time can in fact be considered the beginning of
our Universe. .
It is possible to calculate this zero-time of the Universe —
 ae eo a “aa ites ie

The Beginning.-of the Universe 221

‘from the distances separating the galaxies now and the.
rate at which the Universe is now expanding. According
to the scale of distances accepted from 1925 to 1952, this
zero-time must have been approximately 2,000,000,000
years ago.
Two eons is a long time; it is certainly longer than the
twenty million years allotted the Earth by Helmholtz.
However, 2 eons is still not long enough for the geologists
and their. dismay was great. What was the use of saying
that the Universe existed as a single glob of matter 2 eons —
. ago and that all the galaxies had formed since then, when
the Earth itself was found to be more than twice as old as -
that on the basis of uranium-lead determinations o page
123).
The Earth simply could not be twice as old as the
Universe. Something was seriously wrong either with the
uranium-lead ratios or with Hubble’s constant.
It was a standoff until Baade’s work showed that the
trouble lay with the scale of distance of the Universe and
that this had previously given astronomers far too high a
value for Hubble’s constant. Once again the geologists
were right and the astronomers had been wrong. The new
scale of distance of the Universe made it appear that
zero-time was more like 13 eons in the past.
If anything, this figure is still too small; for astrono-
mers, in measuring the age of star-clusters, sometimes ~
come up with ages of 10-to 25 eons. However, no one
pretends that the determination of Hubble’s constant is
beyond correction even today, or that the methods for
_determining the age of star-clusters are better than ap-
proximate. A quarrel over the exact point of zero-time
would therefore be rather premature, and it is enough for
the present to say that the zero-time, if it exists, lies at
least 15 eons in the past and that there has been time
during the life of the Universe since then for the Solar
system to have formed.
But what was it that happened at zero-time? The first to
consider that point effectively was the Belgian astronomer
Georges Edward Lemaitre (1894-. ). In 1927, he
suggested that at zero-time, the matter and energy of the ©
Universe were actually and literally squashed together
into one huge mass. He called it the “cosmic egg” because
out of it, the “cosmos” (a synonym for “universe”) was
formed.
The cosmic egg was unstable and exploded in what we ©
can only imagine to have been the most gigantic and.
 222 _ The Universe
catastrophic explosion of all time, for the fragments of
that explosion became the galaxies, which were sent hur- .
tling out in all directions. The effects of that explosion are
still with us, for we see it as the recession of the galaxies:
and clusters of galaxies from each other.
If the different fragments of the cosmic egg were hur-
tled outward at different velocities (depending on where in
the egg’ the fragment was originally located, and how
much it was slowed down by collisions with other frag-
ments), those that ended with high velocity would natural-
ly gain constantly on those that ended with low velocities.
This, Lemaitre maintained, would give rise to the situa-
tion we now experience: one in which the galaxies are
receding from each other, with the rate of recession pro-
portional to distance. (It is also possible that even if the
cosmic egg had no angular momentum, some of its frag-
ments would pick up clockwise angular momentum as a
result of the explosion and others counterclockwise angular
momentum, the sum adding up to zero.)
The Lemaitre model of the Universe is a physical .
analogue of Sitter’s theoretical model. Sitter’s Universe
expanded simply because that fit a’ set of equations
worked out: by Einstein. The Lemaitre model, on the
other hand, expanded in consequence of a physical event;
an explosion that differed in size, but not in nature, from
that of a firecracker on Earth. The Lemaitre model is
easily grasped; it is concrete, dramatic, and seems famil-
iar, Eddington adopted and popularized it and since his
time, the Russian-American astrophysicist George Gamow
(1904- ) has upheld it enthusiastically. With refer-
ence to that vast initial explosion of the cosmic egg,
Gamow termed the Lemaitre model of the Universe, the ~
“big bang theory,” although it might less dramatically be
- Called the “exploding Universe theory.”
Naturally, one is curious as to the nature of the cosmic
egg. What was it made of? What were its properties?
Perhaps we can get an idea if we try to observe the
Universe (in imagination) running forward and backward
in time. Right now the Universe seems to be roughly 90
percent hydrogen, 9 percent helium, and 1 percent more
complex atoms. As the Universe runs forward in time,
hydrogen continually fuses to helium, and helium to still
more complicated atoms within stellar cores (see page
164). If we run the Universe backward, the quantity of
helium and more complex atoms decreases and the quanti-
ty of hydrogen increases. As we approach zero-time, then,
 eM eR eet ; Po
The Beginning of the Universe 223
we should expect the Universe to consist entirely, or
_almost entirely of hydrogen.
~ . But the matter and energy of the Universe are compact-
ing themselves as we view matters backward in time. At
zero-time, the hydrogen that exists must ultimately be
compressed; all the particles composing it are pressed
together as hard as they can be.
The hydrogen atom is made up of two particles only, a
central proton carrying a positive electric charge and an
outer electron carrying a negative electric charge. As long
as these exist separately, there is a limit to how com-
pressed a mass of hydrogen can be. If some critical |
pressure is surpassed, however, the electrons and protons
may be considered as squashing together to form a mass
of electrically uncharged particles called neutrons.
Such a mass of ultimately compressed neutrons is some-
The BigBang Theory
 224 | The Universe
times called “neutronium” (although Gamow uses the
term “ylem,” a Latin word for the substance out of which
all matter is formed). It would have a density of about
1,000,000,000,000,000 grams per cubic centimeter, and
would be far denser than the densest known white
dwarf.
The Formation of the Elements

Without committing himself on the nature of the cosmic

egg Lemaitre had viewed it as a kind of radioactive
superatom, one which would break down as ordinary
radioactive atoms do but on an incredibly larger scale.
Not only did portions of the cosmic egg form the present-
~ day galaxies but, on a more intimate scale, the cosmic egg
broke up to form the atoms we know today. From the
Lemaitre viewpoint, however, the atoms would form
- from the top down, so to speak. Very massive atoms
would be formed and these would break down further,
producing less massive atoms and so on, progressively,
-until permanently stable atoms were formed. This, howev-
. er, would mean a Universe composed chiefly of such
- atoms as lead and bismuth, which are the most massive
stable atoms in existence. It would not account for the
preponderance of hydrogen i in the Universe.
An alternate view as to the formation of the elements
was offered by Gamow, who presented it in 1948 in
association with Bethe and with the American physicist
Ralph Asher Alpher (1921- yA
According to Gamow’s suggestion, the neutronium cos-
mic egg, at the instant of the big bang, disintegrated with
ferocious violence into separate neutrons, which rapidly
broke down into protons and electrons. (Individual neu-
trons do this today for that matter, the half-life of the
breakdown being about thirteen minutes.) The protons
formed can be considered the nuclei of hydrogen-1 atoms.
As the protons formed, they would sometimes collide
with neutrons that still persisted and gradually build up
additional stable atomic nuclei of greater complexity. The
advantage of this theory is that it makes use of the
1 Gamow is supposed to have selected these associates with puck-
ish humor, for the names Alpher, Bethe, and Gamow resemble the
first three letters of the Greek alphabet, alpha, beta, and gamma.
Using the Greek symbols for these letters, the theory is sometimes
referred to as the a, B, y theory. :
 - The Beginning of the Universe 225
phenomenon of neutron-addition, something which atoms
are prone to do and which can be observed in the labora-
tory.
If a proton combined with a neutron, for instance, it
would form a nucleus of hydrogen-2, or deuterium (one
proton/one neutron). Hydrogen-2, combining with anoth-
er neutron, would form hydrogen-3 or “tritium” (one
proton/two neutrons). Tritium is, however, unstable. One
of the neutrons in its nucleus emits an electron and be-
comes a proton, and the nucleus becomes helium-3 (two
protons/one neutron). The helium-3 nucleus adds a neu-
tron to become the common helium-4 (two protons/two
neutrons). The process continues and gradually, one neu-
tron at a time, the whole list of elements is built up.
At the incredibly high temperatures following the ex-
plosion of the cosmic egg, Gamow envisages the necessary — ~
nuclear reactions to take place very rapidly, even perhaps
within the first half-hour. Gradually, thereafter, as the
temperatures dropped, the various nuclei would attract
electrons to themselves and form atoms; the atoms would
conglomerate into huge volumes of gas speeding outward
from the site of the exploded cosmic egg and gradually
condensing into galaxies and stars as they sped along.
Naturally, only a small portion of the hydrogen-1 nuclei
first formed would undergo collision with neutrons to
form hydrogen-2; only a small portion of the hydrogen-2
nuclei would undergo a further neutron collision to form
helium-3, and so on. Each successively more complex
atom would be less common than the one before, and this
would account for the fact that in the Universe today
there is a more or less steady drop in the abundance of -
atoms with a rise in complexity.
The drop is not an absolutely uniform one. Helium-4 is
much more common than either hydrogen-2 or helium-3,
and iron-56 is much more common than most of the
atoms less complex than itself. On the other hand, simple
atoms such as those of lithium-6, lithium-7, beryllium-9,
boron-10, and boron-11, are less common on a cosmic
scale than they ought to be, considering their simplicity.
The Gamow theory can make a stab at_ explaining this.
Helium-4 and iron-56, for instance, are both examples of
particularly stable nuclei. They would react to form more
complex atoms only with difficulty and would therefore
pile up. The atoms of lithium, beryllium, and boron, on
the other hand, react particularly easily and would “burn
up.” :
 ‘eee
226 ' The Universe
Gamow’s theory would account for the-relative occur-
rence of the different atoms in the interstellar material.
Once: stars form, other changes take place in their
cores. . :
Gamow’s theory has, however, one serious flaw which
no one, so far, has managed to argue away. The atoms
must be formed one neutron at a time, and there is a gap
_‘that cannot be leap-frogged once the helium-4 nucleus is
reached. The helium-4 nucleus is so stable that it has
virtually no tendency to accept either a neutron or a
_ proton. If a neutron does manage to attach itself to the
helium-4 nucleus, it forms a helium-5 nucleus (two pro-
tons/three. neutrons), which breaks down in about
0.000000000000000000001 seconds (a thousandth of a bil-
lionth of a billionth of a second) to form helium-4 and a_
single neutron again. On the other hand, if a proton
manages to attach’ itself to. helium-4, lithium-5 (three
protons/two neutrons) is formed and that breaks dowa to
helium-4 again even more quickly.
Suppose, on the other hand, a helium-4 nucleus is struck
by another helium-4 nucleus and the two fuse. This is a
much less likely occurrence than the fusion of helium-4
with the very common individual protons and neutrons,
and even so it is useless ‘as a way out. Beryllium-8- is
formed and that breaks down to two alpha particles with
super-rapidity too.
In other words, once you have formed helium-4 by
neutron addition, you are stuck. There is a gap at 5 and
another at 8 that seem insurmountable.
It is possible, of course, that two particles may strike
the helium-4 nucleus simultaneously. If a proton and neu-
tron both strike and attach themselves, then lithium-6
(three protons/three neutrons) is formed and that will
last about half a second before breaking up. That might
be long enough to continue the process. ee
Unfortunately, under the conditions postulated by Ga- ©
mow in the first half-hour following the big bang, the
individual nuclei are so widely dispersed that the chance-
of two particles striking the alpha particle simultaneously
is virtually zero. The Gamow model, then, seems to ac-
count only for hydrogen and helium atoms but nothing
beyond that.
_ Opposed to this theory of the formation of the elements
is one that I have implicitly accepted in this book and
used in my discussion of second generation stars (see page
182). This theory is suggested by Fred Hoyle, who consid-
 The Beginning of the Universe 227
ers that hydrogen-1 only is the original material and that
everything else is formed within stars and is added to
interstellar material by way of supernovae.
Hoyle makes use of the same mechanisms proposed by
Gamow, but now there is a difference. In the stellar core,
the density of matter is much higher than in open space.
The chance of a helium-4 nucleus being struck by two
particles in an essentially simultaneous manner is therefore
considerably better than in the Gamow theory. In fact,
since the stellar core is richer in helium-4 than in anything
else, there isa reasonably likely chance that a helium-4
nucleus will be struck by two other such nuclei in suffi-
ciently rapid succession to form a carbon-12 nucleus. This
would bypass the stable atoms between helium-4 and car-
bon-12—the lithium, beryllium, and boron atoms previ-
ously mentioned. Those light atoms. would be formed only
by less common secondary processes and that would ac-
count for their relative rateness in the Universe today.
The formation of the elements in stellar cores not only
avoids the gap at the 5-particle and 8-particle levels but is
also favored by an interesting piece of evidence. The
spectrum of certain unusual stars of spectral class S shows
evidence for the presence of an element called “techneti-
um.” Technetium is a radioactive: element that possesses
no stable variety of atom. The most nearly stable variety is
technetium-99 which has a half-life of about 220,000
years. This is long on the human scale, but after six
million years (no time at all in the life of an ordinary
star) only a billionth of any original technetium-99 would
remain. It follows that if technetium can be detected
spectroscopically now, it cannot have existed at the time
the star was formed but must have been made fresh, so to
speak, within the star’s interior. *
All in all, then, the weight of plausibility and of what
evidence there is, seems, at the. moment, to favor the
Hoyle model of element formation over the Gamow
model.

Before the Big Bang

If we postulate the existence of a cosmic egg marking

the original form of the Universe, with its explosion as
zero-time, we are bound to ask: But where did the cosmic
egg come from?
We can avoid having to answer that by seeking refuge
in eternity. The law of conservation of energy implies that
 228 ~—sS The ~Universe
the substance of the Universe is essentially eternal, so we
can say that the matter making up the cosmic egg was
always there.
Yet, even granting that the matter of the cosmic egg .
was always there, was it always in the form of the cosmic
egg? If the cosmic egg, as such, had always existed, it
would have to be stable. If it were stable, why did it
suddenly cease being stable and explode at what we call
zero-time, after uncounted eons during which it had not
exploded but had merely existed?
- We would face the same problem on a merely stellar
scale if we asked why a star should explode into a super-
nova after having existed for eons under conditions of
reasonable stability. In the case of a star, however, we
have learned enough to account for this in terms of .
progressive nuclear reactions proceeding in the stellar
core.
Unfortunately, we cannot-study a cosmic egg; we have
no knowledge of what can go on within it; we do not
know what forces will suffice to keep it stable or how they
will bring about progressive changes that will eventually
and suddenly make it unstable. ‘
If we ask ourselves, however, in what form the sub-
stance of the Universe might exist in order to remain
stable over countless eons, without our having to strain
‘our imaginations to account for the stability, we might
find: it easiest to think of the Universe as an exceedingly
’ thin gas. The Universe would then be the kind of “empty
' space” that now exists between the galaxies and that is
certainly stable. : ,
Such an exceedingly thin gas would still be subject,
however, to its own vastly diffuse gravitational field. Slow-
ly, over the eons, the gas would collect and the: Universe
would draw closer together. As the substance of the Uni-
verse grows more compact, the gravitational field becomes
more intense until, after many eons, the Universe is con-
tracting at a great rate.
This contraction must heat the Universe, however, a la
_ Helmholtz, and produce a higher and higher temperature .
in matter compressed into a smaller and smaller volume.
The temperature rise increasingly counters the gravitation-
al contraction and begins to slow it down.
The inertia of matter keeps it contracting, however,
past the point where the temperature effect would just
balance gravitation. Finally, the Universe contracts to a
minimum volume, represented by the cosmic egg or a
 The Beginning of the Universe 229°
close approach to it. At some point the outward push
of temperature and of radiation finally gains control, and
‘the substance of: the Universe is pushed outward, faster
_ and faster, in a manner that rapidly builds up to the big
bang.
in this view, the Universe starts in a state of virtual
emptiness, goes through a phase of contraction to max-
imum density, and then through a phase of expansion to
emptiness again. We do not need to puzzle ourselves over
a cosmic egg that existed “to begin with” and then, after
an indefinite period of stability, suddenly exploded. In-
stead, the cosmic egg becomes a momentary object placed
midway in eternity.
_ This model is termed the “hyperbolic Universe.” It can
be pictured graphically by considering its “radius of curva-
ture.” A ray of light traveling endlessly through the kind
of Universe pictured by Einstein as having a Riemannian
_ geometry (see page 200) would describe a vast circle the
tadius of which would be the radius of curvature of the
Universe. In a contracting Universe, this radius would be
decreasing; in an expanding Universe, it would be increas-
/ ing. Ina hyperbolic Universe, it would first decrease to a
minimum and then increase again.
The hyperbolic Universe lasts through eternity, but it is
not truly eternal in the sense that it persists always in an
essentially unchanged condition, or in a condition that

‘The hyperbolic Universe

®
=
=
3s
&>
o
So
°
”"
=
wo
=
ca
‘
TOREa TR te; chs Mage: citedOTEhe CHOye Feioa; Ng a CG: NS aez a eee e
ee
230 | The Universe a
hovers about some unchanged average. Instead, it under-
goes a permanent and irreversible change. It begins as an
empty Universe filled with a thin gas, presumably hydro-
gen. It ends as.an empty Universe filled with innumerable |
white dwarfs. There is a definite beginning and a definite
end, and we inhabit the brief interval of time during
which the Universe deviates for an instant from its eternal
emptiness. M
However, the hyperbolic Universe is not the only pos-
sible model one can deduce from a consideration of the
cosmic egg. If the Universe is pictured as being blown into
outward-hurtling pieces by the force of a gigantic ex-
-plosion, there still remains the force of universal gravita-
tion acting*to pull the pieces back together again, and
conceivably, it might succeed.
To understand what this means, let us consider a similar
‘situation on Earth. An ordinary explosion may hurl ob-
jects forcefully into the air, but the velocity at which they
_Tecede from the Earth is continually decreasing because of
the pull of gravity. The objects eventually come to a
momentary halt and then begin to fall back to Earth, The
more forcefully the object is hurled upward in the first
place, the higher it will travel before halting and return-
ing. :
The Earth’s gravitational field weakens with distance,
however, and if the object is hurled upward forcefully
enough, it reaches regions where Earth’s steadily weaking
gravitational field will never quite suffice to bring the
object’s steadily diminishing velocity all the way to zero.
The object will have been shot upward at more than the
“escape velocity” which, from the Earth’s surface, is just
about seven miles per second.
_ Without knowing the actual size of the cosmic egg, or-
its mass, or the force of the explosion that rent it, it is
difficult to determine whether the fragments flying out-
ward’ from it managed to attain escape. velocity or not.
Are the galaxies receding from each other forever, or will
their velocity of recession slowly- decline with time, réach
a momentary zero mark, and then will the galaxies finally
begin to fall together again—very slowly at first, but then
more and more rapidly?
Suppose the galaxies do someday begin to fall together
again. In such a contracting Universe, the radiation emit-
ted by the galaxies undergoes a violet shift and the
extent of that ‘violet shift increases as the velocity
of
approach grows greater with accelerating contraction. The
 Te ee ey Lae <ts 0 1. S.-i

F The Beginning of the Universe 231

energy pouring into the center of the Universe is com-
pressed, so to speak, and heightened. Under the lash of
that energy outpouring, the nuclear reactions that take
place in an expanding Universe are reversed.
Where fusion from hydrogen to iron would yield energy
in an expanding Universe of generally dimming radiation,
breakdown from iron to hydrogen would absorb energy in
a contracting Universe of generally brightening radia-
tion.
In short, by the time the Universe was condensed to
something approaching the cosmic egg, it would be all
hydrogen again. Following the formation of the cosmic
egg there would be another big bang and the whole
procedure would start over again. The result would be an
endlessly “pulsating Universe” or “oscillating Universe.”
A pulsating Universe might be looked on as a gigantic
Cepheid variable. Such a Universe would be eternal in a
teal way, for although catastrophic changes take place,
these are periodic. There is no clear beginning or ending;
no steady, irreversible change from one grand universal
structure to another quite different one. As the Universe is

The pulsating Universe

_§

eG
&
Md
3
 232 The Universe
right now, so it will be again an indefinite number of eons
in the future after it collapses and explodes again, and so
it was an indefinite number of eons in the past prior to its
last collapse and explosion.

Continuous Creation

Yet prior to 1952, at least, the big bang theory seemed

to have an element of impossibility to it. It placed zero-
time 2 eons in the past, when the Earth was nearly 5 eons
old. Somehow the big bang had to be an illusion, and a
model of the Universe had to be constructed which did
not involve the cosmic egg at all.
' The new model. arose out of the feeling that the
general scheme of appearance of the Universe would be
the same from any vantage point. No matter where an
observer might be placed in the Universe, no matter on
‘which galaxy, he would find all the galaxies distributed
symmetrically about him in all directions; he would find
the general density of matter the same; he would find all
the other galaxies receding at a rate proportional to their
distance; he would find himself the center of an observable
Universe.
This notion of a Universe generally uniform through all
space was advanced by the English cosmologist Edward
Arthur Milne (1896-1950). He termed it the “cosmologi-
cal principle.” :
The cosmological principle is only an assumption but,
barring strong evidence of its falsity, astronomers are
attracted to it, since by using it, the Universe can be made”
simple enough to be reflected in the kind of models
astronomers can construct. Einstein, for instance, assumed
the cosmological principle when he treated the Universe as
though the matter in it were smeared out evenly, for then
the Universe certainly seems the same no matter where
you are located in space. ;
The cosmological-principle seems to require an infinite
Universe, for otherwise you could imagine yourself trans-
ported to the very edge, where you would find all the
galaxies on one side and nothing at all on the other. How
would this jibe with the Riemannian Universe assumed by |
Einstein, a Universe finite in volume? :
_ Actually, it is possible to have a Universe that is finite
in volume and yet contains an infinite number of galax-
ies, 5
By Einstein’s theory of relativity, it is necessary to
 a Ree a a oe Nite ats PRS I et ae ee te TY
Len
ae Ieatehty SP ee eee Te

The ren of the Universe _ 233

suppose that an object moving with respect to ourselves is
found to be, by any measurement we can make, shorter in
the direction of its motion than it would be if it were
motionless with respect to ourselves. The greater the veloc-
ity, the more pronounced this “foreshortening.” If the
object is moving at the velocity of light, its length in the
direction of motion is reduced to zero.
‘The distant galaxies, receding from us, must be fore-
shortened to our view; and the more distant, the more
foreshortened, in view of the increasing velocity of reces-
sion with distance. Near the edge of the observable Uni-
verse, the galaxies are paper-thin and less, and an infinite
number can be squeezed into the very rim. We then have »
- an infinite Universe packed into a finite volume. ‘(Be-
cause of the red shift, the infinity of galaxies at the rim
would deliver only a finite amount of radiation, particles,
or gravitational force to the interior.)
-An observer on one of the galaxies of the rim would
not, of course, find himself and his galaxy paper-thin. His
galaxy would seem normal to him as would the other
galaxies in his vicinity. At distances far removed from
himself, he, too, would observe an infinitely crowded rim,
and to him our Galaxy, if he could observe it, would seem
paper-thin. (It is a matter of viewpoint, just as the Aus-
’ tralians seem upside down to us and we seem upside down
to them, if we imagine ourselves looking through a trans-
parent Earth.)
Such an infinite Universe does not fit well with the
notion of the cosmic egg, for certainly it is easier to view
the cosmic egg a8 possessing a finite size and as exploding
into a finite number of galaxies. Gamow, however, is
ready to consider a cosmic egg of infinite size, and in that
case the cosmological principle would not be inconsistent
with either the hyperbolic or pulsating Universe. +
_ {fo three astronomers in England, Austrian-born Her-
mann Bondi (1919- - ), Thomas Gold (1920- ),
and Fred Hoyle, it seemed that the cosmological principle
was incomplete. It allowed the Universe to be unchanged
with the observer’s position in space, but what about his
position in time?
If the Universe undergoes changes that are irreversible
as in the case of the hyperbolic Universe, or reversible .
only after many eons as in the pulsating Universe,
then an observer would find the Universe varying its
nature radically with time. An observer 10 eons ago, for
instance, might observe a small filled Universe with young
 (234 ‘The Universe ee
-closely spaced galaxies consisting almost exclusively of
young stars made up of hydrogen and virtually nothing
else: An observer 50 eons in the future might observe a
vast empty Universe with galaxies separated. by enormous
distances and made up largely of white dwarfs. An observ-
er a 100 eons in the future might observe a contracting
Universe.
Bondi, Gold, and Hoyle believed it logical to suppose
that this could not be so. The Universe would have to be
essentially the same for observers at all times. as well as in.
all places, and this they called the “perfect cosmological
principle.” ;
Yet the Universe was changing in two important ways;
ways that. were accepted on the basis of the strongest
evidence and with which it was impossible to quarrel.
First, the distance between the galaxies is growing steadily
larger, and secondly, hydrogen is steadily fusing into heli-
“um and more complicated atoms. If the perfect cosmolog-
ical principle is to be valid, there must be processes that -
neutralize these changes.
The solution advanced by the three astronomers in 1948
was to suppose that hydrogen was continually being
created out of nothing, and this suggestion is referred to
as the “continuous creation theory” or the “steady state
theory.” :
Naturally, one’s first reaction to any such suggestion is
to object that it violates the law of conservation of ener-—
gy. Yet that law is merely an assumption based on the
_ fact that mankind has never observed energy to be created
_ out of nothing. But the requirements of the continuous
creation theory are small indeed; matter need be created
Only at the rate of one atom of hydrogen per year in a
_ billion liters of space, and such a rate of creation would
be far too small to be detectable by any instruments we
possess. Such continuous creation would not violate the
law of conservation of energy which does not really say
“Energy cannot be created out of nothing,” but merely
“Energy has never been observed to be created out of
nothing.”
(An alternate view might be that the matter formed
appears at the expense of the energy of expansion of the
Universe, which would thus expand a bit more slowly than
it would if continuous creation were not taking place.)
If continuous creation is allowed, let us next see the
consequences. The galaxies must be viewed as separating
not as a result of some explosion but as a consequence of
 . The Beginning of the Universe 235
some more subtle effect. In 1959, Hermann Bondi and
Raymond Arthur Lyttleton speculated, for instance, that
the positive charge on the proton might be very slightly
larger than the negative charge on an electron. Suppose
the positive charge of the proton were only larger by a
billionth of a billionth of the size of the electron’s negative
charge. This would be far too little to detect with man’s
most refined instruments. It would, however, suffice to
’ build up a generally positive charge on all the galaxies and
- would cause them to undergo a steady mutual recession.
This explanation of the expanding Universe is considered _
quite unlikely by astronomers generally, but it is an exam-.
ple of the sort of physical cause, other than explosion, that
is sought for by those who wish to avoid the big bang.
As the galaxies recede from each other, whatever the
cause, the spaces between gradually accumulate matter .
through continuous creation.
The accumulation is slow, to be sure, but so is the rate
at which galaxies recede from each other, compared to
the vast spaces between. It takes several eons. for the
distance between two neighboring galaxies to double, and
by that time enough matter has been formed between
them to condense into a new galaxy. The density of
galactic distribution never grows less, therefore, as the old
galaxies spread apart, slowly collecting in the paper-thin
Tim or, as Hoyle seems to assume, moving beyond, some-
how, the rim of the observable Universe. New galaxies
would form-between them, and the two effects would just
balance each other.
Then, too, the matter that. is formed in continuous
creation would naturally be simple. A fragment of matter
would be formed as a hydrogen atom, perhaps, or as a
neutron that would break down in a matter of minutes to —
a proton and electron, which would then come into associ- —
ation to form a hydrogen atom. In either case, the new
galaxies formed out of newly created matter would be
young galaxies built up out of fresh hydrogen. This means-
that any observer at any time in the future would see as
many galaxies about him as he does now and as many
young galaxies among them. The Universe would never
become either empty or old although individual galaxies
can age to any extent.
If we look backward in time, the galaxies can be
pictured as contracting, but they need never come togeth-
er. Continuous creation in a time-run-backward becomes
continuous destruction. Complex atoms break down to
 : ; : a
236 The Universe
hydrogen in such a backward Universe and the hydrogen
disappears. Galaxies melt away as they approach each
other and never form a cosmic egg. Other galaxies take
their place from the infinite supply at the rim or from the
infinite supply beyond the rim, depending on one’s view-_
point. In the long run, then, galaxies neither get closer
together nor do they get younger, however far back in
time we go. |

The evolutionary Universe The steady-state Universe

Under such conditions, the overall state of the Universe

does not change with time in either direction but is steady.
Such a model represents a “steady-state Universe”. and
adheres to the perfect cosmological principle.
There is something very attractive about the thought of
an eternal and immortal Universe in which man (or his
evolved descendants) might conceivably live forever, and
it has appealed very strongly to the general public. This
appeal has been strengthened by the fact that the most
assiduous proponent of continuous creation has been Fred
Hoyle, a charming and persuasive writer, whose popular
works on astronomy have been well received by the gener-
al public. (On the other hand, George Gamow, who has
been the. most: prominent of those on the side of the big
 ‘The Reghiniag of the “Lilverse 237
bang, is also a highly successful writer of popularized
science. Rarely in the history of science has there been |
such.a clash of titans in full view of the lay public.)
To be sure, in 1952, when Baade proposed a new scale
of cosmic distances and shoved zero-time for the big bang
back to 6 eons or more in the past, the greatest argument
in favor of the steady-state Universe (the argument that
the big bang was impossible) vanished. By that time,
however, the steady-state Universe had proved too attrac-
tive to abandon easily.
It is quite difficult to decide among the models of the
Universe presented in this chapter.
To make the decision; we must remember that. the
steady-state Universe adheres to the perfect cosmological
principle, while the others do not. This means that if we
could change our position in time, we could solve the
problem. If the general appearance of the Universe does
not change with time, if the galaxies are not closer and
younger in the past or more separated and older in the
future, then the steady-state Universe would be strongly
supported. Otherwise, either the pulsating Universe or the
hyperbolic Universe would be- supported, and from the
extent of the change with time, we might be able to
choose between these two.
To be sure, if our descendants remain in existence for°
several eons and maintain a continuity of culture, they
will be in the far future and will be able to decide, but
astronomers: would like, if possible, to find the answer
now, and surely one cannot blame them.
What is needed, then, is time travel, and one form of
time travel ispossible.
When we say that the ‘Andromeda galaxy is 2,300,000
light-years from us, we mean that it takes light 2,300,000
. years to cross the distance that separates it from our eyes.
_ When we look at Andromeda, either with our eyes or by
some instrument, we are seeing light that left Andromeda
2,300,000 years ago, and we See it not as it is now, but as
it was that length of time ago. In studying the Andromeda
galaxy we are, in effect, time travelers who have -penetrat-
ed 2,300,000 years into the past.
The farther we penetrate out into space, the longer it
takes light to reach us from the objects we can see, and
the farther back in time we find ourselves. Our best
optical telescopes (as far as was known in the 1950’s)
were reaching objects one or two billion light-years away,
 238 ‘The Universe
and in looking at them we are seeing that portion of the
Universe as it was 1 or 2 eons ago. :
If the steady-state view of the Universe is correct, this
difference in time should not matter. The Universe 1 or 2
eons ago would have the same general properties it has
today. The galaxies we see at the far end of our telescopic
capabilities should be spaced .no more closely or no less
closely than they are today, they should be receding from
. each other at the same rate as today, and they should, in
general, have no properties that would distinguish them as
a whole from those galaxies we see in our immediate
neighborhood, é ,
If the pulsating or hyperbolic views of the Universe are
correct, then the difference in time should make for con-_
siderable change, and there should be some important
properties, at least, in which the far-off edge of the Uni-
verse should differ from the regions near us. ; ;
For instance, the very distant galaxies should be young-
er than the galaxies in our neighborhood, richer in hy-
drogen, spaced more closely together, and separating at a
greater rate (since the explosive force had not yet been
weakened then by the slow, steady pull of gravity). Fur-
_ thermore, since those regions represent the youth of the
’ Universe, they might include objects not to be found at all
- in our own neighborhood, objects characteristic only of a
young Universe. Furthermore, by studying the extent to
which these differences make themselves manifest we
might determine whether the pulsating Universe or the
hyperbolic one better fits the facts. ;
! This seems straightforward enough, but it is actually
maddeningly frustrating. The more distant the objects we
study, the more likely we are to decide among the sug-
gested models of the Universe; but the more distant the
ee we study, the more difficult it is to detect anything ~
at all. :
The best that could be done by the middle-1950’s was
to study the red shifts of the most distant galaxies that
could be detected. In a steady-state Universe, Hubble’s
constant ought to be the same for all times and therefore
for all distances. In a pulsating or hyperbolic Universe,
Hubble’s constant ought to decrease with forward-
progressing time and should have been quite high in the
youth of the Universe. In that case, the very distant
galaxies (representing that youth) ought to be receding
more rapidly than we would expect and ought to have red
shifts greater ‘than we would expect. By 1956, it did
 Te SP.
is Pana

‘The Beginning of the Universe _ 239

indeed seem that such too-great red shifts were observed
and this was an argument against the steady-state Uni-
verse. However, the red shifts were too large by such
small.amounts and the difficulty of observing them was so
great that the evidence was anything but conclusive.
Something better was needed, but what?

Red shift at great distance ©

—_—

Increasing
red-shift

Note: If the Universe were steady-state,

St na ee ee eet ee

Increasing distance
——

To answer that question, we must turn to those branch-—

es of modern astronomy that do not depend on visible
light. Visible light fails us; other manifestations of the
Universe may not.
 yi

CHAPTER 1 {

Particle Bombardment |
. - Massless Particles

All the information we obtain concerning the Universe

outside the Earth itself is derived from particles which,
emitted by bodies in space, cross the gap and reach Earth.
On Earth, they interact with particles already here in such
a way that the results of the interaction can be perceived
by our senses.
This is a roundabout way of saying that (to give the
most common and best-known example) the distant stars
and galaxies emit light which we can see, separate by
spectroscope, or impress on photographic film.
To be sure, light was viewed ‘as a wave-form, rather
than as particles, throughout the nineteenth century. In
1900, however, the German physicist Max Karl Ernst
Ludwig Planck (1858-1947) advanced the “quantum the-
ory” in which light and, indeed, all forms of energy, could
be viewed as consisting of discrete little packets which
Planck called “quanta” (or “quantum” in the singular).
This theory was reinforced by Einstein in 1905, and it
became more and more clear that energy quanta could, in
some ways, behave like particles.
Indeed, by the 1920’s, a kind of duality was recognized.
_ All particles had the capacity to behave like wave-forms
in some ways; all wave-forms had the capacity to behave
like particles. The two aspects were not ordinarily well-
balanced. The particle-like properties of protons, for in-
stance, are much more pronounced and easy to detect
than are the wavelike properties, but the latter are there
and can be detected if one goes about it carefully enough.
On the other hand, the wavelike properties of ordinary
light are much more pronounced than are its Pad
240
 Particle Bombardment , 241
properties, but the latter are present also. The shorter the
wavelength of light, the greater the energy content of the
‘individual quanta, and the more pronounced and easy
to detect are the particle-like properties.
In 1923, the American physicist Arthur Holly Compton
(1892-1962), using forms of light of particularly short
wavelength, demonstrated their particle-like properties un-
mistakably. He coined the word “photon” for such parti-
cles of light (the “-on” suffix being commonly used for
subatomic particles and the “phot-” prefix coming from
the Greek word for “light’’).
The light we see the Universe by, then, can be viewed
as a shower of photons descending on us from all direc-
tons: from the Sun, stars, and galaxies directly and from
the Moon and planets by reflection. The interaction of
these photons with the retina of the eye comprised the
whole of astronomy prior to the middle of the nineteenth
century.
Since then we have expanded matters not only by
allowing photons to affect photographic plates rather than
the retina directly but by recognizing the existence of and
dealing with many forms of particles other than photons.
Some of these other forms are subtle indeed, and I will
begin with the most subtle of all—particles that resemble
photons in some ways.
The photon has a “rest-mass” of zero. That is, if it
could be made to stand still, it would turn out to exhibit
none of the properties associated with the possession of
mass. It would have no inertia, ‘and it would neither
produce a gravitational field nor respond to one. It is
therefore considered a “‘massless particle.”
Such masslessness is purely theoretical, however, for a
photon cannot be made to stand still. The moment it is
formed, it moves away from its site of formation at
300,000 kilometers per second.1 While it is moving in this
fashion, the photon does exhibit some of the properties
associated with mass, that of responding slightly to a
gravitational field, for instance:
_At least two other massless particles, the “graviton” and

1 That is, if it is in a vacuum. If it is traveling through transparent ‘

media other than a vacuum, such as air, water, or glass, its velocity
is less, sometimes considerably less. Under no conditions, however,
can a photon be brought to actual rest without being absorbed.
Moreover, the moment a photon passes from some transparent
medium back into vacuum, its velocity instantaneously becomes
300,000 kilometers per second again.
 Nee POR SE ae + Tea ee aa
Universe Sener
242 The
-
the “neutrino,” have been postulated by physicists. Both,
like the photon, ‘have a rest-mass of zero but are never at
Test; they travel only at the speed of light as long as they
exist. This would be true, apparently, for all massless
particles.”
The photon, graviton, and neutrino are all electrically
uncharged, and if all three are massless as well. it seems
fair to wonder in what manner they may be distinguished.
One distinguishing mark stems from the fact that most
subatomic particles can be pictured as though they were
rotating about an axis, either clockwise or counterclock-
wise. The angular momentum associated with this rotation
can be expressed, therefore, by either a positive or nega-
tive number. Physicists use units in such a way as to assign
the photon a “spin” of +1 or —1. On this basis, in order
* to account for the manner in which subatomic particles
_ behave, physicists find they must assign a spin of +% or
—¥ to the neutrino and one of +2 or —2 to the graviton. —
This alone suffices to make the three types of particles
absolutely distinct. s
The graviton remains a theoretically postulated particle
only, for it has never been detected directly. Indeed, such.
are the properties that the logic of the situation has forced
physicists to assign to it, that it may never be detected.
Nevertheless physicists suppose that it is by virtue of the
emission and absorption of gravitons that a gravitational
field is brought into existence.
If we cannot detect the graviton directly, then we can
at least detect it indirectly through the effects of the ©
gravitational fields it produces. Thus, the interchange of
gravitons between the Moon, the Sun, and the Earth
produces the tides and keeps the Moon and Earth in their
interlocked orbit about the Sun. The interchange of gravi-
tons between the Sun and the Galactic center keeps the
Solar system in its mighty revolution about that invisible
core. S
_ On the whole, the gravitational effect on ourselves of
individual stars outside the Solar system, and of individual
galaxies beyond our own, is undetectable and will proba-
bly continue to be so through the foreseeable future.
Nevertheless, there are gravitational interactions be-
tween neighboring stars and between neighboring galaxies
* On the other hand, particles possessing mass, in however small
an amount, can never travel at the speed of light. They may ap-
oe that speed very closely, but they can never actually attain
i y. :
 Particle Bombardment = 243°
which yield information to us. The two stars of a binary
system move about each other, for instance, in accordance
with the gravitational law worked out by Newton, and by
the application of this law, the relative masses of the two
bodies can be determined. Even when one of the members
of the system cannot be seen, the motion of the other in
response to the gravitons emitted and absorbed by the
unseen body, will yield information as to the mass of the
latter. That is how Sirius B was first detected (see page =

51).
Indeed, smaller bodies than Sirius B have been pinned
down through gravitational effects. In 1943, the motions
of 61 Cygni (see page 52) were studied by a team under
the guidance “of the Dutch-American astronomer Peter
Van de Kamp (1901- ). The star, 61 Cygni, is actual-
ly a binary, so that there is both a 61 Cygni A and a 61
Cygni B, revolving about a common center of gravity. The
motion of one of these, however, wavered slightly, but
sufficiently to indicate the presence of a body about 1/120
the mass of the Sun. Such a mass, only eight times that of
the planet Jupiter, cannot support nuclear reactions to an
extent sufficient to qualify it even as a dwarf star. For that
teason, the new body, 61 Cygni C, was considered a
planet, albeit a giant one—the first planetary body to be.
discovered outside the Solar system. Others of the sort
have been discovered since. In: 1963, a close look at
Barnard’s star (see page 45) showed a slight wavering in
its proper motion that indicated the presence of a plane-
tary body only one and a half times the mass of Jupiter.
The neutrino is intermediate between the graviton and
photon in ease of detection. Unlike the graviton, the
neutrino has been detected, but it is by no means as easily
detected as the photon.
The existence of the neutrino was first postulated in
1931 by the Austrian physicist Wolfgang Pauli (1900-
1958), out of a necessity to explain certain interactions of
subatomic particles that otherwise could not be explained.
For a quarter of a century, it remained a theoretical
figment of the scientific imaginations, and then, in 1956,
two American physicists, Clyde Lorrain Cowan, Jr. (1919-
) and Frederick Reines (1918- ), designed a
‘careful experiment that clearly demonstrated the very
occasional interaction of a neutrino with a proton. The
actual existence of the neutrino was accepted at once by
the scientific community.
The neutrinos detected in 1956 were produced by
 Phere244ses ae erThe oa.
Universe
- a eed
uranium fission. within the core of a man-made nuclear
reactor. Neutrinos (of a somewhat different type) are also
produced, and in vastly greater numbers, of course, in the
cores of stars (see page 180). Neutrinos have the proper-
ty, however, of being able to pass, unaffected, through e
great masses of matter. Uncounted trillions of neutrinos,
moving at the speed of light, pass through the Earth each
second—as though the Earth were not there. Only very
occasionally does a neutrino interact with one of the ao
e
particles of the- planet.
After 1956, strenuous efforts were made to detect these
_ Sun-born and star-born neutrinos. Reines, for instance, set
up a large apparatus, designed to detect such neutrinos, in |
a South African gold mine two miles deep in the Earth. It ~
- may seem strange to attempt to study the heavens from a
hole in the ground, but neutrinos can reach that hole (or
any. other portion of the Earth down to its very center)
- without trouble, wheréas no other detectable particle can.
Finally, in 1965, after observations lasting half a year, —
Reines reported the detection of seven neutrinos (seven!).
_ This is the first birth-pang of “neutrino astronomy.” If,
in times to come, the methods of detection become more
efficient, much of value may be learned. Neutrinos come
directly from stellar cores and by analyzing the distribu-
tion of energies among them, the temperature and other
properties of the core’ may be determined directly, rather
than deduced more or less uncertainly by indirect means.
P
Cosmic Rays

_. Much more spectacular is the bombardment of the e

—
Earth by particles possessing mass. :
As early as 1900, nuclear physicists were studying the
manner in which energetic radiation from radioactive
atoms knocked electrons out of atoms in the atmosphere.
What was left of the atoms carried a positive electric
charge and these charged atom-fragments were called
positive ions.”
Gradually, it became clear that no matter how carefully
physicists shielded a sample of air, enclosing it in lead
boxes that were supposedly impervious to radiation, ions
continued to be formed within the sample at a slow rate.
Apparently, radiation still more energetic than had yet
been observed was penetrating a thickness of lead that
would have been an adequate shield under ordinary cir-~
cumstances, ~ - ‘ :

a il
 ie Particle Bombardment 245 .
Physicists, generally, took it for granted that this partic-
ularly energetic radiation was coming from the soil. It was
in the soil, after all, that radioactive substances were to be
found. To settle the matter, the Austrian physicist Victor
Franz Hess (1883-1964) undertook a series of balloon
ascensions beginning in 1911. The plan was to test for the
presence of this ionizing radiation some miles above the
Earth’s surface. The thickness of air between the surface
and the balloon ought to suffice to absorb at least some of
the energetic radiation, and the rate of ion-formation in
the balloon would then prove to be less than that on the
surface.
Precisely the reverse was observed. The rate of ioniza-
tion increased, and the higher the balloon rose, the greater
the increase. Clearly, the origin of the radiation was not
below the balloon on the Earth’s surface, but above it.
Hess called it “high-altitude radiation.” Over the next
decade or so, Hess’s results were confirmed over and over
again, and it became quite plain that the radiation was
striking Earth everywhere and must have its source some-
where out in space. Since the radiation reached Earth
from the outside Universe (or cosmos), the American -
physicist Robert Andrews Millikan (1868-1953) called it
“cosmic rays” in 1925, and that name stuck.
~The question next arose as to just exactly what the
cosmic rays might be. One of two alternatives seemed
most likely: they were either extremely energetic photons,
shorter in wavelength and therefore more energetic than
any ever observed previously, or they were massive parti-
cles moving at extremely high velocity and deriving their
unprecedented energies from the combination of great
mass and high velocity.
All the massive particles known in the 1920’s carried an
electric charge and it was assumed that would be true of
the cosmic rays as well if they fell into this category. It
followed from this that if the cosmic rays approached the
Earth from all directions equally, they ought, nevertheless,
to strike the polar regions with greater frequencies than
they struck the Tropics.
This arose from the fact that the Earth behaved like a
magnet, with its north and south magnetic poles in the
polar regions, and with magnetic lines of force curving
outward from north to south, spreading most widely apart
in the Tropics and coming most closely together in the
polar regions. Any charged particle entering the Earth’s
 = aes

246° — The Universe

magnetic field would be deflected northward or southivntd
according to an interaction that was well-known to the
physicists of the time. Some particularly energetic cosmic
rays might smash right through the magnetic field by main
force and land in the Tropics (if aimed there), but on the
whole the incidence of cosmic rays would increase steadily
as one traveled north or south from the Equator.
Photons, on the other hand, being uncharged, would
remain essentially unaffected by the Earth’s magnetic
field. If cosmic rays, were photons and approached the
Earth from all ‘directions equally, they would, in that case,
strike all portions of the Earth’s surface equally.
Compton, beginning in 1930, carried out extensive
studies of cosmic rays in different portions of the world
- and found that there was indeed an increase in intensity
with latitude, precisely as one would expect if the rays
consisted of charged particles.
There was still the question as to whether the particles
were positively charged or negatively charged. The Italian
physicist. Bruno Rossi (1905- +): pointed out in 1930
that the nature of the deflection of cosmic rays would
vary with the nature of the charge. Positively charged
particles would be deflected in such a way that more
would seem to be approaching the observer from the west
than from the east, while the reverse would be true for
negatively charged particles. By 1935, the decision was

The Earth as magnet :

north magnetic pele

———= magnetic lines of force south mognetic pole

——~—-—— incoming particles
ae Foee Gate oe rr : -

| Particle Bombardment aT
clear: -cosmic rays were made up of positively eed
particles.
This is true at least for the radiation in the form in
which it exists before it hits the Earth’s atmosphere (“pri-
mary radiation”). Once it enters the atmosphere, it strikes
atoms, producing a variety of kinds of particles of lesser,
but still very high, energy (“secondary radiation”). It is
naturally with the-primary radiation that astronomers are
chiefly concerned.
_ To reach the primary radiation one must rise into the
‘stratosphere at least, leaving most of the interfering, in-
teracting air molecules beneath. In the late 1930’s balloon
ascensions began to be made. for the purpose, making use
of new plastic materials, lighter and more leakproof than
the products available earlier. Altitudes as high as thirteen
miles were achieved, and at that height 97 percent of the
atmosphere had been surmounted.
The primary radiation, it turned out, consisted chiefly of
protons (hydrogen-1 nuclei) moving at nearly the speed
of light. By the 1960’s it was clear that protons made up
about 90 percent of all the energetic particles of the
primary radiation. An additional 9 percent were made up
of helium-4 nuclei, and the remainder of nuclei of still
heavier atoms, up to and including iron-56 at least.
In short, cosmic rays were made up of the general
_matter of interstellar space, set into violent motion, and
possessing almost unimaginable concentrations of energy. ~
In the last few years, physicists have managed to build
costly, enormous, and intricate “particle accelerators” that
can endow single particles with as much as thirty billion
electron-volts (“30 Bev”) of energy. The size of this
energy concentration may be judged from the fact that
the energy of an ordinary photon of light might be some-
thing like two electron volts (“2 ev”).
Man, then, can produce protons with energies up to
fifteen billion times that of a light photon, and ‘such
energies do indeed represent an approach to the level of
energies in cosmic rays (which is the reason why one giant
particle accelerator was given the name “Cosmotron” by
its designers).
Nevertheless, only the lowermost levels of cosmic ray
energies are reached even by man’s. proudest creations.
Many cosmic ray particles strike the Earth’s atmosphere
with energies of far more than 30 Bev; some reach us
with energies of 1,000,000 Bev and more!
 PRPS
ek ashe ery yee, Sale he ey aa ee Ba

248 ihr The Universe

Cosmic Ray Sources

But where do the cosmic rays come from? And what

gives some of them such incredibly tremendous ener-
gies? : :
A logical origin might be the Sun, but that was almost
immediately eliminated as a main source since cosmic rays
approach the Earth from all directions equally, from the
direction opposite that of the Sun just as frequently as
from the direction of the Sun. Granted that cosmic rays, if —
originating in the Sun, would be deflected by the Earth’s
magnetic field, and that some would even be pushed
around to the rear, it is quite unthinkable that the end
result of any such defiection would be to smear the cosmic
rays so completely evenly all about the Earth. The source
had to be, then, from somewhere outside the Solar sys- A

tem. The even smear of cosmic ray influx made it impos-

sible, however, to pin the phenomenon down to any par-
ticular objects in the Universe.
And yet the Sun was not to be counted out com- —
pletely. ;
The Sun’s radiational system is not perfectly even and —
smooth. The Solar surface breaks into “sunspots,” regions
of comparatively low temperature which, therefore, show
up black against the hotter and brighter surroundings.
_ These sunspots are accompanied by magnetic fields, and
the energy store in -these fields can manifest itself in
violent manner. KE.
‘One of these manifestations is that of the “Solar flare,”
the sudden brightening of an irregular area near a sun-
spot. The first report of a Solar flare was made in 1859 by
the English astronomer Richard Christopher Carrington
(1826-1875). He thought the sudden brightening on the
Solar disc was caused by the fall of a large meteor (in line
with the suggestions Helmholtz was making at the time).
Almost immediately after that observation, however,
disorders were reported in the behavior of compasses and
the aurora borealis in polar regions grew particularly
brilliant.
Since then the association of Solar flares with such
“magnetic storms” has been unmistakable. Not all flares
give rise to them, of course, only those that are more or
less central on the Sun’s face and that are therefore aimed
directly at us. It seems clear that the great energies
associated with fiares suffice to eject quantities of subatom-
_1¢ particles out into space. Since flares and other ener-
 ee rete ree ns
oe: Particle Bombardment . 249
getic phenomena are always taking place here and there
on the Sun’s surface, we might picture the Sun as sur-
rounded by a cloud of energetic charged particles shooting
outward from it in all directions. This is the “Solar wind.” —
By 1958, rocket experiments had proved the actual |
existence of this Solar wind. The velocity with which -
particles stream outward from the Sun can be as high as
450 miles a second, and such particles carry a surprising
amount of kinetic energy. Until the last decade; for in-
' starice, it had been thought that comets’ tails streamed
away from the Sun because of the pressure of Solar radia- ©
tion against the tiny particles making up the tails. Appar-
ently, this is not so; it is the force of the Solar wind that
does it.
Naturally, the charged’ particles in the Solar wind, on
approaching the Earth, would have to interact with the —
planet’s magnetic - field. As predicted by the Greek
amateur scientist (and now professional physicist) Nich-
Olas Christofilos in, 1957, the charged particles making ~
up the wind would be defiected by the magnetic lines of —
force, spiralling about them from north magnetic pole to -
-south magnetic pole and back, contributing to the forma- |
tion of a doughnut-shaped region of dense charge about ~
the Earth, well above its atmosphere. This was not taken
seriously at first (partly because Christofilos was an
amateur), but in 1958, rocket observations: made under —
the direction of the American physicist James Alfred Van —
Alien (1914- ) demonstrated the actual existence of ©
such regions. They came to be known as the “Van Allen —
belts” at first and then, later, as the “magnetosphere.”
_ The limits of the magnetosphere are shaped into a
streamlined teardrop by the Solar wind. In the direction of ~
the Sun, the magnetosphere is bluntly rounded, but it
curves into a long tail on the side away from the Sun. ©
The presence of a Solar flare pointed in our direction
produces a local strengthening of the Solar wind, a kind -
of stormblast of charged particles bearing down on us and
filling the magnetosphere to overflowing. The particles
pour down on the polar region of the Earth, particularly,
giving rise to the strengthened aurora and introducing
such irregularities into the magnetic field as to send com- —
passes skittering. More important in a practical sense,
nowadays, a Solar flare alters the properties of those
sections of the upper atmosphere which normally contain
a high concentration of electron.charge in the form of
ions. (That_section of the atmosphere is therefore called
 250 - The Universe
_ the “ionosphere”). As a result, magnetic storms upset the
_ workings of radio and, indeed, any device that interacts
with the radiation of the ionosphere. :
Naturally, the energy of the particles released into the
Solar wind varies according to the size and force of the
flare. What if a really large flare burst out? One took
place in 1942 and it was quickly followed by a temporary
increase in the cosmic ray influx. This has been observed a
number of times since, and it became clear that the Sun
can serve as a source of cosmic rays, at least occasionally.
These solar cosmic rays are “soft”; that is, they are
relatively unenergetic, with energies ranging from 0.5 to 2
Bey, but the principle remains. ~
What if stars generally produce cosmic rays as a result.
of flares or other phenomena? Perhaps in passing across the
vast interstellar distances, local magnetic fields would
defiect the cosmic ray particles over and over again so
that all signs of the original direction of travel would
disappear. In the end, ‘then, the motions of cosmic ray
particles would be randomized to the point where they
_ would seem to come from all directions equally rather
than in highest concentrations from the plane of the Milky
Way, where most of the stars are.3 : ;
. This explanation is not quite sufficient. If all stars con-
tributed equally to cosmic ray production, the Sun, being
much the nearest, would drown out all the rest, as it does ©
with respect to light production. In that case, the cosmic
tay influx would appear to be heavily weighted in the
direction of the Sun—and it is not.
It follows, then, that some stars are much richer pro-
ducers of cosmic ray particles than the Sun is. There are
certain variable stars, for instance, whose source of varia-
tion is the periodic production of large flares. Possibly,
_these “flare stars” are rich sources. And then, too,. there
are the supernovae.
_ The cosmic ray influx from such specialized members of
the Galaxy may completely drown out the piffling produc-
pn from ordinary stars, even from our own nearby
un.
But that still leaves the problem of the energies of the

® Not only would most stars be expected to have an associated

magnetic field that would serve to deflect any cosmic ray particles
passing by, but the Galaxy generally is now thought to have a weak
magnetic field of its own—one that may serve as a factor in main-
taining the existence of the arms of spiral galaxies.
 Particle Bombardment 251
_ cosmic ray particles. If the Sun can produce particles with
1 Bev of energy, it is not surprising that a supernova could.
produce them with much larger energies—but billions of
Bev? No known nuclear reaction in even the hottest and
most ferocious supernovae could be expected to produce
particles as energetic as many of those in cosmic rays.
But do the particles actually have to be formed at those
energies to begin with? In 1951, the Italian-American
physicist Enrico Fermi (1901- 1954) suggested an alter-
nate possibility. Suppose that cosmic ray particles were
produced at quite moderate energies of a few Bev and
suppose that the effect of the Galaxy’s magnetic field was
to accelerate such particles and increase their energies.
The process envisioned was similar to that of man-made
cyclotrons, devices which whirled charged particles round
and round under the influence of a magnetic field,
-pumping additional energy into them at every cycle. As
the particles gain energy, they are deflected less and less
by the magnetic field until finally they can no longer
remain within the confines of the cyclotron. They shoot
out at high energies.
The magnetic intensities of man-made magnets are
much greater than those of the Galaxy’s magnetic field,
but the latter extend over many thousands of light-years.
. The cosmic ray particles are accelerated only very slowly
as they travel, but in eons of time they reach high energies
indeed.
At any point in their travels, such cosmic ray particles
may happen to smash into some obstacle such as our own
planet. We would therefore be subjected to a wide spec-
.trum of particle energies, for the energy of any given
particle would depend to a large extent on how long it had
been traveling in space before colliding with us. The
longer the time of previous travel, the higher the energy
at the time of collision.
As a cosmic ray particle gains energy, however, it is
deflected less and less under the infiuence of the magnetic.
field until it is finally traveling in so nearly a straight line
that even the vast width of the Galaxy is insufficient to
hold it. By the time the energy of the particle reaches the
neighborhood of a hundred million Bev, it goes shooting
out of the Galactic cyclotron, so to speak.
It would be expected then that if our Galaxy were the
only source of cosmic ray particles, no energies higher
than a hundred million Bev would be detected. Neverthe- .
less, higher energies are occasionally detected, up to ten -
 252°. |. ‘The Universe
billion Bev at least. It can only be assumed that such
_super-energetic particles must originate in other galaxies
with stronger magnetic fields than our own. These, after
finally being hurled out of their parent galaxies without
having had the ill-fortune of having collided with any
piece of matter, cross intergalactic space, happen to pass
through our own Galaxy and—strike us.
 CHAPTER 17 ; «

Energetic Photons
The Electromagnetic Spectrum |

But let us now return to the photon. Until 1800, the

only photons known to man were those of visible light,
-which he could sense directly. The wavelength of such
light varies from 0.000076 centimeters at the red end of
the visible spectrum to half that value, or 0.000038 centi-
meters at the violet end. The energy of light photons is
_ inversely proportional to the wavelength of the light. If ©
extreme violet light has half the wavelength of extreme.
red light, then those violet light photons -have twice the
energy content of the red light photons. The energy —
content of the photons of visible light varies from 1.5
electron volts (1.5 ev) at the extreme red end of oie: n
spectrum to 3.0 ev at the extreme violet end. ,
In the opening years. of the nineteenth century, infvowed
and ultraviolet radiation were discovered (see page 76).
The energy content of infrared photons was, naturally, less
than 1.5 ev, while that of ultraviolet photons was more
than 3.0 ev. :
How far the infrared region of the spectrum. might
extend in the direction of lower and lower energies and
how far the ultraviolet region might extend toward higher
and higher energies was unknown through most of the.
nineteenth century.
In 1861, however, the Scottish “physicist James Clerk
Maxwell (1831- 1879) evolved an overall theory of elec-
tricity and magnetism that showed the close and, indeed,
inseparable relationship of these two types of energy.
(One can only speak of an “electromagnetic field” as a
consequence, taking the two types of energy together.) He ©
also showed that periodic variations in the intensity of
i 253
 254 _ ‘The Universe
such a field would produce a wave-form that would recede
from the source of variation at the speed of light. Indeed,
light itself was considered a form of such “electromagnetic
Tadiation.”
Since the electromagnetic field can vary with any peri-
od, the electromagnetic radiation can have any wave-
length. There should, therefore, exist electromagnetic ra-
diations with wavelengths far longer than even the in-
frared, and others with wavelengths far shorter than even
the ultraviolet.
It did not take long for the prediction to be verified. In
1888, the German physicist Heinrich Rudolf Hertz (1857-
1894) produced electromagnetic waves with enormous
wavelengths. Such radiation (called, at first, “Hertzian
waves”) came to be used for radiotelegraphic communica-
- tion—that is, communication not by electric currents
along wires as in ordinary telegraphs, but by waves radiat-
ing (hence the “radio-”) through space. Naturally, one
might expect to have such radiation called radiotelegraph-
ic waves for this reason, but the shortened form “radio
waves” came into fashion. i
In 1895, the German physicist Wilhelm Konrad Roent-
gen (1845-1923) demonstrated the existence of a form
. Of radiation that turned out to be electromagnetic in
nature and to have extremely short wavelengths. He called
them “X-rays” as a confession of ignorance as to their
nature, and the name stuck, even after the ignorance had’
vanished.
The three varieties of radiation from radioactive sub-
stances (discovered first in 1896 by Becquerel) were
named by Rutherford after the first three letters of the
Greek alphabet: “alpha rays,” “beta rays” and “gamma
rays.” Of these, the gamma rays proved to be electromag-
netic in nature; a form of radiation with wavelengths even
shorter than those of X-rays.
By the beginning of the twentieth century, then, physi-
cists found themselves in possession of an enormous “elec-
tromagnetic spectrum” stretching over some sixty octaves—
that is, with ‘wavelengths doubling sixty times as one
progressed from the shortest to the longest. The longest
waves were therefore 2°° or about 1,000,000,000,000,000,-
000 (a billion billion) times as long as the shortest. Of this
vast range, visible light covered only a single octave.
The electromagnetic spectrum is continuous, and there
are no gaps between one form of radiation and another.
 Pe ae et teh eh Ne a a at ee re
FA BE ie sek Vi = heal . , Live 16 =

‘Energetic Photons Pee LOS

The boundaries man sets are purely arbitrary, depending ©
on his ability to sense one small portion directly, and on
the accidents of discovery outside that portion. Ordinarily,
these arbitrary boundaries are described in terms of wave-
length, or of frequency (the number of wavelengths
produced per second). I will, instead, describe them here ~
in terms of the energy content of the photons making
them up, a value that is directly proportional to the
frequency.
The electromagnetic radiations of longest. wavelengths
(and, therefore, made up of photons of least energy) are -
the radio waves. At the broadest extent, they contain
photons with energy contents of 0.001 ev and less. This
turns out to be an uncomfortably large range, and it is
often broken up into three regions: long radio waves,
short radio waves, and very short radio waves. The last
are now frequently referred to as “microwaves.” The
energy contents of the photons would be as follows:
Long radio waves—zero to 0.00000001 ev
Short radio waves—0.00000001 to 0.00001 ev
Microwaves—0.00001 to 0.001 ev
The infrared region can, in turn, be divided into the far
infrared, the middle infrated, and the near infrared, as we
move along the scale of shortening wavelength and of
increasingly energetic photons.
Far infrared—0.001 to 0.03 ev
Middle infrared—0.03 to 0.3 ev
Near infared—0.3 to 1.5 ev
The visible region is, as: stated before, in the range of
15 to 3.0 EV. By color, the energies ae be listed as
(on the average):
Red—1.6 ev
Orange—1.8 ev
Yellow-—2.0 ev
Green—2.2 ev
Blue—2.4 ev
Violet—2.7 ev
Electromagnetic radiations made up of photons with
energies higher than those of visible light include, in or-
der, the near ultraviolet, the far ultraviolet, X-rays, and
gamma rays:
Near ultraviolet—3 to 6 ev
Far ultraviolet-—6 to 100 ev
X-rays—100 to 100,000 ev
Gamma rays—100,000 ev and up.
 256 -° The Universe
Neutron Stars

Naturally, the question arises as to how far over the

whole electromagnetic spectrum the observed spectra of
the Sun and stars extend. It is certain that the Solar
spectrum is not confined to the visible octave, for both
infared and ultraviolet radiation were first discovered in
the Solar spectrum.
But there is a sharp limit to how far beyond the visible.
edges of the Solar spectrum investigations can be carried
on, at least on the Earth’s surface. The atmosphere, while
transparent in the visible light region; is quite opaque for
-almost all other sections of the electromagnetic spectrum.
Though the Sun might well be rich in radiation in the far
ultraviolet and far infared, such radiation would neverthe-
less not reach us under our blanket of miles of air, and
until well into the twentieth century, nothing much could
be done about that.
_ By the mid-twentieth century, however, man’s technolo-
gy had made the atmosphere no longer the impenetrable
_ barrier it had once been. Airplanes could climb into the
Stratosphere and remain there for hours; balloons could
Tise even higher and remain there for days: rockets and
satellites could rise beyond the atmosphere altogether and
Temain there for weeks and months, even years.
Astronomic observations made from the stratosphere
and beyond could take advantage of the entire energy
range of photons, and not merely of those few varieties
that could penetrate, relatively uninterfered with, to the
bottom of the ocean of air under which we move and live.
In 1964, for instance, a balloon-borne telescope studied
the infrared portion of the light reflected by the planet
Venus. The absorption bands present in the infrared spec-
trum of the planet made it quite plain that its clouds
contained ice crystals and were therefore probably made
up of water. It is difficult or impossible to reach such a
conclusion, unmistakably, from the surface of the Earth.
From the surface, the necessary portion of the spectrum
remained hidden, and any effect of the water content of
Venus’ atmosphere on that portion of its spectrum visible
from Earth’s surface wasbound to be more or less
masked by the water vapor present in Earth’s own atmos-
phere.
Again, from balloons, photographs of the Sun could be
taken that were much sharper than any that could be
“
 _ Energetic Photons 257
taken from the Earth’s surface. The Solar spectrum could
be carried far out into the ultraviolet and, especially after
rockets came into use, thousands of absorption lines could
be recorded and measured that would have been forever
invisible from down here.
Surprisingly, the Sun’s spectrum was discovered, in
1949, to spread into the X-ray region to a much greater
extent than would have been thought possible. In general,
' the higher the temperature of a. body, the greater the .
quantity of high-energy photons it emits. This was, in
effect, shown by Wien (see page 133) in the days before
the existence of photons was suspected. And-the Sun’s
surface temperature was certainly not high enough for
X-rays.
The Sun’s surface temperature of 6000° C. placed the
_ peak of its radiation in the visible light region, and the
photons it emitted most profusely from the surface would
_ have an energy content of several electron-volts. To emit
large quantities of X-ray photons with energies of hun-
dreds and even thousands of electron-volts meant that
much-higher temperatures had to be available. Such tem-
peratures did not occur on the untroubled surface of the
Sun, but they did occur in the Sun’s “corona” (its rarefied
outer atmosphere). Here a thin scattering of particles was
whipped into extremely rapid motion, probably by shock
waves from the lower atmosphere. The energies imparted
to these particles were equivalent to temperatures of 500,-
000° C., at least. When a Solar flare erupted the tempera-
_ ture of the corona above the flare would rise ee into
the millions of degrees.
To be sure, although individual particles of the corona
were exceedingly energetic, the total heat of the corona
was small because the total number of particles was com-
paratively small. The heat of the corona carries no danger
to the Earth, therefore. (Indeed, there is some reason to
think that the outermost wisps of corona extend outward
from the Sun to beyond the Earth’s orbit so that, in a
sense, the Earth actually revolves within the Sun’s atmos-
phere—but without being sensibly affected thereby.)
It is the corona and the flares that are the source of
Solar X-rays. And it is the high temperature- equivalences
of these regions that strip so many electrons from atoms
as to produce utterly novel spectral lines—lines that in
earlier years were interpreted as representing a new ele-
ment called coronium (see page 144).
 bree ogee
258 The Universe :

The search for energetic photons from sources other

_than the Sun was also undertaken. Energetic photons
would have enormous advantages over the energetic parti-
cles of cosmic rays in some ways. Photons, being
uncharged, would be unaffected by magnetic fields and
would travel in undefiected trajectories. Photon source
could, therefore, be identified; at least, the direction from
which they approached could be determined. Thus, regions
particularly rich in ultraviolet radiation were uncovered in
1956 and afterward, and it was not difficult to tell that the
most prominent of these were centered in the constella-
tions Orion and Virgo. a
In Orion, the energetic radiation seemed associated with
the luminous nebulae surrounding particularly hot stars. In
a way, these nebulae seemed to be analogous to vastly
extended coronas, heated by the superabundant energy of
the stars within, in the same manner that the solar corona
is heated by the Sun.
Again, X-ray photons produced some surprises. The Sun
radiated X-rays, true, but only to a minor extent, in the
absence of flares. Solar X-rays were detected only because
.the Sun was so close. If other stars did no better, they
could not be expected to register their X-rays over vast
interstellar distances. Ses
Nevertheless, a team of investigators, including Bruno
Rossi, was interested in finding out whether the Solar
X-rays might be reflected from the Moon (the surface of-
which, unlike that of the Earth, was not protected by an
atmosphere). Special rockets sent up in 1962, with instru-
ments designed to detect X-rays, managed to detect them,
not from the Moon, but from the general direction of the
Galactic center.
The next year rockets were sent up by a group under
the leadership of the American astronomer Herbert
Friedman (1916- ). It was their intention to scan
the sky for X-ray sources and pinpoint their locations as
closely as possible. Some dozen regions showing X-ray
activity were detected in this way over the next couple of
years. The strongest of these was found in the constella-
tion Scorpio, and it is this that may have tripped the
detecting devices in that first 1962 rocket flight. The
source 1s not associated with anything that can be iden-
tified optically,
The second strongest source, one-eighth the strength of
was
Energetic Photons 259
the Scorpio source, Seemed to come from the Crab Ne-
bula.1
The nature of the objects radiating so strongly in X-rays
that they could be detected across many light-years of
space was a most intriguing question. Again, temperatures
in the millions of degrees had to be evolved, but the total
energy would have to be many times greater than that of
the Sun’s corona to produce so much. One might almost
- expect that a star’s core would have to be exposed.
Such temperatures might arise out of catastrophic con-
tractions even greater that those involving the conversion
of supernovae into ordinary white dwarfs. The degenerate
matter that composes a white dwarf is made up of pro-
tons, neutrons, and electrons. Of these, the electrons are
the crucial components. They are less compact than are
the protons and neutrons and therefore resist compression
more than the latter do. As long as the mass of the white
dwarf is no greater than 1.4 times that of the Sun (see
page 178), the electrons will manage to keep the star
expanded to planetary volumes, at least, even against the
enormous compressing force of its extremely compacted
gravitational field. If the white dwarf possesses a mass
greater than 1.4 times that of the Sun, then even the
electrons could not resist the compressive force of the still
more intense gravitational field.
An ordinary star with a mass more than 1.4 times that
of the Sun would, in undergoing a supernova explosion,
lose a large part of its mass, and what is left may usually
be expected to be below the crucial mark for white dwarf
formation. But what if it is not?
In that case, compression does not stop at the white
dwarf stage. There is further compression; the electrons
melt into the protons, forming neutrons, and the whole
comes together into a mass of neutrons-in-contact. Such a
“neutron star” would contain all the mass of a couple of
Suns in a sphere not more than a dozen miles in diameter.
It would be made up of neutronium, or of Gamow’s ylem.
It would be a tiny fragment of the substance thought by
those favoring the big bang theory to have made up the
cosmic egg. The very existence of such neutron stars
1 The Crab Nebula added to its remarkable properties in another
way in 1964. In that year, cosmic-ray counts taken by balloon
showed higher value when the Crab Nebula was in view of the
detectors. If this is not coincidence, then the Crab Nebula is the
only known specific cosmic-ray source other than the Sun itself.
 260 The Waiveee
aula be a small point in favor of the big bang. theory for.
that very reason.
Theorists suggest that a neutron star, for a period after
its formation, would have an equal temperature through-
out its structure, a temperature of about 10,000,000° C.
It would consequently release a furious flood of X-rays
and would thus be a most effective X-ray source. A
neutron star would therefore be an “X-ray star” as well.
Moreover, if supernovae turning into neutron stars
were the actual sources of X-rays, then those sources
should be concentrated toward the plane of the Milky
Way; this actually seems to be the case with the first
X-ray sources discovered. However, by 1966 at least two
of the newer X-ray sources detected seemed to be associat-
ed with galaxies: Cygnus A and M-87. These were the first
“X-ray galaxies,” both of which emit radio waves, too.
There may be as many as 10,000 detectable X-ray galaxies
in the Universe.
The question of the existence of a neutron star must be
put, if possible, to the test of observation. If the X-ray
source is really a neutron star, then the X-rays are emerg-
ing from a point in space. No possible detecting device we
could build would make a body a mere dozen miles in
diameter anything more than a point when it is at a
' distance of thousands of light-years from us as the Crab
Nebula is. On the other hand, if the X-rays were emerging
from a comparatively large area of space, then the source
is likely to be a turbulent region of gas and dust, and the
case for a neutron star is weakened. (It is not entirely
eliminated, however, since the area of gas and dust might —
still ee a neutron star, with both radiating X-
Tays.)°
It was possible to differentiate between a point source
and an area source of X-rays as a result of an interesting
astronomic accident. The Crab Nebula is situated in such
a position that it is periodically covered (or “occulted”)
by the Moon. As the Moon passes before it, the X-ray
source might be cut off instantaneously, indicating it to be
a point, or it might be cut off slowly and gradually,
indicating an area.
In 1964 an occultation of the Crab Nebula by the
Moon was scheduled to take place. If astronomers missed
their chance, they would not have another for eight years.
A rocket was sent up in time by Friedman’s group, and its
instruments, worked. The X-ray flux was found to fade out
gradually, and the source was shown to be an area about
>a
Energetic Photons 261
a light-year in diameter at the center of the Crab Neb--
ula.
A more serious blow at the neutron star hypothesis
came in 1965 when new calculations, taking additional
factors into account, seemed to show that if a neutron star
were formed, it would cool off in a matter of mere weeks
to the point where it was no longer hot enough to radiate
X-rays. If so, the chances of finding a neutron star in just
that brief interval of X-ray emission would be so small as
to be not worth considering.
It seems more likely, at the moment, then, that X-rays
are produced by the same processes (on a tremendously
larger scale) that work in the Solar corona. This seems
plausible enough for Crab Nebula where the fury of a
gigantic supernova still leaves its mark, but what about the
other sources where no such fury is visible?
Here the steady-state theory comes into play. I have
stated earlier (see page 235) that in the steady-state view
there might be a continuous creation of hydrogen atoms
(protons plus electrons) or of neutrons. In the former
case, the hydrogen atoms take their place in the Universe
without further change (until they find themselves in the
- core of a star). This is the “cold steady-state theory.” If
the continuous creation is of neutrons, the neutrons
formed break down in a matter of minutes to protons and
electrons, liberating energy in the process. This is there-
fore the “hot steady-state theory.”
The energy radiated by neutron breakdowns might be
sufficient to groduce X-rays generally throughout space,
with local accumulations here and there, for some as yet
unspecified reasons, which would serve as the X-ray source-
es detected in the 1960’s. Unfortunately, the best calcula-
tions seem to show that the general X-ray background
produced by a hot steady-state Universe ought to be at
least one hundred times what it actually is. At the mo-
ment, then, the hot steady-state seems out, and if there is
a steady-state Universe at all, it must be a cold steady-
State.
Which, of course, leaves the question of the nature of
the X-ray sources unanswered.

Antimatter

Gamma-ray photons, more energetic even than X-ray

photons, have also been detected in outer space, by means
of satellite-borne instruments. In 1961, twenty-two such
 Me SR oe ae
4 ee

262 The Universe

photons were detected by instruments on the Explorer XI

satellite. The direction from which these photons arrived
did not seem to be confined particularly to the plane of
the Milky Way, and the conclusion was that they came
from other galaxies. In 1965, the first point-source of
gamma rays was detected, so that now we will be dealing
with “gamma-ray stars.”
Naturally, one must try to think of a source for these
gamma rays. They seem to be rare events so that one
need not imagine a massive source but can postulate
isolated subatomic events. A collision of a highly energetic
cosmic ray particle with any atomic nucleus would set off
‘a train of consequences that would include the production
of gamma rays. The gamma rays we detect may, then,
mark the burial spot of dead cosmic ray particles.
Another possibility is more dramatic and requires a bit
of prologue. By 1932, three types of subatomic particles
were known. These were the proton (massive, positively
charged), the neutron (massive, uncharged), and the elec-
tron (light, negatively charged). Of these, atoms are com-
posed; and of atoms, matter is composed.
In 1930, however, the English physicist Paul Adrien
Maurice Dirac (1902- ) had suggested, out of pure-
_ ly theoretical considerations, that for each type of particle
an “antiparticle” ought to exist. These antiparticles would
be marked by key properties exactly opposed to those of
the corresponding particles.
Thus, equivalent to the negatively charged electron
would be an “antielectron” similar in all respects except
for the possession of a positive charge equal in size to the
electron’s negative charge. Similarly, balancing the posi-
tively charged proton would be a negatively charged “anti-
proton.” The neutron is uncharged, but it has a magnetic
field oriented in a certain direction. Balancing it there
would be an uncharged “antineutron” with a magnetic
field oriented in the opposite direction.
It seemed a rather far-out suggestion at first but in
1932, the American physicist Carl David Anderson (1905-
), in the course of his studies of cosmic rays, discov-
ered the antielectron. In reference to its positive charge,
he called it the “positron,” by which name it is most
commonly known to this day although antielectron is the
more appropriate. The antiproton and antineutron were
detected in 1956.
The antiparticles possess all the properties of the parti-
cles (aside from their mirror-image reversals) and can do
 Energetic Photons 263
anything particles can do. If protons and neutrons can
combine to form atomic nuclei, there is no reason why
antiprotons and antineutrons cannot combine to form
atomic “antinuclei.” Indeed, in 1965, antiprotons and anti-
neutrons were combined at Brookhaven to form a combi-
nation made up of one of each. Since a proton plus a
neutron make up the nucleus of hydrogen-2, or deuteri-
um, the proton-neutron combination is called a
“deuteron.” The combination of an antiproton and an
antineutron is therefore a “antideuteron.”
Then, just as an atomic nucleus can surround itself by
electrons to form a neutral atom, so an atomic antinucleus
can surround itself by antielectrons (positrons) to form a
neutral “antiatom.” And as atoms make up matter, so
antiatoms make up “antimatter.”
The difficulty of constructing such antimatter in the
laboratory rests in the evanescent nature of the antiparti-
cles. Left to themselves the antiparticles would be as
stable as the particles to which they correspond, but they
are not left to themselves. When an antiparticle is formed,
it comes into existence in a Universe of ordinary particles;
it is isolated in a vast ocean of its opposites.
If an antielectron is formed, it is only a matter of time
(a millionth of a second or less) before it meets and
collides with an electron. The result is “mutual annihila-
tion.” The charges cancel, and the total mass of the pair is
converted into energy in the form of photons. The same is
true of the collision of a proton and antiproton.
The mutual annihilation of particles and antiparticles
offers a new and unprecedentedly powerful source of
energy. The most energetic ordinary nuclear reactions, -
such as those which fuse hydrogen to iron in stellar cores,
involve the loss of only some 1 percent of the total mass.
In the mutual annihilation of matter and antimatter, all
the mass is converted into energy. Mass for mass, then,
mutual annihilation produces a hundred times as much
energy as nuclear fusion.
The reverse process can also take place. A gamma ray
of appropriately high energy can be converted into an
electron-anti-electron pair. A still more energetic gamma
Tay can be converted into a proton-antiproton pair. (This
last process, long predicted by theory, was finally observed
in 1965).
But now adifficulty arises. Careful observation makes it’
appear as though there is a “law of conservation of
electric charge.” Negative electric charge can neither be
 a
eee
rere ae
ee ae . ,*

“2 264 The Universe

created nor destroyed by itself; nor can positive electric
‘charge. What can happen is that equal quantities of posi-
tive and negative electric charge can undergo mutual
annihilation and conversion to gamma ray photons, in
which case there is no destruction of net electric charge.
Similarly, positive and negative electric charge can be
created in equal quantities out of gamma ray photons so
that there is no creation of net electric charge.
But in that case, how is it that we are surrounded by a
Universe of matter, without any appreciable sign of anti-
matter? After all, any process that creates particles ought
to create antiparticles as well, and in equal quantities.
Still, how do we know that we inhabit a Universe made
up of matter only?
We can be sure that the Earth itself is exclusively
matter without any significant admixture of antimatter,
for if any antimatter were present, it would interact with ~
matter at once and disappear in a blaze of gamma rays.
The Moon, too, is matter, if only because man-made
rockets have landed on it without producing a colossal
explosion.
_ Meteorites are composed of matter? and the Sun is
matter, if only because the Solar wind is composed of
particles rather than antiparticles. Consequently, it is safe
to conclude that the Solar system is matter.
Since the cosmic rays are composed almost exclusively
of particles and not antiparticles, we can even say that the
Galaxy (and perhaps nearby outer galaxies as well) is
matter.
But in that case, how can one explain the nonexistence
of antimatter?
Consider the steady-state model of the Universe. If
there is a continuous creation of hydrogen atoms (protons
and electrons), why is there not an equivalent creation of .
hydrogen antiatoms (antiprotons and antielectrons)? If
there is a continuous creation of neutrons, which then
break down to protons and electrons, why is there not a
continuous creation of equal numbers of antineutrons that
break down to antiprotons and antielectrons?
We know of no reason to explain the preference of one
over the other, and all the evidence gathered by physicists
* Occasionally, the existence of large meteorite strikes without much
in the way of meteoric matter at the site of the strike leads to
speculation concerning the possible landing, once in a long while,
of an antimatter meteorite.
 Energetic Photons _. 265
so far would lead us to believe that no preference can be
shown. :.
One possibility, of course, is that both matter and anti-
matier are indeed continuously created but that some
mechanism balances the process in such a way that they
are created in different places. An atom may come into
existence here and be balanced by an antiatom coming
simultaneously into existence there. In that case, perhaps,
there are both matter-galaxies and antimatter-galaxies; or,
to put it more compactly, both galaxies and antigalaxies.
If this is so, could we distinguish between a galaxy and
an antigalaxy?
Not, apparently, through the light they emit. A photon
is its own antiparticle, so that both matter and antimat-
ter, interacting among themselves, produce photons of
identical nature. The light from an antigalaxy is just like
the light from a galaxy.
The question of gravitational effects is less certain.
There are speculations that while matter undergoes a
mutual gravitational attraction, and antimatter the same,
the gravitational interaction between matter and antimat-
ter is a-repulsion. No gravitational repulsion has ever been
observed, but then antimatter has never been studied in
quantities sufficient to produce a perceptible gravitational
field, so the matter must be considered unsettled.
If such gravitational repulsion exists, it should be en-
countered between galaxies and antigalaxies. None has
been, which might indicate that no antigalaxies exist in the
Universe. It might also mean that the effects would be
small indeed at intergalactic distances as seen across inter-
galactic gaps of space, and that such effects exist but have
not yet been observed or been interpreted properly.
Neutrinos offer a more definite hope. Galaxies release
floods of neutrinos and antigalaxies floods of antineu-
trinos. If areas of the sky can be pinpointed as rich
sources of antineutrinos, antigalaxies may be located.
However, neutrinos are extremely difficult to detect and
astronomic art is not yet at the point where such a feat is
practical.
Again, antigalaxies ought to produce cosmic-ray anti-
particles. Those likely to reach us would be very few in
number, but some might and those that do would be
useful. Antiparticles of billion Bev energies and more (the
energies required for cosmic rays to shake loose of their
galaxies and streak across intergalactic space toward us)
would be but little affected by galactic magnetic fields,
 266 The Universe
The direction of their arrival might serve to pinpoint
antigalaxies.
It is important to remember, however, that gaiaxies are
not in isolation, but that innumerable galaxies exist and
that some may interact despite the general expansion of
the Universe.A galaxy and antigalaxy might belong to the
same cluster, for instance, and might approach. If they
approach closely enough for their dust and gas to begin
intermingling at the fringes, vast quantities of energy
would be emitted at once. There are, in fact, cases of vast
energy-release in the depths of galactic space that might
just possibly indicate such matter-antimatter mutual anni-
hilation. I will return to this subject later in the book.
It may be that galaxies and antigalaxies are actually
protected from massive annihilation by such reactions at
the fringes. An analogy can be drawn with a drop of
water falling on a hot stove. The water does not evaporate
immediately in an explosion of boiling. Instead, it hops
and skips on the red-hot surface for a surprisingly long
time. The reason for that is that the portion of the drop
initially touching the hot surface vaporizes, and a cloud of |
steam pushes the drop upward and insulates it somewhat
from the heat.
In the same way, if a galaxy and antigalaxy approach,
the first mutual contact at ‘the fringe might produce a
flood of energy that will tend to keep them apart and, so
to speak, insulate them from each other.
In that case, though, the insulating interaction will be a
tich source of gamma ray photons, which would streak
through the Universe generally. The gamma rays detected
by Explorer XI could conceivably be derived from such
sources and could be a signal that the Universe contained
as many antigalaxies as galaxies. :
And how would that fit the big bang theory?
A contracting Universe containing as many antigalaxies
as galaxies ought to undergo more and more mutual
annihilation as the galaxies approach, and when the cos-
mic egg is formed it may consist of gamma rays only.
Perhaps it is the radiation pressure of these photons that
eventually forces the big bang itself, and in the first
Tavening moments of expansion, the gamma rays may give
‘Tise to an equal quantity of particles and antiparticles.
_ Again, these particles and antiparticles must be visual-
ized as being separated, or they will eventually fall back
Into mutual annihilation and gamma rays. We might pos-
tulate that for every particle formed on one side of the
 bi

Energetic Photons 267

exploding cosmic egg, a balancing antiparticle is formed
on the other side. The Austrian-American physicist Mau-—
rice Goldhaber (1911- ) has indeed suggested some-
thing of this sort, visualizing the formation of a Universe
of matter and an Antiuniverse of antimatter (which he
calls a “cosmon” and an “anticosmon,” respectively).
Do the two intermingle in the course of expansion to
form a combined Universe of galaxies and antigalaxies in
equal number? If so, we might picture a pulsating Uni-
erse, in which there is a merger of matter and antimatter
in the course of each contracting phase and a separation
in the course of each expanding phase.
Or perhaps the Universe and Antiuniverse experience a
mutual gravitational repulsion, separating entirely so that
our Universe is made up of matter only. Perhaps, in the
pulsating model of the Universe, the: Universe and Anti-
universe merge in the process of contraction to form a
single cosmic egg and separate again in the process of
expansion.
Or perhaps (and here is my own personal contribution
to the speculation on the problem—one which, to my
knowledge, has never been suggested elsewhere) the Uni-
verse and Antiuniverse are permanently separated and
pulsate in balance. One expands while the other contracts
and vice versa.
In such a “double-Universe model,” two apparent asym-
’ metries of our own Universe would be wiped out. The
double Universe would, as a whole, be perfectly balanced
in matter and antimatter, while leaving our own Universe
to be virtually pure matter. Again, the double Universe
would be perfectly balanced in radial motion. The Universe
and Antiuniverse taken together would be essentially stat-_
ic, while our own Universe would be left, at this point in
the cycle, to its expansion.
CHAPTER 18

Radio Astronomy
The Sun

While the cosmic rays, X-rays, and gamma rays that

reach Earth from outer space are intensely interesting to
astronomers, the true breakthrough of the mid-twentieth
century came at the other end of the spectrum—in the
long-wave, feebly energetic, radio wave portion.
There were two chief reasons for this. First, the atmos-
phere, which is transparent to visible light but opaque to
most other portions of the electromagnetic spectrum, hap-
pens to be transparent also to a broad band in the very
short-wave radio, or microwave, region. Astronomers
were thus offered a second “window” into the heavens.
Any microwave radiation from the skies could therefore
be studied at leisure from the Earth’s surface. There
would be no absolute need to send instruments aloft in
balloons or rockets.
Secondly, the use of radio waves in wireless communi-
cation led to the eventual development of refined
techniques for receiving and amplifying weak radiation of
this sort.
Indeed, the possibility of radio waves from the heavens
occurred to scientists quite early in the game. It was only
a few years after the discovery of the radio wave region -
of the spectrum that speculation arose as to the possibility
of detecting radio emission from the Sun. Among others,
the English physicist Oliver Joseph Lodge (1851-1940),
a pioneer in radio communication, tried to detect solar
tadio waves about 1890, but failed. Efforts in this direc-
tion then languished for a generation, and when success
finally came, it was by accident.
The discovery was registered by the American radio
268
 Mette aay sae se
Rte
269
| Radio Astronomy
engineer Karl Jansky (1905-1950) who, in 1931, was
engaged in the purely nonastronomic problem of counter-
ing the disruptive effects of static in radio communication.
There was one source of static he could not at first pin
down, and which he decided, at last, was due to interfer-
ence from an influx of very short radio waves from outer
space. He published his findings in 1932 and 1933, but his
-_papers roused little interest among astronomers.
The only person to take fire, in fact, was another
American radio engineer, Grote Reber (1911-- ). In
1937, he built a “radio telescope” in his. back yard, a
parabolic device thirty-one feet in diameter, designed to
gather the weak microwave radiation from space over a
sizable area and concentrate it on a receiving device at the
focus of the parabola. For several years, Reber painstak-
ingly located the radio sources in the sky. He was the first
and, for quite a while, the only “radio astronomer.” He
published his first paper on the subject in 1940.
Astronomers only slowly began to grow interested. One
_ trouble was that the very short radio waves from the sky
were much shorter than those ordinarily used in radio
communication, so that radiation from space did not ordi-
narily interfere with radio reception and, in this fashion,
_ force itself on man’s consciousness. Furthermore, technol-
ogy had not yet developed efficient ways of handling such
short radio waves.
In the late 1930’s, however, key steps were being taken
that eventually broke the dam. Great Britain and the
United States were developing “radar”—a device whereby
a beam of microwaves was sent outward with the expecta-
tion that it would strike an obstacle and be reflected in
such a way that a microwave echo might be detected.
_. From the angle from which the echo was received, the
direction of the obstacle would be determined. From the
time-lapse between emission of the original beam and
reception of the echo, the distance of the obstacle could
be determined (since microwaves travel at the velocity of
light). Radar was ideal for detecting distant objects accur-
ately and quickly, particularly under conditions in which
ordinary optical methods were inadequate. It worked as
well by night as by day; and whereas clouds and fog were
opaque to light, they were virtually transparent to mi-
crowaves.
Radar was used by the British to give early warning of
the arrival of German planes and was a crucial aid in
winning the Battle of Britain. Any interference with the
 s So aE

270 The Universe ©

working of radar was therefore bound to be of the most
intense interest to Great Britain and its allies, and in
1942, such interference occurred. The entire radar system
was jammed severely by a flood of extraneous microwave
radiation. Britain’s warning system was temporarily useless.
If this was deliberate jamming on the part of the Ger-
mans, the consequences could be most serious. Investiga-
tion revealed, however, that it was the result of the giant
Solar flare, which happened to give the first indication of
the existence of Solar cosmic rays (see page 250). The
flare had, apparently, sent out a flood of microwaves
toward the Earth that had succeeded in handily drowning
the man-made radiation that fed the radar system. -Thus,
it was discovered that the Sun radiated in. the radio wave
region of the spectrum. By the time World War II was
over, astronomers were ready to turn to “radio astrono-
my” in earnest.
Once the radio wave spectrum of the Sun was placed
under study by techniques that had reached high refine-
ment in connection with radar technology, it was quickly
apparent that the Sun was emitting far more microwave
radiation than could be accounted for by its surface tem-
perature. Some wavelengths were being emitted at intensi-
ties that would only be produced at temperatures of 1,-
000,000° C. or so. These, of course, were emitted by the
corona, which was hot enough to radiate X-rays as well
(see page 257). ,
The Sun also emitted bursts of high-intensity microwave
radiation in connection with sunspots, flares, and other
disturbances.

The Planets

Nor is the Sun (surprisingly enough) the only mi-

crowave source in the Solar system. The planets shine only
by reflected light, but some of them emit microwaves of
their own with sufficient intensity for them to be detected
on Earth.
In 1955, for instance, Jupiter was recognized as the
source of certain microwave bursts that had been puzzling
observers over a period of five years. Some of the general
microwave emission of Jupiter was thermal in origin; that
is, it originated simply because Jupiter’s surface was at a
certain temperature and therefore radiated energy over a
broad band of the electromagnetic spectrum—a band that
included the microwave region. At certain wavelengths,
 Radio Astronomy 271
however, the radio emission was much more intense than
_ could be accounted for in purely thermal fashion. (After
all, no one expected Jupiter to have a high-temperature
corona as the Sun did). The nonthermal radiation was -
eventually explained by postulating the existence of a
Jovian magnetic field far larger and more intense than
that of the Earth, as I shall explain in the next section
(see page 277).
Equally interesting were the microwaves received from
Venus. These were first detected in 1956 and presented
astronomers at once with an interesting discrepancy. The
measurement of ordinary infrared radiation from Venus
had indicated “surface temperatures” of about —43° C.
Microwave radiation, however, bespoke temperatures hun-
dreds of degrees higher, well above the boiling point of
water, in fact.
But was this really a discrepancy? The infrared radia-
tion from Venus had to originate in the planet’s upper
atmosphere. If it originated at or near the solid surface of
_ Venus, the infrared radiation would be absorbed by its
atmosphere. The atmospheres of both Venus and Earth
are, however, transparent to microwaves. Even Venus’
Opaque and permanent cloud cover, which has kept its
solid surface from ever having been seen by the eye of
man, is transparent to microwaves. There is a good
chance therefore, that the temperature indicated by in-
frared radiation is at the low level one would naturally
expect of the upper atmosphere, while the temperature
indicated by microwave radiation is that of the actual
solid surface. ;
Still a surface temperature well above boiling water is
rather surprising for Venus. Could it be that the mi-
crowave radiation is not thermal in origin but is, at least
in part, produced by a magnetic field as in the case of
Jupiter? The chances of the latter are small. It was widely
believed by astronomers that Venus rotated about its axis
very slowly. Since it is also strongly suspected that a
magnetic field about a planet originates only when it has a
rotation rapid enough to set up eddies within a molten
core, it did not seem likely that slowly rotating Venus
could have a significant magnetic field.
The matter was settled by the Venus-probe, Mariner II,
-a well-instrumented rocket which passed within 21,600
miles of Venus in December 1962. No significant magnet-
ic field was detected. If Venus had one at all, it could not
be more than a hundredth as intense as that of the Earth.
 272 The Universe
Mariner II data also showed the radiation was not coming
from Venus’ ionosphere but from its surface. The mi-
crowave emission must therefore be thermal in origin, and
Venus’ surface must be hot. Its intensity as measured from
Mariner II indicates a surface temperature for Venus of
approximately 400° C.
The Solar system can also yield information through
reflected microwave radiation. The first example of this
came in 1945 when microwaves were bounced off meteor-
ite showers. Such showers could in this way be detected
and studied even in broad daylight when they are ordinari-
ly invisible. To use obstacles still farther away as mi-
crowave reflectors required only refinements in technique—
the ability to send out very strong impulses and to detect
and amplify very weak echoes, from amid environmental
radiation of the same nature (“noise”).
The Moon, for instance, could be used as an obstacle
from which to reflect a microwave beam, and this was
accomplished for the first time in 1946. In 1958, echoes
were received from Venus, in 1959 from the Sun, and
since then from other members of the Solar system such
as Mercury, Mars, and perhaps even Jupiter.
The time-lapse between emission and echo-return can
be used to determine planetary distances and, indeed,
reflections from Venus gave a new and unprecedentedly
accurate method of determining the scale of the Solar
system. It represented a significant improvement over the
parallax determinations of the asteroid, Eros, a generation
before (see page 31).
Furthermore, microwave reflections can yield informa-
tion concerning the nature of the reflecting surface. If a
reflecting body were a smooth and perfect sphere, only the
portion directly facing the Earth would return an echo.
However, if the surface is rough and uneven, sloping
ground would be expected to return an echo toward the
Earth in regions where no echo would be returned if the
ground were smooth. This sloping ground would beatrifle
farther from the Earth, however, because of the Moon’s
curvature, and the echo would be smeared out somewhat,
lasting longer than the original emitted pulse. The mi-
crowave echoes would also be distorted in certain fashions
by Doppler effect if the reflecting object is rotating.
Naturally, much that radar can tell us about the Moon’s
surface can be checked by evidence of sunlight reflected
from that surface. This is not so in the case of Venus,
where the solid surface is hidden from us optically, but
 Radio Astronomy 273
where that surface can be touched by cloud-piercing mi-
crowaves. Thus, in 1965, microwave reflections seemed to ©
indicate the presence of at least two huge mountain ranges
on the surface of Venus, one running north and south, the
other east and west.
Even more interesting was the question of Venus’ rota-
‘tion. Since nothing could be seen on the globe of that
planet but a featureless cloud cover, there was no way of
determining accurately the period of its rotation. As late
as 1962, far more was known about the rotation of distant
Pluto than about the rotation of our nearest planetary
neighbor in space. Many guesses, or estimates from inade-
quate data, had been made, and the most popular sugges-
tion was that the period of rotation was equal to that of
the planet’s revolution about the Sun—225 days.
If this last situation were so, however, one side of the
planet should face the Sun perpetually and the other face
away from it just as perpetually (as one side of the Moon
faces the Earth perpetually and the other side faces
away). One would expect, then, that the “sunside” of
Venus would be exceedingly hot, while the “nightside”
would be exceedingly cold.
Microwave emissions, rather surprisingly, seemed to in-
dicate, however, that the temperature of Venus’ surface
did not vary as much as might be expected if there were
indeed a sunside and a nightside. One might have to

Microwave reflection

eadar pulse lost

radar pulse reflected

back to Earth
 274 The Universe 5
suspect the existence of strong winds that served to carry
heat from the former to the latter. Or else, perhaps
Venus’ period of rotation was not quite equal tc its period
of revolution so that there was no sunside or nightside, but
instead (as on Earth) every point on the planetary surface
was exposed to the periodic presence and, later, absence
of the Sun. ;
The latter suggestion is borne out by microwave reflec-
tions. Rather to the surprise of astronomers, Venus was
found in 1962, to have a period of rotation of 247 days
in the retrograde direction. That is, its surface turned in
the clockwise direction when viewed from high above its
north pole, rather than counterclockwise as in the case of
Earth and almost all the other planets. To put it another
way, Venus’ surface rotated from east to west rather than
from west to east as in our case. The period of revolution, — .
combined with the motion of the planet about the Sun
_means that from any given point of the planet’s surface,
the Sun would be seen (if the clouds did not exist) to rise
in the west and set in the east about twice each planetary
year. ;
Why Venus should rotate in the retrograde direction
and why it should be so hot now exercises astronomic
thinking but, again, better unanswered questions than no
questions. :
Nor is Venus the only planet about which notions con-
cerning rotation have had to be revised because of the new
Microwave techniques. In 1965, the planet Mercury was
shown by microwave reflection to have a period of rota-
tion that was not equal to its eighty-eight day period of
revolution about the Sun. This was even more surprising
than the case of Venus, for Mercury has no cloud cover
and its surface can be seen (albeit with difficulty in view
of its closeness to the Sun), so that its period of rotation
can be observed directly by following the shifting pattern
of its surface features. As long ago as the 1880's, the
Italian astronomer Giovanni Virginio Schiaparelli (1835-
1910) had studied those surface features and had
maintained that it rotated once per revolution. This had ©
been accepted for eighty years. :
When the microwave data came in, Mercury’s surface
was reobserved with painstaking closeness and Schiaparelli ~
was found to have been wrong. The mistake was, howev-
er, understandable. Mercury rotates in 58% days, or just
about two-thirds of its year. The particular face it shows
to the Sun at one perihelion, it therefore shows again at
 Radio Astronomy 275
the third return to perihelion, the sixth, the ninth, and so
on. Anyone making observations separated by some multi-
ple of three revolutions would see the same surface
features in the same place and could be pardoned for
supposing that the planet rotated once each period of
revolution since that is exactly what one would observe in
that case, too.

The Stars

Microwave emissions from the bodies of the Solar sys-

tem, however interesting, are by no means all there is to
radio astronomy. Indeed, the very first observations made
in the field dealt withmicrowave sources lying far beyond
the Solar system. E
Jansky’s first Gidenvstone were made at a period when
the Sun was relatively quiet and was emitting microwaves
in comparatively minor quantities. It was not therefore the
source detected by him. Jansky thought it was, at first,
since the source traversed the sky with the Sun. However,
as the days passed he noted that the source gained on the
Sun by four minutes a day. That meant that the source
Was maintaining a fixed position with respect to the stars,
and as it turned out it was located in the direction of the
constellation Sagittarius. There was no question that the
source was the center of the Galaxy.
This was a discovery of the first importance. The clouds
of dust that forever blocked off the center of the Galaxy
from optical study were quite transparent to microwaves.
If we could never see the Galactic nucleus in the ordinary
way, we could “see” it by microwaves.
Reber, in the course of his lonely investigations of the
microwave emissions received from the heavens, plotted
what might be called the “radio sky,” the background
intensities of microwave emission from place to place. The
chief feature of the radio sky was the band of high
emission that marched along the Milky Way, with its
greatest intensities in the direction of the Galactic nucleus
and falling off on either side to a minimum in the direc-
tion opposite that of the Galactic nucleus.
However, the plane of the Milky Way did not by any
means include all there was to the radio sky. There were
knots of high intensity here and there, even in places far
-removed from the Milky Way. Most of these radio sourc-
es could not be identified, at first, with any visible
feature, but it was plain they could not represent ordinary
 276 ‘The Universe
stars like our Sun. If stars generally emitted microwaves
only as the Sun did, the Galactic center could not radiate
enough microwaves to reach us with the intensity they
do.
The unusual nature of the radio sources was made
perfectly plain by the fact that one of them was quickly
identified with the Crab Nebula. I have already discussed
that object as a source of X-rays and cosmic rays (see
page 259). It proved also to be a source of microwaves,
the third strongest source outside the Solar system.
At first, one might suppose that the microwave emission
of the Crab Nebula, far far stronger in intensity that that
from our Sun, was merely the result of the same high
temperatures that produced X-rays and cosmic rays. Ap-
parently not, however. If the temperature was high
enough to produce microwaves in the intensities observed,
the Crab Nebula would have to be much brighter optical-
ly. Secondly, while the intensity of microwaves produced
by high temperature should decrease as wavelength in-
creases, this does not happen in the case of the Crab
Nebula. ; 4
A suggestion arose out of findings in the field of nuclear
physics. It takes energy to make a moving body change
its speed or direction of motion, and the moving body, in re-
sponse to this, can radiate away some of the energy. This
was noted particularly where instruments called “synchro-
trons” made use of intense magnetic fields to whirl elec-
trons in circles with energy being poured into them con-
stantly. The electrons emitted “synchrotron radiation” with
‘wavelengths and intensities that depended on their energy
content,
In 1953, the Soviet astronomer Iosif Samuilovich
Shklovsky (1916- ) suggested that the Crab Nebula
might have an intense magnetic field and that the magnet-
ic lines of force might cause electrons to spiral, emitting
synchrotron radiation as they did so, and that this radia-
tion might include microwaves as well as light.
If this were so, then the wave forms emitted would be
guided by magnetic lines of force oriented in a fixed
Manner in space. The wave forms would themselves,
therefore, have a fixed plane, and the light from the Crab
Nebula would have the property of being “polarized.”
Astronomers can test whether light is polarized or not,
and the Soviet astronomer V. A. Dombrovsky was the
first to show that indeed it was,.a point quickly confirmed
by other astronomers, Views of the Crab Nebula taken ©
 Radio Astronomy ie S2TT
through Polaroid filters oriented in given directions, show
straight-line features in each case, in directions perpendic-
ular to that orientation. That would be expected in the
case of polarized light.
It was the success of this view in connection with the
Crab Nebula that led to the suggestion that some of
Jupiter’s nonthermal emissions might be synchrotron radi-
ation (see page 271). This meant that Jupiter wotild have
to have a magnetic field some tens of times stronger than
the Earth’s, and in view of Jupiter’s rapid rotation (ten
hours as compared with the Earth’s twenty-four, although
Jupiter has eleven times the diameter of the Earth) that
seemed quite likely. By 1960, Jupiter’s microwave emis-
sions were found to be polarized in the direction that
would be expected if Jupiter’s magnetic poles were near its
geographic poles, as is true of the Earth.
By that time, the efficiency with which particles were
trapped by magnetic lines of force had come to be under-
stood, and the magnetosphere about the Earth (see page
249) had been discovered. A much more intense magneto-
sphere about Jupiter had to be inferred. Moreover, in
1962, when a nuclear bomb was exploded high above the
atmosphere, the Earth’s magnetic lines of force were
flooded with charged particles which were trapped and
which, as they spiralled back and forth about the lines,
actually emitted detectable synchrotron radiation. That
filled out the picture, and the synchrotron radiation theory
of microwave emission by the Crab Nebula (and many
other radio sources) came to be generally accepted.
The Crab Nebula is, of course, almost certainly the
remnant of a supernova, and other radio sources can also
be identified with supernovae known to have exploded in
our Galaxy. Those of Brahe and of Kepler are examples.
The most powerful radio source of all is not, however,
connected with a known supernova. It is called “Cassio-
peia A,” for it occurs in that constellation. Nothing start-
ling corresponds optically to Cassiopeia A; all that shows
up are clouds and wisps of gas about 10,000 light-years
from us. This gas, on closer study, proved to be ferocious-
ly hot and in violent movement. It may well be the
remnant of a supernova that exploded as recently as 1700
but went unnoticed because its distance kept it from being
particularly bright, at a time when interest in “new stars”
was so small that only particular brightness would have
brought it to the attention of observers. Another interest- .
ing radio source is IC443, also a nebula-that may be a
 278 : The Universe

‘supernova remnant, one that is perhaps 50,000 years old.

The white dwarfs associated with such nebulosities cannot
be seen—again because of distance. -
It is reasonable to suppose that the microwave emission
of the Milky Way generally might be produced by the
supernovae that have appeared within it, but they need
not be the only sources. There is a type of red dwarf star
that flares up occasionally in an irregular fashion. Pre-
sumably these stars brighten by means of flares, like those.
of the Sun, only more intense, and these flares liberate
microwaves as do those of the Sun. Cooperative work
between the English astronomer Alfred Charles Bernard
Lovell (1913-1) and the American astronomer Fred
Lawrence Whipple (1906- ) showed that the intensi-
ty of microwave emission did indeed parallel the brighten-
ing effect, and these “flare stars” were the first reasonably
ordinary individual stars to be identified as radio sources. .

The Galaxy

Not all microwave emission from the Galaxy originates

in stars, or in the remnants of supernovae. There is also
the matter between the stars, the thin interstellar gas that
consists mostly of hydrogen. If this hydrogen happens to
be heated by some nearby star, the atoms can be pumped
full of energy and ionized. This energy can be radiated so
that the astronomer can observe luminous clouds and
detect spectral lines associated with hydrogen.
This is better than nothing, but not much better, be-
cause only a small proportion of the interstellar hydrogen
of the Galaxy is heated sufficiently to produce these lines.
At least 95 percent of the hydrogen of interstellar space is
relatively cold and, radiationally speaking, quiet. Further-
more, the lines emitted by hot hydrogen can only be seen
where dark nebulae do not interpose their obscuring
clouds, which means that only our own section of the
Galaxy can be studied in this manner. ;
In 1944, however, a Dutch astronomer, Hendrik Christ-
offel van de Hulst amused himself with pen-and-paper
calculations concerning the behavior of cold hydrogen.
(He was forced into this by the fact that the German
occupation of Holland during World War II made ordi-
nary astronomic work impossible.) He worked out the
manner in which the magnetic fields, associated with the
proton and the electron in the hydrogen atom, were ori-
ented to each other, They could both line up in the same .
 Radio Astronomy 279
direction or in opposite directions. There is a slight energy
difference between the two, and every once in a while a
hydrogen atom in the less energetic form might absorb a
photon of just the right size that happens its way and
move into the more energetic form. Again, every once in
a while a hydrogen atom in the more energetic form
might emit a photon and sink to the less energetic form.
The energy taken up or given off is so small that only very
unenergetic photons are absorbed or produced, photons in
the microwave region with wavelengths of about 21 centi-
meters. ,
The emission or absorption of a 21-centimeter mi-
crowave photon by any one hydrogen atom ought to take -
place only very rarely, but there are so many hydrogen
atoms in space generally that a steady and perhaps detect-
‘able drizzle of such events must be taking place.
After the war was over, astronomers began to search
for some evidence of this, and in 1951, the Swiss-American
physicist Felix Bloch (1905- ) and the American
physicist Edward Mills Purcell (1912- ), working in-
dependently, detected the 21-centimeter radiation. Hydro-
gen absorption at the 21-centimeter mark was also detected
eventually.
Now a method existed for detecting the interstellar
hydrogen in space and for telling where it was present in
relatively thick profusion and where it was not. Further-
more, since the photons were those of microwaves and not
of visible light, they could penetrate dust clouds easily and
astronomers could “see” the interstellar hydrogen in many
parts of the Galaxy which were optically invisible.
Since it was to be expected that the dust and gas of the
Galaxy are concentrated in the spiral arms, the mapping
of the 21-centimeter sources ought to give one an idea of
the spiral structure of our Galaxy.
Actually, before the detection of the 21-centimeter ra-
diation, attempts had already been made to trace the
spiral arms by means of the luminous clouds of hot
hydrogen. They would surround the particularly hot Popu-
lation I stars that formed in the dusty arms. The line of
blue-white giants, lighting up the hydrogen all about for
light-years, would trace out the arms. Using this tech-
nique, the American astronomer William Wilson Morgan
(1906- ) and his associates prepared a map of the
spiral arms of our Galaxy in 1951.
Sections of three separate arms were marked out. One
of these included features in Orion and was therefore
 280 The Universe
called the “Orion Arm.” It includes our own Sun. Closer
to the Galactic center than that is the “Sagittarius Arm”
and farther out from the center than the Orion Arm is the
“Perseus Arm.”
Further investigations of this.sort made the map more
intricate, but then the technique became obsolete in the
light of the 21-centimeter radiation. Suddenly, it became
possible to work for much larger distances within the ©
Galaxy and in much greater detail. Maps were prepared
of the spiral structure of the Galaxy, and one could begin
to think of it, schematically, as a rather symmetrical
double spiral.
Nor is the cold, neutral hydrogen gas of the Galaxy
static. Studies by Oort and Van de Hulst seem to indicate
that the hydrogen flows outward from the center to the
outskirts of the system at a surprisingly rapid rate. Oort
estimates that the quantity of hydrogen transported each
year from the center outward is equal to the mass of the
Sun. This flow of gas outward along the spiral arms may,
according to some speculations, serve to keep the arms in
being, maintain their rich supply of gas and their ability to
form new stars. On the other hand, it is difficult to see
how the source of hydrogen at the center persists. It
should have run dry long ago, unless there is a general
Spiral armsoftheGalaxy (schematic)

Cuppis
210°
ERNE TRS srt iert ee
TSe at com ieee
. Radio Astronomy 281
circulation by which the supply at the center can be
replenished, perhaps at the expense of a gigantic “halo” of
hydrogen gas that seems to encompass the Galaxy gener-
ally. What keeps the hydrogen circulation in being is not
known as yet. .
Other galaxies also have their supply of hydrogen, of
course, and the amount may depend on the type of
galaxy. Spiral galaxies have more hydrogen, it seems, than
elliptical galaxies do. Study of the 21-centimeter radiation
- makes it appear that more open spiral galaxies have more
- interstellar hydrogen than tighter ones have, while irregu-
lar galaxies have most of all. What bearing this has, if
any, on the question of the evolution of galaxies, is, as
yet, uncertain.
Naturally astronomers are anxious to find other types of
microwave radiation that might be useful in studying in-
terstellar gas, and for that reason considered deuterium
(hydrogen-2). Hydrogen is by far the most dominant
component of interstellar gas, and a certain small percent-
age of its atoms must be hydrogen-2. The hydrogen-2
atom differs from the ordinary hydrogen-1 atom in having
a neutron added to the proton of the nucleus. The magnet-
ic field of the proton-neutron nucleus ought to interact
with the magnetic field of the electron as in the case of
- the simple proton nucleus, theory predicted, and a mi-
crowave with a 91-centimeter wavelength should be emit-
ted. Strong radio sources such as Cassiopeia A were
combed for the telltale wavelength, but the search failed.
Next to hydrogen and helium, the most common com-
ponent of interstellar gas is oxygen. An oxygen atom can
combine with a hydrogen atom to form a “hydroxyl
group.” This combination would not be stable on Earth,
for the hydroxyl group is very active and would combine
with almost any other atom or molecule it encountered. It
would, notably, combine with a second hydrogen atom to
form a molecule of water. In interstellar space, however,
where the atoms are spread so thinly that collisions are
- few and far between, a hydroxyl group, once formed,
would persist undisturbed for long periods of time.
Such hydroxyl groups would, calculations showed, emit
or absorb four different wavelengths of microwaves. Cas-
siopeia A was searched for these and in late 1963, two of
them—absorption lines in the neighborhood of the 55-cen-
timeter mark—were detected.
The hydroxyl absorptions turned out to be unexpectedly
useful. Since the hydroxyl group is some seventeen times
 282 The Universe

as massive as the hydrogen atom alone, it is more sluggish

and moves at only one-fourth the velocity of the hydrogen
.atom at any given temperature. In general, movement
_ blurs the line, and the hydroxyl absorption line is much
sharper than is that of hydrogen. It is easier to tell
whether the hydroxyl line has shifted slightly toward short-
er or longer wavelength and to determine, in that way,
the radial velocity of the gas cloud.
Furthermore, the ratio of hydroxyl groups to hydrogen
atoms varies, for some reason, from place to place, and
the relative quantity of hydroxyl groups seems to increase
rapidly as one investigates objects closer and closer to the
Galactic center. Astronomers hope, therefore, that the
hydroxyl group map of the Galaxy will pinpoint the center
sharply and make possible a clearer interpretation of the
events taking place there. However, in these early stages
of hydroxyl investigation, enough uncertainties and puzzles
are arising to lead some astronomers to dub the hy-
droxyl rich clouds “mysterium.”
 CHAPTER 19

The Edge of the

Universe

Colliding Galaxies

By the mid-1950’s nearly 2000 separate radio sources

had been marked out in the sky. A number of them were
obviously part of the Milky Way complex of general
microwave emission. About 1900 were not, however.
These were not smeared out over a sizeable area as the
“microwave emission from the Milky Way was. Rather, as
the Australian astronomer John G. Bolton first showed in
' 1948, they were point sources, with microwaves emerging
from small areas in the heavens.
It seemed logical to suppose such microwaves arose
from stars which, for one reason or another, radiated
heavily in that region of the spectrum. Indeed, Bolton
called them “radio stars.”
It was clear, of course, that such radio stars, if they
really existed, could not be ordinary stars, but were proba-
bly remnants of supernovae. Certainly this was true of the
Crab Nebula, which was the third brightest radio star, and
somewhat less certainly, of Cassiopeia A, which was the
brightest.
The trouble was, though, that very few radio stars could
be associated, in those early years of radio astronomy,
with any object that was visible optically, not even with
faint nebulous patches as in the case of Cassiopeia A. Part
of the trouble was that the sharpness with which any
object is viewed is dependent on the wavelength of the
radiation by means of which it is viewed. Radiation with
long wavelength makes for fuzzier vision. Microwaves are
roughly 400,000 times longer in wavelength than are the
light waves by which we ordinarily see the stars, and
“vision” is correspondingly fuzzier. The result is that
283
 284 The. Universe
viewing the sky by microwaves is like viewing it through
optical equipment that is badly out of focus. Instead of a
sharp point, we get a dim patch of fog, and where in that
fog is the actual point we are trying to see?
Radio astronomers had to locate their radio source as
well as they could, pin it down to a certain small area—
small to the naked eye, but gigantic to a powerful optical
telescope—and then try to see if somewhere in that area
something suspicious is visible in the light-wave region. If
there is, every attempt is made to sharpen the microwave
- focus and see if it seems to be zeroing in on the suspicious
site. Efficiency in this direction increased with passing
_ time as radiotelescopes grew larger and detecting equip-
ment more refined.
But as the years passed, astronomers grew increasingly
uneasy over the notion of radio stars. No matter how the
position of these objects was boxed in with increasing
accuracy, nothing visible within our Galaxy could be
pinned down, except in a very small minority of cases.
Worse still, the greater the number of radio stars located,
the more it became evident that they were spread out all
over the sky quite evenly, whereas all objects within our_
Galaxy from ordinary stars to supernovae remnants were
heavily concentrated in the plane of the Milky Way. In
fact, the only optically visible objects that were spread out
evenly all over the sky were the galaxies. Could it be then
that the so-called radio stars were galaxies? Ought one to
speak of “radio galaxies” rather than of radio stars?
The first real breakthrough in this direction came in
connection with the second brightest radio source, one
called “Cygnus A.” Microwave emission from its general
direction had been noted by Reber in 1944, but by 1948,
Bolton had shown it to be one of the radio stars. It was
the first source, in fact, that he could identify as sufficient-
ly sharp to warrant being called a radio star. By 1951, the
position of Cygnus A had been boxed into an area about 1
minute of arc squared. The problem, then, was to locate
something within that square.
Baade studied that square with the 200-inch telescope
and spotted an oddly shaped galaxy within it. On closer
investigation, it seemed to be not one distorted galaxy, but
two galaxies with their nuclei in near contact.
The explanation seemed clear. Two galaxies were in
collision! Just as there were catastrophes on the stellar
level which, as in the case of the Crab Nebula, resulted in
 The Edge of the Universe 285
an out-spewing of microwaves, so there were still more
colossal catastrophes on the galactic level, with still larger
microwave emissions.
And it seemed clear that the colliding galaxies were
indeed undergoing colossal travail. When the optical spec-
trum was finally obtained (a difficult task in view of its
faintness), that spectrum showed the lines of highly ion-
ized atoms, lines that could be present only if the temper-
ature were extraordinarily high. (Baade had bet Minkow-
ski a bottle of whiskey that this- would be the case, and
won.)
Suspicion arose at once that all or almost all the so-
called radio stars were actually galaxies in collision, and
the search was on for other cases of the sort; or, indeed,
for any “peculiar galaxy”—one with some oddity of shape
or structure that might indicate an unusual event on a
huge scale.
They were found in Conadeable numbers. More than a
hundred “radio galaxies” have now been identified, and
_many of them are peculiar indeed. There is, for example,
galaxy NGC 5128 which seems to be a spheroidal galaxy
with a thick band of dust running down its middle. It was -
suggested that this, too, might represent a galactic col-
lision and that what we saw was a spiral galaxy, viewed
edgewise with the dust of its arms obscuring its center,
knifing its way through the spheroidal galaxy.
Astronomers calculated the probabilities of galactic col-
lisions and decided they might be much more likely than
stellar collisions. Our Sun, for instance, is 860,000 miles in
diameter and is 25,000,000,000,000 miles from its nearest
neighbor. If this is typical, then the distance between stars
is just about 30,000,000 times their diameters. A star
moving randomly is much more likely, therefore, to pass
through the vast empty spaces between stars than to zero
in on the comparatively minute body of astar itself.
_ Our Galaxy, on the other hand, is 100,000 light-years in
diameter and our nearest large neighbor, the Andromeda
galaxy, is 2,300,000 light-years away. If this is typical,
then the distance between galaxies is about twenty times
their diameter. Space is much more crowded (relatively)
with galaxies, than galaxies are with stars, and intergalac-
tic collisions are correspondingly more likely than inter-
stellar collisions.
In the heyday that followed the discovery of the collid-
ing galaxies in Cygnus, it was calculated that in our own
 ran . , | OO

286 . The Universe

neighborhood of the Universe there ought to be five col-
lisions per billion galaxies, while within galactic clusters
collisions ought to be even more common. In a 500-galaxy
cluster in the constellation Coma Berenices it was esti-
mated that at least two collisions ought to be proceeding
at any given moment and that every galaxy in it ought to
undergo several collisions in its lifetime.
To be sure, when galaxies collided and passed through
one another, there is no question of wholesale stellar
collisions. The stars are widely separated compared to
their size and one galaxy can pass completely through
another without much danger that any of the stars of one
would actually collide with any of the stars of the other.
Nevertheless, the dust clouds of one are apt to collide and
pass through those of the other, and this might supply the
actual source of the microwave emission.
That this might be so was evidenced by the fact that as
the actual radio source in Cygnus was pinpointed further,
it seemed not to arise from the dust-free galactic nuclei in
collision, but from two points well on either side of those
nuclei—in the spiral arms, presumably, where the dust
would be concentrated.

Exploding Galaxies

The life of the colliding theory was merry, but short.

For one thing, the question of energy arose to plague
astronomers.
An ordinary galaxy, like our own, will emit about
10,000,000,000,000,000,000,000,000,000 (ten thousand
trillion trillion) kilowatts of energy in the form of mi-
crowaves. This is about the equivalent of a thousand
individual radio sources such as Cassiopeia A.
_ This is a comfortable fact. It is perfectly reasonable to
explain the microwave emissions of an ordinary galaxy by
supposing it to contain several thousand supernovae rem-
nants. Such a figure is certainly not excessive. The mi-
crowave emission of an ordinary galaxy represents only
about a millionth of the energy emitted in the form of
light and this, too, is easy to accept.
Among the radio galaxies, however, even the weakest
pour out into space a hundred times the microwave ener-
gies that ordinary galaxies do. Cygnus A radiates a million
times as much microwave energy as ordinary galaxies do.
Indeed, Cygnus A radiates about as much energy in the
form of microwaves as in the form of light.
eee ee
“The Edge of the. Universe 287
F
This began to seem more and more puzzling, and to
account for such large batches of microwaves became more
difficult, the more the matter was considered. It turned |
out, for instance, that the energy of microwave emission
of Cygnus A was about equal to the total energies of
motion of the supposedly colliding galaxies. It seemed
completely incredible that the energy of colllision could be
converted completely into microwaves. All the mass of the
galaxies would have to be brought to a standstill with
_ Tespect to each other-and how could that be done? By tens
_ of billions of stellar collisions? Impossible! And even if
that happened, how could all the energy come off as
microwaves? Surely much of it would be emitted as radia-
tion in other regions of the spectrum.
Furthermore, as the 1950’s progressed toward their
close, it became more and more generally accepted that ~
the microwave emission from the various radio sources
arose from the synchrotron radiation of high-energy elec-
_ trons trapped in a magnetic field (see page 277). This
meant that the kinetic energy of collision would not be
converted into microwaves directly but into very high-
energy electrons which would then be trapped in a magnet-
‘ic field. No reasonable mechanism could be advanced for
~ such a conversion of kinetic energy into high-energy elec-
_ trons.

Radio sources within Galaxies:

e

ae NGC fose

Za Ui
@ strong radio emission
YZzweakradio emission
 288 The Universe
There was observational evidence against the colliding-
galaxy theory, too. As more and more microwave sources
were pinned down to galaxies, it became more and more
difficult to interpret everything one saw in those galaxies
as representing collisions. Indeed, some “peculiar galaxies”
might be peculiar in the intensity of their microwave
emission, but they were not peculiar at all in shape or
appearance. They seemed perfectly ordinary galaxies,
abiding in single blessedness with no sign of any collision,
and yet they were strong microwave emitters.
An alternate view began to make itself felt. Perhaps it
was not a matter of the collision of two galaxies at all, but
the explosion of a single galaxy.
For instance, the galaxy NGC 1068 is a weak radio
galaxy with microwave emission only about a hundred
times that of a normal galaxy, but that emission seems to
arise entirely from a small area right in the center of the
galaxy. A collision of galaxies, involving dust clouds,
would be expected to spread the source over a wider
area and certainly not in the dust-free center. An ex-
plosion, on the other hand, would be expected to go off
right there in the center, where stars are most crowded
together and where some sort of catastrophe involving
large numbers of stars over a relatively short span of time
might most easily take place.’ If so, we may be viewing the
beginning of such a central catastrophe in NGC 1068. The
microwave emission is still highly concentrated in that
beginning-to-explode center, and it is still weak.
A larger stage in the process is perhaps displayed by
galaxy NGC 4486, which is better known as M87 from
‘its position on Méessier’s list. It, too, has an intense mi-
crowave source at the center, but it emits microwaves
more weakly throughout a halo about that center, one
that takes up almost the whole volume of what can be
seen optically. It is as though the ravening fury of the
central explosion may have spread out for tens of thou-
sands of light-years in every direction, and M87 emits
microwaves with a hundred times the intensity of NGC
1068. What is most interesting is that close optical study
of M87 shows that a luminous jet is emerging from its -
center. Can that be material hurtled out into intergalactic
space by the fury of the central explosion? The light from
this jet was shown, by Baade, to be polarized; another
piece of backing for ‘Shklovsky’s theory of synchrotron
radiation as the microwave source.
Perhaps a later stage still is one in which the main
 The Edge of the Universe 289
source of microwave emission moves out of the galactic
nucleus entirely and emerges on either side. In the case of
NGC 5128, which radiates microwaves as intensely as
M87, there are four regions of microwave emission. A
pair of intense ones are found, one on either side of the
central dust band; and a pair of weaker, more extended,
ones are on either side of what can be seen optically. The
source of microwave emission has split and moved apart
toward the edge of the galactic nucleus with some of it
having been hurled far beyond the nucleus in either direc-
tion. Can it be then that the dust band is not the rim of a
spiral galaxy moving into a spheroidal galaxy, as was first
suggested, but that it is instead the product of whatever
had been going on in the catastrophe-ridden core of the
galaxy? Could the dust band be a vast cloud of disinte-
grated star-stuff, perhaps, that happens to be hurled out in
our direction?
_ NGC 5128 is relatively close to us (only about fifteen
million light-years away), and it can be seen in some
detail. If it were much farther away, the dust band and all
that surrounded it would shrink until ,all that might be
made out would be two patches of light not quite
touching. The result might be interpreted as two galaxies
coming together broadside, like a pair of cymbals.
But this is exactly the interpretation of the galaxies in
Cygnus A. Perhaps that represents a case just like that of
NGC 5128, but one that is seen more dimly because it is
700 million light-years away and not merely 15 million. If
so, the explosion may be more advanced because now all
the material giving rise to microwave emission has been
hurled beyond the galactic nucleus, part to either side. The
Same is true of other galaxies where the radio sources
exist on either side of the galaxies themselves. The galax-
ies, nevertheless, still show signs of the catastrophe, for
their optical spectra indicate ferociously high tempera-
tures.
And perhaps the latest stage of all takes place when the
radio sources grow more diffuse and dim until they reach
the point where they can no longer be detected, and the
galaxy is once more (as far as we can tell by radio’
astronomy) a normal galaxy.
Yet, while the notion of colliding- galaxies slowly died
and that of exploding galaxies became prominent, the
evidence in favor of the latter remained almost entirely in
the form of deduction from the nature of the microwave
emission throughout the 1950’s. The only piece of optical
 290 The Universe
evidence in favor of the explosion theory was the case of
the jet in M87, and that was dubious because it was
emerging in only one direction, where one would expect
such a phenomenon to take place in two opposite direc-
tions. _ ;
The necessary optical evidence came as the 1960's
opened. In 1961, the American astronomer Clarence Rog- —
-er Lynds (1928- ) was trying to pinpoint a weak
radio source listed as 3C231. The area fuzzily covered by
the source included a number of galaxies in the constella-
tion Ursa Major, of which the largest and most prominent
- was M81. It had been supposed that M81 was the radio
source. However, as Lynds pinpointed the source more
carefully, it zeroed in not on M81 but on a smaller neigh-
boring galaxy, M82.
Certainly, M82 qualified as a peculiar galaxy, consider-
ably more so than M81. Earlier photographs taken of it
had shown it to be unusually dusty, and individual stars
could not be made out within it although it is only ten
million light-years from us (which is close enough to allow
some stars to show up). In addition, there were faint signs of
filaments of gas or dust above and below it.
Once M82 was pinpointed as a radio source, however,
its optical properties gained new interest. The American
astronomer Allan Rex Sandage (1926- ) took pho-
tographs with the 200-inch telescope, using a special red
filter that let through chiefiy the light associated with hot
hydrogen. He reasoned that if something were going on in
the center of that galaxy, something that spewed out
matter, then that matter would be chiefly hydrogen and it
would be seen most clearly if light from other sources
were excluded,
He was right. The galaxy, M82, was clearly and visibly
undergoing a vast explosion. A three-hour exposure re-
vealed jets of hydrogen up to 1000 light-years long, burst-
ing out of the galactic nucleus. The total mass of hydro- ~
gen being shot out was the equivalent of at least 5,000,000
average stars. From the rate at which the jets were
traveling and from the distance they had covered, the
explosion, as now seen from the Earth, must have been in
progress for 1,500,000 years. Apparently, this is still an
early stage of the process, too soon to develop the late-
stage appearance of a double source on either side of the
galaxy.
_ The light from M82 is polarized in such a way as to
indicate that the galaxy has a strong magnetic field.
 The Edge of the Universe 291
Again, the synchrotron radiation theory is backed. (In
- 1965, it was discovered that synchrotron radiation was
arriving from a halo about M81; perhaps as a response to
the outpouring of energy from its exploding neighbor.)
Can it be that galactic explosions are comparatively
common, that it is a stage that galaxies often pass
through—just as many stars may pass through a nova
stage? Has our own Galaxy passed through such a stage?
Has our Galactic nucleus exploded? If it has, the explosion
cannot have been a very large one, or a very recent one,
for there is no sign of strong radio sources on either side
of the Galaxy. On the other hand, there is the continuing
outstreaming of hydrogen from the center to the outskirts
of the Galaxy (see page 280). Is that a normal process
for galaxies generally, or is it a last dying effect of an
eons-old explosion?

The Distant Radio Sources

But what causes the explosions?

Since the most colossal explosion we know of, on a
sub-galactic scale, is the supernova, it seems reasonable to
suppose that a galactic explosion results from the simul-
taneous or nearly simultaneous explosion of a great many
_ supernovae. This was suggested by the American astrono-
mer Geoffrey R. Burbidge.
We can picture a large number of massive stars having
reached the stage where they are on the point of a
supernova explosion. If one of them explodes, its radiation
may heat up neighboring stars a trifle, pushing some of
them over the edge and driving them into explosions. They -
will, in turn, touch off others. This is most likely to happen
at the galactic center, of course, where the stars are most
closely spaced and where this domino effect of supernova
formation can proceed with the greatest speed. (It takes
time for radiation to cross the gap between stars.)
But even such multiple-supernovae supply barely
enough energy. It would take all the hydrogen in ten
billion stars like our own to supply the total microwave
energy released by a giant radio source such as Cygnus A,
even if it were supposed that all the energy from hydrogen
fusion would go into microwaves with 100 percent effi-
ciency. Such a catastrophe would involve every star in the
nucleus of a medium-sized galaxy.
This wholesale slaughter of stars is a rather drastic way
eRe? ; Rem a
292 The Universe tae
out of the difficulty. Is there any way we can economize
on stars?
Suppose, for instance, that what is happening is the
interaction of matter and antimatter (see page 263).
Matter-antimatter annihilation converts all matter in-
volved into energy, and not merely 1 percent of it as
nuclear fusion does. The energy of Cygnus A could then
be supplied not by ten billion supernovae, but by a mere
hundred million stars (half matter and half anti-matter)
undergoing mutual annihilation.
And yet this thought is not a comfortable one, either.
We have no direct evidence of the existence of antimatter
in huge masses. If such antimatter existed, it might explain
something like the jet in M87 which might be pictured as
caused by the invasion of a wedge of antimatter from
outside the galaxy (or the invasion of a quantity of matter
if M87 is itself antimatter). But how explain the appear-
ance of radio sources on each side of a galaxy? The
simultaneous and symmetrical invasion of antimatter
clouds from both sides would seem unlikely. If one sup-
posed the actual annihilation took place at the center and
the two clouds to either side were the result of an ex-
plosion, how could the antimatter get to the center of the
galaxy without first annihilating the outskirts?
Fred Hoyle points out, however, that one can go back
to gravitational. energy. If one begins with a mass great ©
-enough and postulates a sufficiently catastrophic collapse,
one can account for the production of enormous quantites
of energy. Indeed, the amount of energy produced in that
fashion per unit mass can be a hundred times as great as
that produced by ordinary nuclear fusion and therefore
just as great as that produced by matter-antimatter anni-
hilation.
One-can picture a galactic nucleus containing stars so
crowded together that the mutual gravitational field is
intense enough to overcome those factors that tend to
keep them apart. A number of stars start to move togeth-
er, intensifying the gravitational field, drawing other stars
into the vortex until, at the end, the mass of something
like a hundred million stars, forming a single super-star,
smash together into one huge neutron star. The energy
then produced would suffice to account for the existence
of Cygnus A.
The advantage of the gravitational theory over the
matter-antimatter theory is that no collision of any sort
 The Edge of the Universe 293
is required. A matter-galaxy can do it all on its own. So
can an antimatter-galaxy.
But are we correct in turning only to that which is
known? A century ago, Helmholtz tried to explain the
enormous energy output of the Sun by means of forces
then known and was forced to postulate a very short-lived
Solar system. The radiation of the Sun and other stars
could not be explained with real elegance until a new kind
of energy, nuclear energy, was recognized. In the same
way, does the still more massive output of certain radio
sources indicate kinds of energies we have not yet recog-
nized even today? Some astronomers cannot help wonder-
ing.
But whatever the source of the ferocious energy of the
intense radio sources, its mere existence is of great poten--
tial use to those interested in the problems of cosmology
and cosmogony.
Through the middle 1950’s, the most distant ordinary
galaxies that could be seen by optical telescopes were
perhaps two billion light-years away. That is about one-
sixth the distance to the edge of the observable Universe,
and this is not quite far enough to be able to tell distinctly
whether the Universe is SEE SEES: pulsation, or steady-
state (see page 230).
Viewing the Universe by microwave, however, gives
hope of a longer reach. Light sources are almost endlessly
many, while microwave sources are few. In our own
Galaxy there are billions of stars crowding in on telescopes
and only a hundred radio sources. Outside our Galaxy,
there are billions of ordinary galaxies but only thousands
of radio, galaxies. This means that individual microwave
sources, even though very distant, are not likely to be lost
in the clutter of nearby sources, and are therefore much
more easily detected and studied than equally distant opti-
cal objects are.
Thus, Cygnus A, which is 700,000,000 light-years away,
has so little competition, so to speak, from nearer radio
sources that it is the second brightest radio star in the sky.
Only Cassiopeia A is brighter. Furthermore, our radio
telescopes can pick up a radio source as intense as Cygnus
A at distances that would far outdo anything our best
optical telescopes could attain. The microwave emission of
Cygnus A could be picked up clearly at distances that
would reduce its light to an indetectably dim flicker. The
distribution of the very distant radio sources might, there-
fore, help us choose a model of the Universe, where the
aie Lakes Tae ae 27 OE ee ae pea

294 - The Universe

distribution of the less-distant ordinary galaxies, detectable _
only by light, would fail. “
As a first approximation one can assume that on the ©
whole fainter radio galaxies are more distant than more
intense ones, just as Hubble could work on the assumption
that dimmer galaxies were more distant than brighter ones
(see page 197). Using that assumption, the English as-
tronomer Martin Ryle (1918- ) attempted to an-
alyze the manner in which the number of radio sources
increased as their intensity diminished, much as Herschel
had once done in the case of stars (see page 59).
If the steady-state theory of the Universe. is correct,
then the average distance between galaxies has always
been what it is now. The radiation from distant radio
sources, which were first produced eons ago and therefore
represent a Universe eons younger than the Universe in
our own neighborhood, should show those radio sources to
be spread no more thickly through space than are the
radio sources among our own neighboring galaxies. In that
case the number of radio sources ought to increase, as
their intensity diminished, according to a fixed formula.
Herschel had found that in the case of stars, the in-
crease with dimness fell short of the formula, from which
he deduced a finite Galaxy. Ryle found the reverse. The
number of radio sources increased with dimmess more
rapidly than would be indicated by the formula. It seemed
that radio sources in the far distance were more thickly
spread than in our own neighborhood.
When he announced this in the mid-1950’s, his analysis
seemed to favor the big-bang and to indicate either a
hyperbolic or a pulsating Universe. After all, in the youth
of the Universe shortly after the big bang, the galaxies
were spaced more closely and therefore the radio galaxies
were spaced more closely, too. This means that in the far
reaches of space, where the radiation reaching us was
produced in the youth of the Universe, radio sources
would be more numerous than they are here.
_ Ryle’s data seemed, at that time, even to fit in with the
colliding-galaxy theory of radio sources. If the galaxies
-were closer together in the far past (and therefore in the
far distance), one would expect more frequent collisions°
and therefore more numerous radio galaxies.
The passing of the colliding-galaxy theory does not, how-
ever, weaken Ryle’s case if the data he presents are valid.
It may well be that exploding nuclei are a feature of a
galaxy’s youth, that a galaxy is more likely to experience
 The Edge of the Universe 295
the catastrophe in its first eon of existence than in its
second, more likely to do so in its second than in its third,
and so on. In that case, one would expect the explosions
to be more numerous in the youth of the Universe, and
therefore in the far reaches made out by our radio tele-
scopes, than in the present Universe of our own neighbor- —
hood.
In fact, one need not account for the difference at all.
The mere existence of the difference is sufficient. If there
is any overall difference between the Universe here and
-the Universe near the edge, then the steady-state theory is
eliminated, since it is the essence of the steady-state that
there be no significant difference through space or time.
Of course, one can question Ryle’s data. They rest on
the detection and measurement of very faint radio sources
and these can only be, at best, of limited accuracy and
reliability. Hoyle, for instance, clung stubbornly to the
steady-state theory, despite Ryle’s. findings, maintaining
that the microwave data were not yet sufficiently sturdy to
support a final decision.
But, then, quite unexpectedly, came a phenomenon that
extended the astronomer’s reach even more startlingly
than did Ryle’s distant radio sources and, moreover,
moved him back into the optical portion of the spectrum,
where, all things being equal, he could see more sharply.

Quasars

As explained earlier, many radio sources had been

shown, as far back as 1948, to be point sources, originat-
ing from quite restricted areas of the sky. The average
diameter of such radio sources is about 30 seconds of arc.
This means that all the “radio stars” could be fitted
_comfortably into an area not much more than half the
size of the full Moon.
And yet there are a few radio sources that are unusual-
ly small even by those standards. As methods of pinpoint-
ing the sources grew more refined, it seemed that a few
were only 1 second of arc or less in diameter.
This is unusually compact, and it could not help but
Touse the suspicion that if ordinary “radio stars” were
actually radio galaxies, then very compact “radio stars”
might really and literally be radio stars. Certainly, that
would account for their compactness.
Among these compact radio sources were several
known as 3C48, 3C147, 3C196, 3C273, and 3C286. The
 296 es The Universe
“3C” is short for “Third Cambridge Catalog of Radio
Stars,” a listing compiled by Ryle and his group, while the
remaining numbers represent the placing of the source on
that list. :
- In 1960, the areas containing these compact radio sources
- were combed by Sandage with the 200-inch telescope,
and in each case a star seemed to be the source. In the
case of 3C273, the brightest of the objects, the precise
position was obtained by Cyril Hazard in-Australia who
recorded the moment of blankout as the Moon passed
_ before it. The stars involved had been recorded on previ-
ous photographic sweeps of the sky and had always been
taken to be nothing more than faint members of our own
Galaxy. Painstaking photographing, spurred by their un-
usual microwave emission, now showed, however, that
- that was not all there was to it. Faint nebulosities proved
to be associated with some of the objects, and 3C273
showed signs of a tiny jet of matter emerging from it.
The compact radio sources, although they looked like
stars, might not be ordinary stars at all. They eventually
came to be called “quasi-stellar sources,” where “quasi-
stellar” means “star-resembling.” Asthe term became
more and more important to astronomers, quasi-stellar
radio sources became too inconvenient a mouthful, and it
was shortened to “quasar” (“quasi-stellar’), an ugly word
that is now firmly embedded in astronomic terminology.
Clearly, the quasars were interesting enough to warrant
investigation with the full battery of astronomic techniques
and that meant spectroscopy. Jesse L. Greenstein and
Maarten Schmidt labored to obtain the spectra and when
they accomplished that task they found themselves with
Strange lines they could not identify. Furthermore, the
lines in the spectra of one quasar did not match those in
any other. This was puzzling, but the quasars were still
accepted as objects of our own Galaxy. .
In 1963, however, Schmidt returned to the spectrum of
3C273. Six lines were present, of which four were spaced
in such a way as to seem to resemble a series of hydrogen
lines—except that no such series ought to exist in the
place in which they were found. What, though, if those
lines were located elsewhere but were found where they
were because they had been displaced toward the red end
of the spectrum? If so, it was a large displacement, one
that indicated a recession at the velocity of over 25,000
miles per second. This seemed unbelievable and yet if such
a displacement existed, the other two lines could also be
x
Peewee mmr sf Ss Pasy= eae Sena ortega
The Edge of the Universe 297
identified; one represented oxygen minus two electrons,
_ the other magnesium minus two electrons.
Schmidt and Greenstein turned to the other quasar
spectra and found that the lines there could also be
identified, provided huge red shifts were assumed.
Such enormous red shifts could be brought about by the
general expansion of the Universe; but if the red shift was
equated with distance in accordance with Hubble’s law
(see page 198), it turned out that the quasars could not
be ordinary stars of our own Galaxy at all. They had to
be among the most distant objects known—billions of
light-years away.
This was a hard thing to accept, and alternate explana-
tions for the red shift were sought. Could the quasars be
very massive objects with gravitational fields so intense as
to produce an enormous Einstein red shift (see page
171)? If that were so, they might be close by. It was
clearly shown, however, from theoretical considerations,
that this could not be so. The red shift had to signify a
velocity.-
Well, then, they might still be relatively close by and
have attained huge velocities not through the expansion of
the Universe but through some explosion at the center of
_ our Galaxy (a “small bang” so to speak). If so, all the
quasars would be streaking away from the center of the
Galaxy. To begin with, some of them might have been-
approaching us, and these would then show an enormous ~
violet shift. However, if the explosion had taken place so
long ago that all those emerging in our direction had
passed beyond us, no violet shifts would be involved. Even
so, some quasars would be moving more or less across our
line of sight and if they were members of our Galaxy,
even distant members, they would (considering their enor-
mous velocities) show a measurable proper motion. None
did. k
By the mid-1960’s, in fact, some forty quasars were
definitely identified, and the spectra of more than half of
these had been studied. Every single one of them showed
a large red shift, larger ones indeed than that of 3C273.
None showed a perceptible proper motion.
Furthermore, at least one direct piece of evidence,
independent of the red shift, has been advanced in favor
. of the great distance of the quasars. The microwave
emission of 3C273 seems rather deficient in. the 21-centi-
meter region; there is an absorption line there. The ab-
sorption line itself shows a red shift that indicates that the
 298 The Universe
hydrogen cloud which may be expected to absorb in that
region must lie at a distance of forty million light-years.
Between ourselves and 3C273, and at a distance of forty
mnillion light-years, there happens to be a cluster of galax-
ies in the constellation Virgo. It seems quite reasonable to
suppose that there is a cloud of hydrogen gas associated
with this cluster and that it is responsible for the 21-centi-
meter absorption.
In that case, 3C273, being farther than the Virgo
cluster, could certainly not be a member of our Galaxy.
And if it is beyond the Virgo cluster, then it seems
impossible not to accept it as being as far away as its red
shift indicates. Then, if one quasar is very far away, and if
its red shift represents this vast distance, it becomes diffi-
cult to argue that the even more extreme red shifts of the
other quasars represent some other factor than distance.
Most of the more recently discovered quasars are more
distant than those of the first batch. In May 1965,
Schmidt found the quasar 3C9 to have a red shift so large
as to indicate a recession at 80 percent the speed of light,
and later that year one was found to be receding -at 81
percent the speed of light. (This represents a velocity of
some 150,000 miles per second.) Such objects must be
nearly nine billion light-years away, according to Sandage,
and the light that reaches us from them must therefore
have been emitted nearly 9 eons ago. We may be seeing —
most of the way back toward the big bang, if there was
one.
And yet the acceptance of the huge distances of the
quasars involves astronomers in some puzzling and diffi-
cult points. If the quasars are indeed as incredibly distant
as they seem to be from their red shifts, then they must be
extroardinarily luminous to appear as bright as they do.
They seem not only to be emitting a huge quantity of
microwaves, but a huge quantity of visible light also. The
quasars are anywhere from thirty to a hundred times as
luminous as an entire ordinary galaxy!
Yet if this is so, and if the quasars had the form and
appearance of a galaxy, they ought to contain up to a
hundred times as many stars as an ordinary galaxy and be
up to five or six times as large in each dimension. Even at
their enormous distances they ought to show up as distinct
oval blotches of light in the 200-incher. The fact that they
appeared to be starlike points seemed to indicate. that,
despite their unusual brightness, they had to be far smaller
in size than ordinary galaxies.
 ee ee et: ARERR (Soe
. é a ee ore

The Edge of the Universe 299

The smaliness in size was accentuated by another
phenomenon, for the quasars were found to be variable in
the energy they emitted, both in the visible light region
and in the microwave region. Increases and decreases of
close to 50 percent were recorded over the space of a few
years.
For radiation to vary so markedly in so short a time, a
body must be small. Small variations might result from
-brightenings and dimmings in restricted regions of a body,
but large variations must involve the body as a whole.
If the body is involved as a whole, some effect must make
itself felt across the full width of the body within the
time of variation. But no effect can travel faster than
light, so that if a quasar varies markedly over a period
of a few years, it cannot be more than a light-year Or so
in diameter.
This cOmbination of tiny volume and great luminosity
poses such problems to the astronomer that there is a
-constant tendency to find some way of accepting the
quasars as nearby bodies after all. If the quasars were
nearby, they would not have to be exceedingly luminous
-to be as bright as they appear; they would be no more
luminous than one would expect a body a light-year across
to be. But if we assume, as it seems at the moment we
must, that the quasars are really vastly distant, then we
are faced with a body that is only a light-year across and
yet is up to a hundred times as luminous as an ordinary
galaxy that isa hundred thousand light-years across. How
can that be explained? A number of “far-out” explana-
tions have been suggested, including, for instance, the
formation, in a quasar, of colossal concentrations of -neu- —
trinos. A quasar might, by that view, be a “neutrino star.”
One interesting possibility is that Hoyle’s concept of a
massive implosion may fit here as well as in the case of
intense radio sources such as Cygnus A. What if an
implosion involved virtually all of a galaxy, and if the
energies released by the implosion blew away what matter
on the outskirts was not involved? Would not a quasar-
then be the bare naked center of an imploding galaxy?
Might it form tremendous quantities of neutrinos in the
course of the implosion?
It is tempting to draw an analogy between catastrophic
events among stars and those among galaxies. On a stellar
-scale, ordinary novae radiate much energy but retain their
substance and general starlike form; while supernovae
blow away much or most of their substance and implode,
 300 The Universe
contracting catastrophically into a white dwarf. Could it
be that on a galactic scale, ordinary exploding galaxies are
“novae” that radiate much energy but retain their sub-
stance and general galaxylike form, while some particular-
ly enormous catastrophes are like “supernovae” so that
much of the galactic substance is blown away while what
remains contracts catastrophically into a quasar? (And,
indeed, the light received from quasars is similar in some
respects to that received from white dwarfs!)
But, if this is the proper interpretation of a quasar, ate
can only be a short-lived object. It cannot radiate such
immense quantities of radiation for very long. Some calcu-
lations indicate that it can exist asa quasar for only a
million years or so. In that case, the quasars we see only
became quasars a short time ago, cosmically speaking, and
there must be a number of objects that were once quasars
but are quasars no longer.
Sandage, in 1965, announced the discovery of objects
that may indeed be ‘aged quasars. They seemed like ordi-
nary bluish stars, but they possessed huge red shifts as
quasars did. They were as distant, as luminous, as small as
quasars, but they lacked the microwave emission. He
called them “blue stellar objects” which can be abbrevi-_
ated as BSO’s.
The BSO’s seem to be more numerous than quasars,
perhaps fifty times as numerous, and one estimate places -
the total number of BSO’s within reach of our telescopes
at 400,000, though others suggest a considerably larger
figure. If they develop out of quasars, then they are fifty
_ times as numerous because they endure in the BSO form
fifty times as long—say 50,000,000 years. Still older’ qua-
Sars must dim to the point where they can be detected
neither by microwave emission nor by light emission.
What shape they may then take or how they may then be
recognized is as yet unknown.
The mere existence of quasars and BSO’s is a heavy and
perhaps an even fatal blow at the steady-state theory. All
of them are very far away and were therefore formed
many eons ago. Since none can be detected in our own
neighborhood, it seems that whatever processes formed
them. are not operative now (although there may be
long-dead quasars in our neighborhood that we have not
yet learned to recognize). This, in turn, means that the
Universe was different in important ways, eons ago.
If the big bang took place some 15 eons ago, this makes
sense. The Universe was smaller, hotter, younger, and
 The Edge of the Universe 301
- fuller eons ago than it is today, and it is not at all
surprising that catastrophic events such as quasar-forma-
tion should be common then while in today’s larger, cooler,
older, emptier Universe they do not occur. :
Indeed, by studying the distribution of the quasars and
BSO’s, Sandage believes not only that the big bang is
established, but that it is even. possible to decide between
the hyperbolic and pulsating versions of the big-bang theo-
ry. The Universe, Sandage says is pulsating with a cycle of
something like 82 eons.
After the big bang, in other words, the Universe ex-
pands at a slowly decreasing rate until it comes to a
momentary halt after 41,000,000,000 years. It then begins
to fall together again, contracting for another 41,000,000,-
000 years till it is a cosmic egg again, and there is a
‘second big bang. We are living perhaps a quarter of the
way along one of the expansion cycles and in the very
distant quasars we are seeing almost back to the most
recent big bang.
In fact, it was reported from Bell Telephone Laborato-
ries in New Jersey in 1965 that after all microwave
emission was accounted for, there was still a faint general
background of radiation. According to some theories, the
fiash of radiation that accompanied the big bang ought by
now, many eons later to make itself perceived as a gener-
alized microwave emission just like that detected at Bell
- Telephone. If so, it may be that man actually hears the
faint remnant of the big bang itself.
If this interpretation is correct, the present temperature
of residual matter from the explosion is 3°K., or three
degrees above absolute zero—judging from the nature of
the microwave emission. A temperature of 10°K. had
been predicted. The discrepancy seems to indicate that the
original temperature of the big bang was less than had
seemed likely. This could mean that the Universe at birth
_ Was not pure hydrogen but contained a considerable ad-
mixture of helium.
The steady-state theory can only be saved now if it
turns out—against all probabilities—that quasars and
BSO’s are nearby objects after all. In that case, they are
not very luminous and may be spread evenly through the
Universe with our instruments only capable of detecting
the nearest, Their existence would not, in that case, indi-
cate important differences between the far-distant reaches
of the Universe and our own neighborhood.
Furthermore, it is the quasars and BSO’s that account
 302 The Universe
for the abnormally high numbers of far distant radio
sources observed by Ryle. If these are considered to be
- not far distant at all, then the radio sources would in-
crease only by the expected amount with distance, as
called for by steady-state theory. And, to be sure, some
Teports in the spring of 1966 offer the chance that quasars
are associated with “peculiar galaxies’—-which are distant
but not at “cosmological distances”; that is, distances that
would account for the red shift. Such relative nearness for
some quasars would weaken the argument for the big bang.
But these indications of the nearness of quasars do not
seem to impress most of the men in the field and the
salvation of the steady-state theory seems now a rather
forlorn hope. Even Hoyle, in 1965, finally gave up and
accepts the big bang. He speculates, however, that the big
bang may still be a local phenomenon; that is that we are
living in a big-bang “bubble” within a much larger Uni-
verse in which steady-state may hold sway, generally.
There seems at present, however, no way of checking on
such a far-ranging supposition.
* * *£+ *& *€

And so, as I promised at the start, we have followed the |

quest of mankind out toward the endlessly receding hori-
zon. We began with man’s narrow vision of a patch of flat |
Earth and have paused now at the point where man
pictures a Universe 26,000,000,000 light-years in diameter
pulsating in a vast period of 82,000,000,000 years. .
Nor need we feel that we have now plumbed the
Universe to the utmost. Astronomy has been advancing at
an ever-accelerating pace for four centuries now, and
there are no signs, as yet, of any leveling off. More has
been learned about the Universe in the last quarter-
century than in all man’s history before; what, then, may
lie ahead in the next quarter-century?
If it is exciting to probe the unknown and shed light on
-what was dark before, then more and more excitement
surely lies ahead of us.
 Suggested Further Reading

Bonnor, William, The Mystery of the Expanding Universe, The

Macmillan Co., New York (1964).
The Flammarion Book of Astronomy, Simon & Schuster, Inc.,
New York (1964).
Glasstone, Samuel, Sourcebook on the Space Sciences, Van
Nostrand Co., Inc., Princeton (1965).
Hoyle, Fred, Astronomy, Doubleday & Co., Inc., New York
(1962).—-Galaxies, Nuclei and Quasars, Harper & Row,
Publishers, New York (1965).
McLoughlin,. Dean B., Introduction to Astronomy, Houghton
Mifflin Co., Boston. (1961).
Pannekoek, A., A History of Astronomy, Interscience Pub-
lishers, New York, (1961).
Rudaux, Lucien, and Vaucouleurs, Larousse Encyclopedia of
Astronomy, Prometheus Press (1959).
Smith, F. Graham, Radio Astronomy, Penguin Books, Inc.,
Baltimore (1960).
- Struve, Otto, and Zebergs, Velta, Astronomy of the 20th Cen-
tury, The Macmillan Co., New York (1962).

303
 INDEX

Aberration, light, 48-49 Antigalaxies, 266

Absolute magnitude, 53-54 Antimatter, 263
Absorption spectrum, 129 big bang theory and, 266-267
Adams, Walter Sydney, 167, 171 exploding galaxies and, 292
Aldebaran, 42, 54 formation of, 264-265
Algol, 68 steady state theory and, 264-
- Alpha Centauri, 52-53, 54 265
Alpha Crucis, 133 Antineutrinos, 265
Alpha rays, 254 Antineutron, 262
Alpher, Ralph Asher, 224 Antinuclei, 262-263
Altair, 53-54 Antiparticles, 262
Anaximander, 17 Antiproton, 263
Anderson, Carl David, 262 Antiuniverse, 267
Andromeda galaxy, 106, 174, 237 Apex, 64
age of, 188-189 Aphelion, 32
class of, 187 Apogee, 24
distance of, 105, 213-214, 216 Arcturus, 43, 54
globular clusters of, 214 Argelander, Friedrich W. A., 61
_ movae in, 214 Aristarchus, 24, 26
_tadial velocity of, 195 Aristotle, 18, 66, 98
satellites of, 204 Asteroids, 31
- size of, 214 Astronomic unit, 33
stars of, 189 Atom(s), 120 i
Andromeda nebula, 95-98 temperature and, 142
cepheids in, 105 Atomic nucleus, 120
color of, 98 A: U., 33
distance of, 96-97, 104-105
novae in, 103-104
stars in, 105 Baade, Walter, 174, 189, 214,
Angstrom, Anders Jonas, 130 221, 237, 284, 288
Angular momentum, 113 Barnard, Edward Emerson, 45,
Antares, 149, 163 85, 173
color of, 78, 141, 145 Barnard’s star, 150
luminosity of, 54 color of, 145
Antiatom, 263 distance of, 53
Anticosmon, 267 planet of, 243
Antideuteron, 263 proper motion of, 45
Antielectron, 262 transverse velocity of, 83
305
 ie an

306 The Universe

Barred spiral galaxies, 187 stellar populations and, 214-
Becquerel, Antoine Henri, 120, 216
254 : Ceres, 31
Beryllium, 225-226 Cesium, 129
Bessel, Friedrich Wilhelm, 51, Chamberlin, Titomas Chrowder,
65 125
Beta Centauri, 54 Chandrasekhar, Subrahmanyan,
Beta Persei, 67 178
' Beta rays, 254 Chandrasekhar’s limit, 178
Betelgeuse, 149, 163 Chiu, Hong-Yee, 180
color of, 145 Christofilos, Nicholas, 249
diameter of, 146 Clark, Alvan Graham, 51, 166
Bethe, Hans Albrecht, 135, 224 Clusters, stellar, 64-66
Big bang theory, 222, 301 Coal Sack, 86.
antimatter and, 266-267 Color, 78-80
Binary stars, 50-51, 150n, 157- Comets, 35-37,
: 158 ' parallax of, Line
eclipsing, 67 Compton, Arthur Holly, 241,
spectroscopic, 158 246
Black dwarfs, 150, 152 Comte, Auguste, 128
Bloch, Felix, 279 Continuous creation theory, 234
Blue stellar objects, 300 Copernicus, Nicolas, 26, 40, 84
Bolston, John G., 283-284 Corona, 144, 257 ;
Boltwood, Bertram Borden, 122 Coronium, 144, 257 =*
Bondi, Hermann, 234-235 Cosmic egg, 221, 233
Boron, 225 composition of, 222-224
Bowen, Ira Sprague, 144 origin of, 227-229 -
ees moe 48 Cosmic rays, 262, 264
BSO’s,3 Crab Nebula and, 2590
Bunsen, Robert Wilhelm, 128, discovery of, 245
142 high-energy, 251
Burbridge, Geoffrey R., 291 nature of, 245-247
solar, 250
sources of, 248-251
Calcium, 130, 159 Cosmological principle, 232
Californium, 182 perfect, 234
Canopus, 42, 54 Cosmology, 201
Capella, 54, 141, 147 Cosmon, 267
Carbon, 163 Cosmos, 221, 245
Carrington, Richard Christo- Cosmotron, 247
pher, 248 Cowan, Clyde Lorrain, Jr., 243
Cassini, Giovanni Domenico, 29 Crab Nebula, 176
Cassiopeia A, 277, 281 central star of, 181
radio emission by, 283, 286 cosmic rays from, 259n
Cataclysmic variables, 101-102 light polarization of, 276
Celestial sphere, 22 microwave emission by, 276-
Cepheids, 68 x 278
distance of; 82-84 Moon occultation of, 260
dust-dimming of, 88 radio emission of, 283.
luminosity of, 69-71 X-rays from, 257-259
Magellanic clouds and, 69-70 Curtis, Heber Doust, 103-104
mass-luminosity relation and, Cygnus A, 260, 289, 291-293
152n distance of, 293
pulsation of, 135 microwave energy of, 286
revision of scale of, 213-218 radio emission by, 284
 Index ~ 307
_ Dark nebulae, 85 Electrons, 120
Darwin, Charles Robert, 120 energy emission by, 276
Degenerate matter, 170 ; spectral lines and, 142-143
Delta Cephei, 68 temperature and, 142-143
Delta Orionis, 158 Elements, 128-130
Demon star, 67 formation .of 224-227
Deneb, 54 - Ellipse, 26-27
Deuterium, 225, 263, 281 Ellipsoidal galaxies, 187
Deuteron, 263 Emission spectrum, 97
Dirac, Paul Adrien Maurice, Encke, Johann Franz, 31
262 Energy, conservation of, 115,
Dombrovsky, V. A., 276 - 234
Doppler, Christian Johann, 73 mass and, 120-121
Doppler, effect, 74 nuclear, 120
light and, 75-76 Epsilon Aurigae, 146, 149-150,
Doppler-Fizeau effect, 80 168
Dust, interstellar, 78 Eratosethenes, 19
Eros, 31
Escape velocity, 230
Evolution, stellar, 149-150
Earth, 13-21, 27 Exploding galaxies, 289-291
age of, 119, 123 - _ Explorer XI, 262
central position of, 25
circumference of, 20
composition of, 139 Fabricius, David, 68
core density of, 168 . Fermi, Enrico, 251
crust density of, 168 First law of thermodynamics,
curvature of, 16-17 185
cylindrical, 17 Fixed stars, 22
diameter of, 20 Fizeau, Armand H. L., 80
flat, 14-16 Flare stars, 250
magnetic properties of, 245, microwave emission by, 278
249. Fraunhofer, Joseph von, 79, 129
orbit of, 39-40 - Fraunhofer lines, 80
orbital velocity of, 46 Friedman, Herbert, 258, 260
second-generation nature of,
182
size of, 19-21 Galactic nuclei, 107
spherical, 18, 198 stars of, 189-190
surface area of, 20 Galaxies, antimatter and, 265
Van Allen belts of, 249 classes of, 187-189
Eclipsing binary, 67 clusters of, 204-206, 286, 296-
Eddington, Arthur Stanley, 134, 298
155, 168-169, 222 colliding, 284
Edlen, Bengt, 144 continuous formation of, 234-
Einstein, Albert, 121, 171, 198- 235
203, 207, 229, 232, 240 distances of, 198, 210-213
Einstein shift, 171 electric charge on, 235
Electra, 42 expanding Universe and, 203
Electric charge, conservation of, exploding, 289-291
263-264 gravitation and, 204
Electromagnetic field, 253 imploding, 292, 299
Electromagnetic radiation, 254 microwave energies of, 283-
Electromagnetic spectrum, 254- 291
256 . number of, 106, 207-209
fae:ote
308 The Universe
peculiar, 285, 288 Hartmann, Johannes Franz, 158-
radio, 284-291 159 :
recession of, 195-198, 219, Hazard, Cyril, 296
— 235-236 Hecataeus, 16
red shift of, 195-198 Heliccentric system, 26
spectral class of, 189-190 Helium, 130-131
structure of, 106-108 cosmic rays and, 247 —
X-ray, 260 fusion of, 163
_ Galaxy a Way), 58 interstellar gas and, i59
age of, planets and, 139 ;
center a on 84, 88-90 stellar formation of, 179
class of, 187 Universe and, 301
evolution of, 188-189 Helium-3, 225
explosion in, 291 Helium-4, 131, 225-226
insterstellar gas of, 160, 278- Helium-5, 226
282 Helmholtz, Hermann L, F.
magnetic field of, 250n, 251 115, 117, A190 4222 135, 39,
mass of, 149, 153,219,248, 293
microwave emission by, 275 Henderson, Thomas, 51
number of stars in, 92 Herschel, John, 66, 69, 106
rotation of, 91, 195 Herschel, William, 34, 49-50, 58-
satellites of, 92, 204 61, 65, 76, 84-85, 106
shape of, 59-60 Hertz, Heinrich Rudolf, 254
size of, 60, 61, 83, 87-89, 213 Hertzian waves, 254 5
spiral arms of, 2250n, 279 Hertzsprung, Ejnar, 83, 145
Galileo Galilei, 29, 37, 42, 49, Hertzprung-Russell diagram,
55, 58, 64, 101 148-150
Gamma ray(s), 254 Hess, Victor Franz, 245
antimatter and, 263 Hind, John Russell, 102
energy of, 255 Hipparchus, 24-25, 41-42, 100
Gamma ray stars, 161-162 Horsehead Nebula, 85
Gamow, George, 222, 223-226, Hoyle, Fred, 137, 183, 226, 233, .
233, 236 236, 295, 302 -
Beat Sergi Ilarionovich, H-R diagram, 148-150
star clusters and, 161
- Gas, interstellar, 156-160 stellar evolution and, 152-155
Globular clusters, 65-66 Hubble, Edwin Powell, 104-105,
- Andromeda galaxy and, 214 187-188, 196-198, "294
distribution of, 66, 71-72 Hubble’s constant, 210-212, 217,
relative distance of, 71 221, 238
Gold, Thomas, 233 Hubble’s Law, 198, 203, 206,
Goldhaber, Maurice, 267 210-213
Goodricke, John, 67 Huggins, William, 81
Gravitation, 34, 110-111, 150n
' Humason, Milton Lasell, 196
antimatter and, 265
galaxies and, 203-204 Hutton, James, 119
Graviton, 241 Huygens, Christian, 84
Great Hercules Cluster, 65 Hydrogen, 130
Greenstein, Jesse L., 296 continuous formation of, 235
cosmic egg and, 222-223
fusion of, 131
interstellar gas and, 159, 278-
Hades, 38 SMB ae
Halley, Edmund, 35, 42, 65 microwave emission by, 278-
Halley’s Comet, 36 279
 Index 309
planets and, 140 Kant, Immanuel, 96, 103
spectral lines of, 143 Kapteyn, Jacobus Cornelius, 61,
Hydrogen-1, 131 84, 90
Hydrogen-2, 225, 263, 281 Kapteyn’s star, 53
Hydroxyl group, 159 Kepler, Johannes, 26, 39, 40,
in interstellar gas, 281-282 101
Hypersphere, 200 Kepler’s supernova, 176
Kirchhoff, Gustav Robert, 128-
129, 142
Kruger 60B, 169
1C443, 277
Kuiper, Gerard. Peter, 37
- Imploding galaxies, 292
Infrared giants, 146-147
Infrared radiation, 76
energy of, 253, 255 Lagrange, Joseph Louis, 112
- Interferometer, 145 Laplace, Pierre Simon de, 96,
Interstellar gas, 156-161, 278- 103, 112, 149, 176
282 Large Magellanic Cloud, 69
hydroxyl in, 281-282 distance of, 89
oxygen in, 281 interstellar gas in, 160
supernovae and, 181 size of, 92
Tntrinsic variable, 68 stars in, 92
Ton(s), 143, 244 structure of, 108
atmosphere and, 249 Lassell, William, 37
Tonization, 143 Lead, 122
Ionosphere, 249 Leavitt, Henrietta Swan, 68,
Iridium, 168 215
Tron, 144 Lehrer, Tom, 200n
cosmic rays and, 247 Lemaitre, Georges Edward, 221-
spectral lines of, 142-143 222, 224
stellar formation of, 164, 179 Leverrier, Urbain, J. J., 34
Island universe, 97 Light, aberration of, 48-49
chemical composition and, 79
colors of, 74
Doppler effect and, 75-76—
Jansky, Karl, 269, 275 energy of, 253, 255
Jeans, James Hopwood, 126 gravity and, 201
Jeffreys, Harold, 126 invisible, 77
Jupiter, 23 nature of, 254
angular diameter of, 45 particle properties of, 240-241
angular momentum of, 125 path of, 200
brightness of, 42 photons of, 253
composition of, 140-141 scattering of, 78, 87-88
diameter of, 37 temperature and, 78-79, 133
distance from Sun of, 33 “tired,” 197
hydrogen and, 140 velocity of, 33, 121
magnetic field of, 271, 277 visible, 77, 253
microwave emission by, 270- wave length of, 253
271, 277 wave nature of, 74-75
microwave reflection from, Light-hour, 33
272 Light-year, 47
orbital motion of, 89 Lithium, 225-226
planetesimal hypothesis and, Lobachevski, Nikolai Ivanovich,
126 200
Satellites of, 38 Local Group, 205-206
temperature of, 153 Lockyer, Joseph Norman, 130
 Soe

310 The Universe

Lodge, Oliver Joseph, 268 Meteorites, 99, 264
Long-period variable, 68 tmicrowave fefiection from,
Lovell, Alfred C. B., 278 272 :
Luminosity, stellar, 53-54 solar energy and, il
Luyten, Willem Jacob, 171 Meteoroid, 99
Lyell, Charles, 119 Michelson, Albert Abraham, 145
- Lynds, Clarence Roger, 290 Microwaves, 32 ,
Lyttleton, Raymond Arthur, 235 atmosphere and, 268
fal
ols
Maa
at
aa
energy of, 255
from Crab Nebula, 276-277
from flare stars,278 — Nie
ic
Mi, 175 from Galactic center, 275-276
M3, 66 from galaxies, 289-291
Mi3, 65 from interstellar hydrogen,
M22, 66 278-279
M31, 95, 204 from Jupiter, 270-271, 277
M32, 205 from Milky Way, 275, 278
M33, 205 from Sun, 270
M51, 106 from supernovae, 277
M8i, 290 from Venus, 271
M82, 290 synchrotron radiation and, 276
M87, 260, 288 Milky Way, 58
Magellan, Ferdinand, 18, 21, 69 dark nebulae in, 86
Magellanic Clouds, 68-69, 192, microwave emission by, 275,
- 204 278
Magnetic storms, 248-249 Millikan, Robert Andrews, 245 tte
it
iiell
Magnetosphere, 249 Milne, Edward Arthur, 231-232
Magnitude, stellar, 41-42 Minkowski, Rudolf Yeo B., 174,
Main sequence, 149 285
stellar stay on, 154-155 Mira, 68
Mariner II, 271 Mizar, 158
Marius, Simon, 95 Moon, 23
Mars, 23 angular diameter of, 45
brightness of, 42 brightness of, 42
distance from Sun of, 33 circumference of, 24
Microwave fefiection from, diameter of, 24 -
272 distance of, 23-25
orbit of, 26 eclipse of, 17, 24
parallax of, 29 microwave feficction from,
Mass, energy and, 121-122 272
mage inosity relation, 15i- parallax of, 18, 29.
satellite nature of, 26
Maury, Antonia C., 158 surface of, 272
Maxwell, James Clerk, 253 3C273 and, 296
Menzel, Donald Howard, 130 _ X-tay stars and, 260
Mercury, 23 Morgan, William Wilson, 279
distance from Sun of, 33 Moulton, Forest Ray, 125
microwave refiection from, Mysterium, 282
272
orbital motion of, 89 Nebulae, 85
rotation of, 274 dark, 85
Messier, Charles, 65, 84, 95, 106, extra-galactic, 105
175, 288 planetary, 176
Meteor, 99 spectra of, 97
Meteorology, 99n spiral, 106

ol:
Be
 Index : 311
ae hypothesis, 96, 114, 118, Osmium, 168 +
Oxygen, 143
Nebulium, 144 interstellar gas and, 159-160,
Neptune, 34, 37 * 281
Nereid, 38 Sun and, 139
Neutrino(s), 180, 242, 265
detection of, 243-244
quasars and, 299 Parallax, 27-29
supernovae and, 181 stellar, 39-40, 46
Neutrino astronomy, 244 Parsec, 52
Neutrino stars, 299 Parsons, William, 106
- Neutrons(s), 120, 235 Particle(s), subatomic, 241-242
big bang and, 223-226 Particle accelerators, 247
Steady state theory and, 261 Pauli, Wolfgang, 243
Neutronium, 224, 259 Peculiar galaxies, 285, 288
Neutron stars, 259-261 Perfect cosmological principle,
Newton, Isaac, 34, 75, 110, 150n 234
NGC 891, 107 Perigee, 24
NGC 1068, 288 Perihelion, 32
NGC 4486, 288 Perseus Arm, 280
NGC 5128, 285, 289 Perturbations, 111
NGC 6611, 156 Philolaus, 18
NGC 7293, 176, 181 Phoebe, 38
NGC 7619, 196 Photons, 241-242
Nicholas of Cusa, 44 antimatter and, 265
Piazzi, Giuseppe, 31
Pickering, Edward Charles, 142
North star, 42 Planck, Max K. E. L., 240
Nova(e), 101 Planet(s), 23
Andromeda galaxy and, 214 composition of, 140-141
luminosity of, 172 distance from Sun of, 33
_ number of, 102 elliptical orbits of, 26-27
repeating, 172 hydrogens and, 140
white dwarfs and, 178-181 microwave fefiection from,
Nova Aquilae, 102, 173 272 7)
Nova Persei, 103 Telative distance of, 24
Nuclear energy, 120 Planetary nebulae, 176
Nuclear reactions, 120 Planetesimal hypothesis, 126, 136
Nucleus, atomic, 120 Platinum, 168-169
Nutation, 49 Pleiades, 42, 64-65, 161
proper motion of, 65
size of, 65
Oberon, 38 Pliny, 100
Olbers, Heinrich W. M., 56-57, Pluto, 53
60 distance from Sun of, 34
Olbers’ paradox, 56-58, 95, 108 Population, Cepheids, 214-216
expanding Universe and, 208- Population, stellar, 190-194
209 Polaris, 41
Omicron Ceti, 68 Positron, 262
Oort, Jan Hendrik, 37, 89, 176, Primary radiation, 247
280 Procyon, 42
Open cluster, 65 brightness of, 42, 167
Orion Arm, 280 companion of, 166-167
Orion Nebula, 84, 156 distance of, 53
spectrum of, 97 luminosity of, 54
 312 The Universe » n
Procyon A, 166 Roentgen, Wilhelm Konrad, 254
Procyon B, 166-167 Rosette Nebula, 156
mass of, 171 Rosse, Lord, 175 :
Proper motion, 43 Rossi, Bruno, 246, 258
~ Proton(s), 120 | RR Lyrae, 215
big bang and, 223-226 Rubidium, 128
cosmic rays and, 247 Russell, Henry Norris, 130, 148
Ptolemaeus, Claudius, 25 Rutherford, Ernest, 120, 254
Ptolemaic system, 25 Ryle, Martin, 294, 302
Ptolemy, 25, 42, 84
Purcell, Edward Mills, 279
Sagittarius Arm, 280
’ Sandage, Allan Rex, 290, 296, —
Quasars, 296-300 300
appearance of, 298 S Andromedae, 103, 173-174
formation of, 299 Satellite, 26
nature of, 298- 300 Saturn, 23
- neutrinos and, 299 diameter of, 37
red shift of, 297-298 distance from Sun of, 33
spectra of, 296 planetesimal hypothesis and,©
supernovae and, 299 127
Quasi-stellar radio sources, 296 satellites of, 38
Schaeberle, John Martin, 167
Schiaparelli, Giovanni Virginio,
Radar, 269 274
Radial velocity, 81 Schmidt, Maarten, 296, 298
Radioactivity, 120 S Doradus, 92, 155
‘Radio astronomy, 269 ’ Secchi, Pietro Angelo, 142
Radio galaxies, 284-291 Secondary radiation, 247
Radio sky, 275 Second law of thermodynamics,
Radio stars, 283, 295 186
Radio telescope, 269 Shapley, Harlow, 71, 83-84, 89,
Radio waves, 254 104, 152, 215
astronomy and, 268 Shklovsky, losif Samuilovich,
energy of, 255. 276, 288
Ramsay, William, 131 Shooting stars, 99
Reber, Grote, 269, 275, 284 Sirius, 42
Red dwarfs, 145, 152, 155, 278 brightness of, 42, 46
Red giants, 145, 162-165 color of, 141
formation of, 162 companion of, 51, 166-167
H-R diagram and, 149 distance of, 53
Red shift, 76 luminosity. of, 54
distances and, 238-239 radial velocity of, 81
galactic, 195-198 surface temperature of, 133
gravitational, 171 Sirius A, 166-167, 178
Olbers’ paradox and, 209-210 Sirius B, 166, 243
quasars and, 296-297 brightness of, 166
Regulus, 41, 54 color of, 167
Reines, Frederick, 243 density of, 171, 177
Relativity, theory of, 198-203 diameter of, 167
Richer, Jean, 29 luminosity of, 166
Riemann, Georg F. B., 200 surface temperature of, 167
Ring Nebula, 176 Sitter, Willem de, 202, 222
Rigel, 54 61 Cygni, 52
Ritter, Johann Wilhelm, 77 planet of, 243
 Index
: 313
61 Cygni C, 243 explosions of, 172-174
Slipher, Vesto Melvin, 195-196 formation of, 156
Small Magellanic Cloud, 68 gamma ray, 262 :
distance of, 89 ‘ gaseous nature of, 155-156
stars in, 92 giant, 144-146
structure of, 108 helium-burning, 163
Sodium, 129-130 H-R diagram of, 149-150
Solar flare, 248-250, 257, 270 invisible, 55
Solar spectrum, 76-77, 129-130 lifetimes of, 151-155
lines in, 79-80 luminosity of, 53-54
Solar system, 26 magnetic fields of, 250
age of, 137 main sequence, 149, 153-155
angular momentum and, 113- mapping of, 42, 60, 100
114, 125 masses of, 150, 151
energy in, 115-118 motion of, 80-81
ee of, 96, 114-115, 126- neutrinos and, 244
ay neutron, 259-261
Kepler’s model of, 26-27 new, 100-102
origin of, 182-183 numbers of, 58-60
size of, 32-34, 272 parallax of, 39-41, 46
stability of, 111 photography of, 61
Solar wind, 249-250, 264 planets of, 243
Sombrero galaxy, 107 populations of, 190-194
Sound waves, 73-74 proper motion of, 43
Spectral classes, 141 radio, 283 —
surface temperature and, 144 second-generation, 182
Spectral lines, 79-80 size of, 145-147
electrons and, 142-143 spectra of, 80, 97
elements and, 128-130 technetium in, 227
ions and, 144 variable, 66-68
_. Spectroscopic binary, 158 visible, 55
Spectroscopy, 128 X-ray, 260
Sphere, celestial, 22 Steady state theory, 234, 261,
Spheroidal galaxies, 187 264, 294-295, 300-302 3
Spiral arms, 107, 279 Struve, Friedrich G. W. von, 51
stars of, 189-190 Struve, Otto, 159
Spiral galaxies, 106-107, 187 Subatomic particles, 169-170
Spitzer, Lyman, Jr., 136 Sun, 26
Stars, 22 - absolute magnitude of, 54
beyond the main sequence, age of, 124, 127, 132, 137
161-165 angular momentum of, 125
binary, 50-51, 150n, 157-158 balloon photography of, 256
brightness of, 41-42 brightness of, 42
clusters. of, 64-66 composition of, 130, 162
collapse of, 178-182 constitution of, 128-130
’ color of, 78, 141 contraction of, 118
cosmic rays and, 250 core composition of, 162.
diameter of, 146-147 core density of, 168
distance of, 41, 45 core radiation of, 181
distribution of, 58-60 corona of, 257
dwarf, 145 cosmic rays of, 248, 250
Earth’s curvature and, 16-17 degenerate matter in, 170
element formation in, 226-227 density of, 168
evolution of, 149-150, 151-156, diameter of, 37
160-165 distance of, 24, 30-32
 314 The Universe
evolution of, 151, 162 | Tarantula Nebula, 93
energy of, 115-118 Technetium, 227
flares on, 248 . Telescope, 29
galactic location of, 107 Thermodynamics, laws of, 185
galactic position of, 84, 280 3C9, 298
helium in, 130-131, 138-139 3C48, 295
helium formation in, 131-132 3C147, 295
hydrogen in, 130, 138-139 3C196, 295
- hydrogen fusion in, 131-132, 3C231, 290
135 3C273, 295-298
‘internal temperature of, 134 3C286, 295
iron in, 142 “Tired” light, 197
life on main sequence of, 155 Tombaugh, Clyde William, 34
magnetosphere and, 249 T Pyxidis, 172
mass of, 168 Transuranium elements, 182
mass-loss of, 121, 131-132- Transverse velocity, 81.
microwave reflection from, 272 Tridium, 225
microwaves from, 270 Triton, 37
motion of, 23n, 62-64 Trumpler, Robert Julius, 88, 162
neutrinos and, 180 Tycho Brahe, 26, 101
‘oxygen in, 139 Tycho’s supernova, 176
parallax of, 30n \
period of galactic revolution,
91 Ultraviolet radiation, 77
reddening of, 78 energy of, 253, 255
red giant formed from, 162 stars and, 258
Tevolution about galactic cen- Uniformitarian principle, 119
ter, 92 Universe, 200
rotation of, 125 age of, 184-187, 221, 231
second-generation origin of,' antimatter and, 264-267
182 birth of, 220-224
solar system and, 110 composition of, 159
spectral class of, 142 contracting, 228
spectrum of, 76-77, 257 edge of, 294-295 -
spots on, 248 end of, 220
surface temperature of, 133 evolution of, 219-239
temperature of, 257, 270 expanding, 202-203, 208, 234-
Venus’ transit across, 31 237
volume of, 168 exploding, 222-224
X-rays from, 257-258 “flat,” 200
Sunspots, 248 general shape of, 200-202 _
Supergalaxy, 207 homogeneity of, 198-201
Supernovae, 174, 259 hyperbolic, 229
brightness decline of, 182-183 limit of, 212-213
chains of, 291 number of galaxies in, 231-234
cosmic rays and, 251 observable, 210, 218
galactic, 174-176 oscillating, 231, 300-301
interstellar gas and, 181 pulsating, 231
microwave emission by, 277 relativistic, 232-233
Riemannian, 200
neutrinos and, 180
rim of, 232
numbers of, 181 scale of, 213
quasars and, 299 size of, 218
types of, 174 steady-state, 236
Synchrotron radiation, 276 time-changes of, 233
 ‘Index 315
Uranium, 120, 182 Water, 189 é
half-life of, 122 © Weizsicker, Carl Friedrich von,
; Uranus, 34, 37 36
Whipple, Fred Lawrence, 278
Whirlpool galaxy, 106, 188
Van Allen, James Alfred, 249 White dwarf, 167, 259, 300
Van Allen belts, 249 formation of, 177-181
Van de Hulst, Hendrik Chris- mass of, 178
toffel, 278, 280 novae and, 179-181
Van de Kamp, Peter, 243 numbers of, 171
Van Maanen’s star, 53 Wien, Wilhelm, 133
Van Maanen 2, 178 Witt, Karl Gustav, 31
Variable stars, 66-68 Wolf, Max F. J. C., 85
Vaucouleurs, Gerard Henri de, Wolf 359, 53
108, 207 Wollaston, William Hyde, 79
Vega, 42 WZ Sagittae, 179
color of, 78, 141
distance of, 51-52 X-ray(s), 254
~ luminosity of, 54 cosmic sources of, 258-261
Veil Nebula, 176 energy of, 255
Venus, 23 galaxies and, 260
brightness of, 42 neutron stars and, 260
distance from Sun of, 33 solar, 257-258
magnetic field of, 271-272 steady state theory and, 261
microwave(s) from, 271 X-ray galaxies, 260
microwave reflection from, 32, X-ray stars, 260
272-273
mountains on, 273
rotation of, 273-274 Yellow dwarfs, 148
temperature of, 271-272 Yellow giants, 148
transit of, 31 Ylem, 223, 259
water on, 256
Violet shift, 76 Zwicky, Fritz, 174, 181
 @2 DISCUS BOOKS
DISTINGUISHED NON-FICTION

THEATER, FILM, AND TELEVISION

ACTORS TALK ABOUT ACTING _
Lewis Funke and John Booth, Eds. ; 15062 1.95
| ACTION FOR CHILDREN’S TELEVISION 10090 1.25
ANTONIN ARTAUD Bettina L. Knapp 12062 1.65
A BOOK ON THE OPEN THEATER Robert Pasoli 12047 1.65
THE CONCISE ENCYCLOPEDIC GUIDE _TO SHAKESPEARE
Michael Martin and Richard Harrier, Eds. 16832. 2.65
THE DISNEY VERSION Richard Schickel 08953 1.25
EDWARD ALBEE: A PLAYWRIGHT IN PROTEST
Michael E. Rutenberg 11916 1.65
_ THE EMPTY SPACE Peter Brook. 32763 1.95
_ EXPERIMENTAL THEATER James Roose-Evans 11981 1.65
FOUR CENTURIES OF SHAKESPEARIAN CRITICISM ?
Frank Kermode, Ed. 20131 1.95
_ GUERILLA STREET THEATRE Henry Lesnick, Ed. 15198 2.45
THE HOLLYWOOD SCREENWRITERS
_ Richard Corliss 12450 1.95
IN SEARCH OF LIGHT: THE BROADCASTS OF ‘
EDWARD R. MURROW Edward Bliss, Ed. 19372 1.95
INTERVIEWS WITH FILM DIRECTORS :
Andrew Sarris 21568 1.95
_ MOVIES FOR KIDS Edith Zornow and Ruth Goldstein 17012 1.65
_ PICTURE Lillian Ross 5 08839 1.25
THE LIVING THEATRE Pierre Biner 17640 1.65
_ PUBLIC DOMAIN Richard Schechner , 12104 1.65
RADICAL THEATRE NOTEBOOK Arthur Sainer 22442 2.65

~ GENERAL NON-FICTION
ADDING A DIMENSION Isaac Asimov 22673 1.25
A TESTAMENT Frank Lloyd Wright 12039 1.65
AMBIGUOUS AFRICA Georges Balandier : 25288 2.25
‘THE AMERICAN CHALLENGE
J. J. Servan Schreiber 11965 1.65
AMERICA THE RAPED Gene Marine 09373 1.25
ARE YOU RUNNING WITH ME, JESUS?
_ Malcolm Boyd 09993 1.25
~ THE AWAKENING OF INTELLIGENCE J. Krishnamurti 27342 2.65
BLACK HISTORY: LOST, STOLEN, OR STRAYED
_ Otto Lindenmeyer 09167 1.25
THE BOOK OF IMAGINARY BEINGS
Jorge Luis Borges 11080 1.45
BUILDING THE EARTH Pierre Teilhard de Chardin 08938 1.25
CHEYENNE AUTUMN Mari Sandoz 09001 1.25
THE CHILD IN THE FAMILY Maria Montessori 28118 1.50
THE CHILDREN’S REPUBLIC Edward Mobius 21337 1.50
CLASSICS REVISITED Kenneth Rexroth 08920 1.25
(1) DDB 5-77
 @& DISCUS BOOKS
DISTINGUISHED NON-FICTION

THE CONCISE ENCYCLOPEDIC GUIDE TO

SHAKESPEARE Michael Rheta Martin and
Richard A. Harrier 16832
CONSCIOUSNESS AND REALITY
Charles Musees and Arthur M. Young, Eds. 18903
CONVERSATIONS WITH JORGE LUIS BORGES
Richard Burgin 11908
DIALOGUE WITH SAMMY
J. McDougall and S. Libovici 15644
DISINHERITED Dave Van Every 09555
DIVISION STREET: AMERICA Studs Terkel 22780
' DRUG AWARENESS
Richard E. Horman and Allan M. Fox (editors) 11064
ESCAPE FROM FREEDOM Erich Fromm 29215
THE FEMALE IMAGINATION Patricia Meyer Spacks 28142
FRONTIERS OF CONSCIOUSNESS
John White, ed. 24810
GERMANS George Bailey 20644
GERTRUDE STEIN: A COMPOSITE PORTRAIT
Linda Simon, Ed 20115
THE GREAT POLITICAL THEORIES, VOL. I
Michael Curtis 23119
THE GREAT POLITICAL THEORIES, VOL. Il
Michael Curtis 23127
THE GREEK WAY Edith Hamilton 30908
GROTOWSKI Raymond Temkine 12278
HARD TIMES Siuds Terkel 22798
HOMOSEXUAL: LIBERATION AND OPPRESSION
Dennis Altman 14214
HOW OLD WILL YOU BE IN 1984?
ed. by Diane Divoky 09142
THE HUMAN USE OF HUMAN BEINGS
Norbert Wiener 21584
THE INCAS Garcilaso de la Vega 11999
INTERPRETATION OF DREAMS Freud 12393
JESUS IN BAD COMPANY Adolf Hoil 19281
JUSTICE Richard Harris 12120
THE LIFE AND DEATH OF LENIN’ Robert Payne 12161
THE LIFE AND WORK OF WILHELM REICH
M. Cattier 14928
LIFE IN ACRYSTAL PALACE Alan Harrington 15784
THE LIFE OF JOHN MAYNARD KEYNES
R. F. Harrod 12625
MALE AND FEMALE UNDER 18
Nancy Larrick and Eve Merriam, Eds. 29645
A MALE GUIDE TO WOMEN’S LIBERATION
Gene Marine 18291
MAN’S RISE TO CIVILIZATION P. Farb 21576
MAN IN THE TRAP Elsworth F. Baker, Ph.D. 18309
THE ME NOBODY KNOWS ed. Stephen M. Joseph 33969) =1-58
THE NATURE OF HUMAN NATURE Alex Comfort 08797 1.25
(2) DDB 5-77 ©
 2 DISCUS BOOKS
DISTINGUISHED NON-FICTION

NATURE OF POLITICS -M. Curtis 12401 1.95

THE NEW GROUP THERAPIES
Hendrick M. Ruitenbeek 27995 1.95
NOTES OF A PROCESSED BROTHER Donald Reeves 14175 1.95
OF TIME AND SPACE AND OTHER THINGS
Isaac Asimov 24166 1.50
THE OMNI-AMERICANS Albert Murray 11460 1.50
THE RISE AND FALL OF LIN PIAO
Jaap Van Ginneken 32656 2.50
POLITICS AND THE NOVEL Irving Howe 11932 1.65
THE POWER TACTICS OF JESUS CHRIST AND
OTHER ESSAYS Jay Haley 11924 1.65
PRISONERS OF PSYCHIATRY Bruce Ennis 19299 1.65
THE PSYCHOANALYTIC REVOLUTION ; é.
Marthe Robert 08763 1.25
THE LIFE OF EZRA POUND Noel Stock 20909 2.65
THE QUIET CRISIS Stewart Udall 24406 1.75
(RADICAL SOAP OPERA David Zane Mairowitz 28308 2.45
THE ROMAN WAY Edith Hamilton 33993 1.95
RUSSIA AT WAR Alexander Werth 12070 1.65
THE SCHOOLCHILDREN: GROWING UP IN THE SLUMS
-Mary Frances Greene and Orletta Ryan 18929 1.65
SHOULD TREES HAVE STANDING? :
Christopher Stone 25569 1.50
STUDIES ON HYSTERIA Freud and Breuer 16923 1.95
THE TALES OF RABBI NACHMAN Martin Buber 11106 1.45
THINKING ABOUT THE UNTHINKABLE ;
Herman Kahn 12013 1.65
THINKING IS CHILD’S PLAY Evelyn Sharp 29611 1.75
THOMAS WOODROW WILSON Freud and Bullitt 08680 1.25
THREE NEGRO CLASSICS
Introduction by John Hope Franklin 16931 1.65
acne Me ON THE TH EORY OF SEXUALI UALITY Satie 1 os

TOWARDS Aeke CULTURE Caleb Gattegno 11940 1.65

THE WAR BUSINESS George Thayer ‘09308 1.25
WHAT WE OWE CHILDREN Caleb Gattegno 12005 1.65
, REMEMBER ME:
REN TBS SreIN IN PERSON W. G. Rogers 15610 1.65
or eton REICH: . A PERSONAL BIOGRAPHY
1 13738 tes

WOMEN’S ROLE IN CONTEMPORARY SOCIETY 12641 2.45

WRITERS ON THE LEFT Daniel Aaron 12187 1.65

Wherever better paperbacks are sold, or direct from the

publisher. Include 25¢ per copy for postage and handling;
allow 4-6 weeks for delivery.
Avon Books, Mail Order Dept.
250 West 55th Street, New York, N. Y. 10019
@) DDB S77
 EQUINOX & NEW LEADER IN
LARGE-FORMAT SCIENCE FICTION

NOW AVAILABLE IN ONE BIG PAPERBACK VOLUM E

THE CLASSIC TRILOGY READ BY MILLIONS
WINNER OF THE HUGO AWARD |
FOR BEST ALL-TIME SCIENCE FICTION SERIES —

TRIOGY
FOUNDATION |
SECOND FOUNDATION
FOUNDATION
AND EMPIRE
The thousand-year epic... the galac-
tic struggle . . .a monumental work
in the annals of distinguished
science fiction.

26930/$4.95
Now available wherever fine paperbacks are sold,
or direct from the publisher. Include 25¢ per
copy for mailing; allow three weeks for delivery.
Avon Books, Mail Order Dept., 250 West 55th
Street, New York, N.Y. 10019.
A fascinating guide to “‘the new geography”
Beginning with man’s narrow vision of a patch of flat earth, the
renowned science writer Isaac Asimov traces the steps by which
our grasp of the Universe has broadened and deepened to its pre-
sent state. Now man pictures an expanse twenty-six billion light years
in diameter, and approaches the realization of his ancient dream of
conquering all of it.
With the scientist’s knowledge, the professional writer’s craft, and
the poet’s imagination, Dr. Asimov leads a breathtaking voyage of
discovery in space and time that makes even the most complex of
theoretical concepts comprehensible, and that furnishes the basic
knowledge with which to understand new developments. Astronomy
has become the new geography; it is now the shape of outer
space that man must learn in order to continue his explorations.
ILLUSTRATED WITH PHOTOGRAPHS AND DRAWINGS

A DISCUS BOOK/PUBLISHED BY AVON BOOKS