Discovering The Universe

Discovering the universe
Âoe
^5;.
c^
Digitized by the Internet Archive

in
2010
http://www.archive.org/details/discoveringuniveOOkauf
DISCOVERING THE UNIVERSE
William
J.
Kaufmann,
W. H. Freeman and Company
New
York
To Al Tomlinson,
a science teacher extraordinaire!
The northern hemisphere of Uranus

was recorded by Voyager 2 on January
Cover image:
25, 1986, as the spacecraft left the planet behind
This view of Uranus
set forth on a Sf-year cruise to Neptune. Voyager
wide-angle picture was taken
this
(JPL,
of the cresent
is
million kilometers (about
The pale blue-green color
atmosphere That gas absorbs red wavelengths

extreme edge
was
of the planet results
and
600,000 miles) from Uranus when

from methane
in
Uranus's
leaving the predominant hue seen here. The white at the
of sunlight,
caused by a high-altitude haze Voyager 2
will arrive at
Neptune on August
24,
1989
NASA)
Book design: Valerie
Pettis
and Sylvia Woodard
Cover design: Lisa Douglis
Library of Congress Cataloging-in-Publication Data
Kaufmann, William
J.
Discovering the universe.

Includes bibliography and index.
1.
Astronomy,
QB43,2K376
Cosmology,
1987
I,
Title
86-9973
520
ISBN 0-7167-1784-C (pbk)
Illustration credits are listed
Copyright
on page 374
1987
by W. H. Freeman
and Company
No
part of this
book may be reproduced by any mechanical, photographic, or electronic process,
phonographic recording, nor may
it
be stored
in
Printed
in
the
United States of America
234567890
KP
4 3 2
or
the form of a
a retrieval system, transmitted, or otherwise copied for public or
private use, without the written permission of the publisher
10
8 9 8
Contents
Preface
vii
Chapter
Astronomy and the universe
11
Chapter 2
Discovering the heavens
30
Chapter 3
Gravitation and the motions of the planets
47
Chapter 4
Light, optics,
64
Chapter 5
The laws
79
Chapter 6
Our solar system
100
Chapter 7
The major
131
Chapter 8
The Jovian planets
157
Chapter 9
The smaller
183
Chapter 10
Interplanetary vagabonds
202
Chapter 11
The nature
221
Chapter 12
Our star
239
Chapter 13
The
260
Chapter 14
The deaths
282
Chapter 15
Black holes
292
Chapter 16
Our galaxy
306
Chapter 17
Galaxies
and telescopes
of light
terrestrial planets
terrestrial
of the stars
lives of stars
of stars
worlds
Table of (unlents
324
Chapter 18
Quasars and active galaxies
341
Chapter 19
Cosmology: The creation and fate
358
Appendixes
362
Glossary
373
Answers
374
Illustration credits
376
Index
382
Star charts
of the universe
Preface
The
popularity of astronomy as a general science course is due in large

to the inherent intrigue of its subject matter. Since earliest
times people have been fascinated bv topics that astronomers explore
measure
today: the creation of the universe, the formation of the Earth

planets, the motions of the stars, the structure of space
and
and other
Armed
time.
with the powers of observation, the laws of physics, and the resourcefulness of the human mind, astronomers survev alien worlds, follow the life
cycles of stars, and probe the distant reaches of the cosmos.
This book was written with the conviction that the subjects astronomers concern themselves with can and should be imderstood bv just
about everyone. In today's increasingly technological \v(jrld, we face a

multitude of scientific issues of great physical and moral import, and we
informed as best we can. In this spirit. Discovbroad view of astronomy without recourse to
sophisticated mathematical and ph\sical arguments. No background in
science or mathematics is presumed; a minimum of etjuations and formulas appear.
Using a largely descriptive narrative, we investigate and explore
realms far removed from our daily experience. .Many of the objects that
astronomers study are too vast, distant, or intangible to sample directly;
indeed, many of the phenomena that they observe today occurred long
ago. Even the scope of space and time take on new meaning when
viewed in the context of the evolution of the universe.
I have tried to convey more than the intriguing nature of our physical
universe, however. To show students how scientists reason must be an
objective of any tnst science course. H(iw have astronomers come to
know what thev know? Bv studNing the methods astronomers ha\e used
owe
enng
in
it
to ourselves to stav
the Unn'erse offers a
exploring the universe and disco\ering the patterns in the observamake, we can learn something about the nature of scientific
tions thev
inc|uir\
Organization
This text was written for the one-term, descriptive astrononi) course. A
major difference from the parent text. Universe, involves the planetary
astronomv which is treated here in onlv five succinct chapters. .\ similar
judicious condensation of stellar and galactic astronomy permits the entire book to be covered in as few as ten weeks.
The traditional Earth-outward organization of the text allows us to see
how our understanding of the universe developed and to share in the
excitement of astronomical discovery. The first celestial objects to be exaiTiined are those that were observed by the ancient astronomers. As we
move outward from
the planets to the stars
and
galaxies, older issues are
supplanted by newer questions and observations, including those from

outside the visible range and from space. These new observations in turn
raise new questions, which draw the reader on to the outer limits of our
imiverse and oin- understanding.
he first li\e hapiers iniroduce the foundations of astronoin\ including descriptions ol sue h naked-eye observations as eclipses and planetarv motions and siidi basic tools as Kepler's laws and the ojjtics of tele1
Prefai
A discussion of the formation of the solar system in Chapter 6

prepares the reader for the next four chapters, which cover the planets.
Chapter 1 1 leads into stellar astronomy with a discussion of the properties of stars, and Chapter 12 introduces the Sun as a typical star. In
Chapters 13 through 15, stellar evolution is described chronologically
from birth to death. Molecular clouds, star clusters, nebulae, neutron
stars, black holes, and various other phenomena are presented in the
sequence in which they naturally occur in the life of a star, thus unifying
the discussion of the wide variety of objects that astronomers find scattered about the heavens.
A survey of the Milky Way introduces galactic astronomy in Chapter
16, followed by a chapter each on galaxies and quasars. The final chapter, on cosmology, emphasizes exciting recent developments in our understanding of the physics of the early universe.
scopes.
The
illustrations
Pedagogical emphasis
Color illustrations are a distinguishing feature of this book. Most astronomy texts contain only a sampling of color photographs, usually segregated from the corresponding narrative. In this text, however, color
photographs are integral to the text and are incorporated throughout.
We feel that color pictures are indispensable to a truly modern view of
astronomy. Color is now routinely used by astronomers in a wide variety
of circumstances. One glance at a color photograph of a planet's cloudtops or the glowing gases of a nebula reveals significant details about the
object that cannot be gleaned from a black-and-white view alone. In recent years, computer processing of data at nonvisible wavelengths has
produced extraordinary false-color views of the X ray, infrared, and
radio sky. To exclude such colorful views from the main body of the text
not only makes reading more difficult but does a disservice to the field.
very
much
care that students find this book a pleasure to read.
The
dynamic, compelling, and relevant, and no book about it should

be otherwise. I hope these pages will be turned with relish and that interest in astronomy will grow daily.
For those taking astronomy as a first science course, I would have the
experience be as positive and rewarding as possible. Ease in understanding will play a large role in this. To this end, I have tried to emphasize
the central ideas arcjund which astronomy (and indeed other sciences)
revolves. Each chapter begins with a one-paragraph abstract that gives a
clear idea of the chapter's contents. The chapter headings are given in
the form of declarative sentences to highlight main concepts, and a formal summary outlines the essential facts addressed in each chapter. Each
chapter concludes with a series of questions grouped by difficulty and
content into three categories: review, advanced, and discussion. Answers
to questions that require computation (marked by an asterisk) appear at
the end of the book. Care has been taken to formulate questions that
test comprehension of central issues. Roger Culver of Colorado State
field
is
University deserves special recognition for his
regard.
many
contributions in this
Preface
Ancillaries
It is a pleasure to here announce the availal)ility ol

standing ancillaries to Discovering the Universe:
The
Imtructor's
University.
It
tlie
lollowing out-
Manual prepared by Thomas H. Robertson,
Ball State
contains chapter commentaries, quizzes, and suggestions
for lecture topics
and student
activities.
The Computerized Test Bank prepared by Allan T. Clark, The University

of Calgary, Alberta, Canada. Included on this IBM-PC testing program
are short-answer questions with answer keys and the special feature of
accepting an instructor's own questions within the program. Both the
Test Bank and the Instructor's Manual are free to adopters.
Fraknoi's very successful Universe in the Classroom: A Resource
Cuide for Teaching Astronomy is an invaluable book for teaching ideas, experiments, and other related activities.
Andrew
For more information and to request copies of these supplements please

contact:
Marketing Department
W. H. Freeman and Company
41 Madison
New
Acknowledgments
York,
Avenue
New York
10010
the
I am deeply grateful to a number of people who paiticipated in
preparation of this book. Foremost among them is my developmental
editor, Carol Pritchard-Martinez, who worked closely with me on both
this book and its parent text. Universe. I also thank my editor, Jerry
Lyons, for his support and encouragement of this project, and Ceorgia
Lee Hadler, who coordinated production of the book. I acknowledge the
fine efforts of my ccjpyeditor, David R. Hall; Lisa Douglis, who directed
the design production; Deborah Graf-Fox, who managed the art program; Ellen Cash, associate production manager; the drawings by Vantage Art; and the airbrush artistry of George Kelvin.
I also extend sincere thanks to the many instructors who consulted
with us on the preparation of this edition:
Jay Bolman
Robert J. Duke.s
Paul Helminger
University of Central Florida

College of Charleston
LIniversity of South Alaiiama
Richard Henry
Hal R. Jandorf
Moorpark
Yong Hak Kim

Michael Stewart
Takaniasa Takahashi
University of
Oklahoma
Ck)llege
Saddleback College
San Antonio College
St. Norbert College
Preface
I was fortunate to have received many constructive comments and useful

suggestions from instructors who executed prepubUcation reviews of this
manuscript:
Gladwin Comes
Roger A. Freedman
Terrv Rettig
Thomas H. Robertson
Harley Thronson. Jr.
Louis Winkler
Broward Community College

University of California
University of Notre Dame
Ball State University
University of
Wyoming
Pennsvlvania State Universitv
from the reviewers of the parent text. Universe, whose comments and suggestions have had a continuing influence:
as well as
Robert Allen
L'niversity
John M. Burns
Mt. San Antonio College

University of Texas
David
S.
Evans
Andrew Fraknoi
Astronomical Society of the Pacific
Owen
Gingerich
J. Richard Gott,
Paul Hodge
John
K.
III
Lawrence
Dimitri Mihalas
L. D.
Opplinger
John
R. Percy
of Wisconsin
Richard Saenz
F. Scanlon
Richard L. Sears
David B. Slavsky
Joseph S. Tenn
Thomas
Nicholas Wheeler
Donat G. Wentzel
Ravmond E. White
Harvard University
Princeton University
LIniversity of Washington
California State University at Northridge
University of Illinois
Western Michigan University
University of Toronto
California Pohtechnic State University
Grossmont College
University of Michigan
Loyola University of Chicago
Sonoma State University
Reed College
University of Marvland
Universitv of Arizona
In spite of the care exercised in preparing this text, errors
mav have
would appreciate hearing from anyone who finds an error or

who wishes to comment on the text. \'ou mav write to me in care of the
publisher. I will personally respond to all correspondence.
crept
in.
William
J.
Kaufmann,
III
Department of Physics
San Diego State University
Discovering the universe
The Horsehead Nebula
Xeu' stars are form-
ing in the clouds of interstellar gas
shvun
this
Astronomy and the universe
photograph, which
area of the sAv approximate^
and
coi'ers
If
y.
2.
dust
an
The
gases glow because of the radiation emitted by
newborn, massive
stars.
Dark
regions such as
the horsehead are caused by dust graiyis that
block light
The bright
from
the
background
ernmost star in Orion's
and
nebulosity.
star to the left of center

belt.
is
Most of
nebulosity in this photograph are
the eastthe stars
1600
from Earth. (Royal

Observatori. Edinburgh)
light years
Astronoim
is
the study of the universe.
chapters provides
an outline of
the scope
brief prex'ieu' here of the following
and
content of astronomy. We inand powers-oj-ten notation.
troduce important tools such as angular measure
We
learn enough about the solar system, stars, nebulae,
and
galaxies to get
a sense of where we will be going in this book. Above all, we leant that
the universe is indeed comprehensible. Although some questions remain
unanswered, there
is
no reason
to
think that any aspect of the physical
arbitrary or unexplainable.
universe
is
Modern
citv
dwellers usually pay lilde attention to the night skv. If they

to see little more than the
do look toward the heavens, they are likelv

Moon and a few of the brightest stars.
For the inan\ generations who lived without electric lights and smog,
however, the breathtaking panorama of the night sky has been one of
the central experiences of life. Thousands of stars are scattered from
horizon to horizon, with the delicate mist of the Milky Way tracing a
faerie path through the patterns oi brighter stars. The Moon and the
from night to night against this glorious stelbackground while the entire spectacle swings slowly overhead from
planets shift their positions

lar
east to west as the night progresses.
and
Our
ancestors learned to
directions froin these changing patterns in the sky.
tell
time
Thev mapped
the stars into picture outlines of the most important legends and ideas in
their cultures.
Asl}onfiin\
iiiii!
till'
nitivfrsc
Like these earlier people, as we gaze at the stars we find our thoughts
turning to profound questions. How was the universe created? Where
did the Earth, Moon, and Sun come from? What are the planets and
made of? And how do we fit in what is our place in the cosmic
scope of space and time?
Speculation about the nature of the universe is one of the most ancient human endeavors. The study of the stars transcends all boundaries
of culture, geography, and politics. The modern science of astronomy
carries on an ancient tradition of observation and speculation, using the
newest tools of technology and mathematics. In the most literal sense,
astronomy is a universal subject its subject is indeed the universe.
stars
Astronomers use angles to

denote the apparent sizes
and positions of objects in
Astronomers have inherited many useful concepts from antiquity. For

example, ancient mathematicians invented angles and a system of angular measure that is still used to denote the positions and apparent sizes
the sky
of objects in the sky.
An angle is the opening between two lines that meet at a point. Angular measure provides a more exact description of the shape or "size"
01 an angle. The basic imit of angular measure is the degree, designated
by the syinbol . A full circle is divided into 360. A right angle measures
90. As shown in Figure I- 1, the angle between the two "pointer stars" at
the front of the Big Dipper is 5.
Astronomy uses angular measure in a wide range of situations. For
example, we can most easily use an angle to describe how big an object
appears in the sky.
Iinagine looking up at the full moon. The angle (oxered b\ the Moon
is nearly i. We therefore say that the angular diameter or angular size
of the Moon is 5. Alternatively, astronomers sa)' that the Moon subtends
an angle of 5. Ten full moons could fit side by side between the two
pointer stars in the Big Dipper.
I'o talk about smaller angles, we subdivide the degree into 60 minutes
of arc (abbreviated 60 arc min or 60'). A minute of arc is further subdi\ided into 60 arc seconds (abbreviated 60 arc sec or 60"). Thus
1
= 60
arc
=60
arc sec
'
min =
60'
= 60"
\'our thumbnail held at arms' length subtends an angle of
dime viewed from
a distance of
1,
and
mile has an angular diameter of about
2 arc seconds.
Figure 1-1
An
angle
An
riiurU' is ilif
iiig hcnivrii luHi lines Ihiil inert al
piiiut.
lingular di.slance between the two stars

/runt of the
Big Dipper
is
5.
the iingiilnr diameter of the
writing
numbers
ill
The
the
For comparismi.
Moon
Powers-of-ten notation
useful shorthand
upeu-
i\
is
system
In the Astronomical Almanac for 1985, we can read that Venus had an
angular diameter of 42.59 seconds of arc on May 3. That is a very con\enient, precise statement of how big the planet appeared in the sky on
that date.
i.
a
of
Astronoiny is a subject of extremes. As we examine alien environments,

we find an astonishing range of conditions, from the oppressive, aciddrenched clouds of Venus to the near-perfect vacuum of interplanetary
space. To describe these conditions accurately, we need a wide range of
large and small numbers. To avoid such confusing terms as "a million
billion billion," astronomers use a shorthand method called powers-of-ten
notation. All of the cumbersome zeros that accompany a large number
are consolidated into one term consisting of 10 followed by a super-
A.sliviioi/n aiifl the uni;
script,
need
or exponent. The exponent indicates how many zeros you would

out the long form of the number. Thus,
to write
10"=
10'
10-
= 100
10
10-'= 1000
= 10,000
10'
and so
forth
The exponent
tells you how many factors of ten must be multiplied

together to give the desired number. For example, ten thousand can be
10^
written as
(read "ten to the fourth") because 10' = 10 x 10 x 10 x 10
10,000.
With
this notation, numbers are written as a figure between 1 and 10

multiplied by the appropriate power of 10. For example, the distance
between the Earth and the Sun can be written as 1.5 x lO'* km. After
you are familiar with it, you will fmd this notation more convenient than
using "150,000,000 kilometers" or "one hundred and fiftv million kilometers."
This shorthand system can be extended to numbers less than one bv

using a minus sign in front of the exponent. A negatixe exponent tells
you the location of the decimal point, as follows:
10"
10"
Figure 1-2
the renter
is
Examples of powers -of-ten

llie
Taj Miilial.
IOmeter world in
grow .smaller
siotu
irhicli
Dimen-
to the left: first, crystalline
(0.1 millimeter) in size and,
10'"' meter
10"-
0.01
10"'
0.001
10"'
= 0.0001
ailunn the
reach nj our senses.
skeletons of single-celled diatoms
atoms,
Al
0.1
10"' meter
far
left,
tungsten
in diameter.
On
the right,
and
so forth
looking across the Indian Ocean toward the
South Pole, we
see the curimture of the Earth.

10' meters in diameter. At the far right is a
galaxy, 10-' meters in diameter

meters
and
l()-^
from Earth. (Courtesy of Scientific

American Books: NASA: USNO)
For example, the diameter of a hydrogen atom is 1.1 x 10 " cm. That is
easier than saying "0.00000001
centimeters" or "eleven billionths of a
centimeter." Thus, with both very large and very small numbers, the
powers-of-ten notation bypasses all those awkward zeros in a convenient
1
fashion.
-.y---V-
Astronomy and
We
Astronomers use the laws of

physics to explore and
understand the universe
begin, in Ciiapter
tions
and ideas of our
2,
the
un
by examining some of the astronomical observaWe shall see that the course of civiliza-
ancestors.
been dramatically affected by the realization that the universe is

indeed comprehensible. This first glimpse of the power and potential of
the human mind is one of the great gifts to come to us from ancient
Greece. By observing the heavens and carefully thinking about what we
see, we can figure out how the universe operates. For example, we shall
see that the ancient Greeks measured the size of the Earth and undertion has
stood eclipses.
Some people
think astronomy deals with faraway places of no possible

here on Earth, but nothing could be farther from the
truth. For example, in Chapter 3 we see that the seventeenth-century scientist Isaac Newton succeeded in describing how the planets orbit the
Sun. In the motions of the planets, we see some of the most fundamental laws of nature revealed in their simplest forms, unhampered by air
significance to
life
From Newton's work we obtained our first comcoherent description of the behavior of the physical universe. The
resulting body of knowledge, called Newtonian mechanics, speaks in
concrete terms about force, mass, acceleration, momentum, and energy.
It is no coincidence that the Industrial Revolution followed hard on the
heels of this knowledge because Newtonian mechanics provided the theoretical and mathematical basis for immediate practical application in the
construction of machines, factories, buildings, and bridges.
Astronomers use Newtonian mechanics along with other physical prinresistance or friction.
plete,
ciples (usually called the laws of physics), to interpret their observations

and to understand the processes that occur in the universe. The laws of
physics, particularly those involving optics and light, can also be used to
develop new tools and techniques with which to examine and explore
the universe.
In Chapter 4
we
discuss the astronomer's most important tool, the
escope. Until recently, everything
Figure 1-3
An
[left]
Moon Humanity
astronaut on the
has taken
we
out intu the universe. As

worlds,
planet
its
first
small step
explore distant
we gain a new perspective on our own

and a broadened understanding of our
relationship to the cosmos. This photograph
shows Apollo 11 astronaut Edwin Aldrin at

the first lunar landing site
21, 1969.
Figure 1-4 [right]

Space Shuttle
scopes
orbit,
and
on fuly
(NASA)
The Space Shuttle
The
will be used to transport tele-
other astronomical equipment into
far above the obscuring
Earth's atmosphere.
effects
of the
NASA's anginal plans
to
launch interplanetary probes from the Space

Shuttle were set back by a disastrous explosion
during a shuttle launch
in
Galileo mission to Jupiter
1986. Both the
and
the Ulysses mis-
sion over the Sun's poles will have to seek
other launch vehicles because they required the

Shuttle to carry a liquid fuel booster in
its
cargo bay. This photograph shows the first
Space Shuttle rising majestically from Cape

Canaveral on its maiden voyage April 12,
1981. (NASA)
we knew about
tel-
the distant universe
was based on visible light. Astronomers would peer through telescopes,

take photographs of what they saw, and analyze the starlight. Toward
the end of the nineteenth century, however, scientists began discovering
nv
ami
the universe
nonvisible forms of light such as X rays and gamma ravs. radio waves
and microwaves, and ultraviolet and infrared radiation.
Astronomers recently have placed telescopes in orbit to detect
nonvisible forms of light. Far above the obscuring effects of Earth's atmosphere, these astronomical instruments give us views of the universe
vastly different from what oui eves can see. This new information is crucial to our understanding ol such familiar objects as the Sun and gives
us important clues about such exotic objects as neutron stars, pulsars,
quasars,
and black
holes.
We
complete our introduction to astronomy in Chapter 5 with a discussion of light. By understanding how objects emit radiation and how
with matter, we acquire the skills to analyse and interpiet
the wealth of information coming to us from the stars and galaxies. In
light interacts
later chapters
we
shall see
how astronomers
finidamcntal information about
use these
stars, galaxies,
skills to
obtain
and the evolution of the
universe.
By exploring the planets,

astronomers uncover clues
about the formation of the
solar system
Fhis book presents the substance of modern astronomy in three segments, corresponding to three major steps out into the universe; the
planets, the stars, and the galaxies. In Chapters 6 through 10, we explore the solar system, beginning with Earth-like worlds and moving outward toward the frigid depths of space where comets spend most of
their time. Throughout this journey, we shall find that our discoveries
are relevant to the quality of human life here on Earth. Until recently,
our knowledge of such subjects as geology, geophysics, weather, and climate was based on data from only one planet. Earth. With the advent of
space exploration, however, we have a range of other planets with which
to compare our own. As a result, we are making important strides in
understanding the creation and evolution of the Earth and the entire
solar svstem. These investigations give us significant insight into the origin and extent of all our natural resources.
Figure 1-5
Jupiter
is
Jupiter, lo,
orbited by
many
and Europa
mootis.
Jour
oj
which are so large they could qualify as plan

eb, in their
elides are
own
slmeu
right.
Two of
in this view:
these giant miI-
ruddy lo on
Ihi
ice-bound Europa on the right. Close- ujj

examination of these worlds has dramatually
left,
broadened our understanding of Earth-like

Thh photograph was taken try Voyager
planet.s.
I
in
1979 when
the spacecraft
kilometers (12 million miles)
colorful cloudtops.
(NASA)
ims
2(1 million
from jupitn\
Astronom-^
Astronomical distances are

often
measured
in
AUs,
parsecs, or light years
and
the
univene
As we turn toward the stars in the second half of this book, we shall
find that some of our Earth-based traditions become cumbersome. It is
fine to use kilometers (or miles) to give the diameters of craters on the
Moon or the heights of volcanoes on Mars. But it is as awkward to use
kilometers for distances to stars or galaxies as it would be to talk about
the distance from New York to San Francisco in inches or millimeters.
Astronomers have therefore invented new units of measure.
When discussing distance across the solar system, astronomers like to

use a unit of length called the astronomical unit (abbreviated AU),
which is the average distance between the Earth and the Sun:
1
AU
= 1.496 X
10**
km
Thus, for example, the distance between the Sun and Jupiter
is
stated as
AU.
5.2
When
one of two
which is the
talking about distances to the stars, astronomers use
different units of length.
The
light year (abbreviated
ly),
distance light travels in one year:

1
parsec
ly
One
9.46 X 10'2
light year
is
km
roughly equal to 6
Proxima Centauri,
is
4.3 light years
The second commonly used
trillion miles.
The
nearest star,
from Earth.
unit of length
is
the parsec (abbreviated
Imagine taking a journey far into space, beyond the orbit of Pluto.
As you look back toward the Sim, the Earth's orbit subtends a small
angle in the sky. One parsec is the distance at which 1 AU subtends an
angle of one second of arc, as shown in Figure 1-6.
pc).
The
1
Figure 1-6
length
to
light years.
the distance at
of
parsec
commonly used
3.26
The parsec, a
by astronomers,
The parsec
which
AU
is
subtends an angle
how
old,
3.09 X 10'^
The
stars are born,
and eventually die
km
ly
distance to the nearest star, for instance, can be stated as 1.3 pc, or
as 4.3 light years.
personal
By studying stars and

nebulae, astronomers
grow
parsec turns out to be longer than the light year. Specifically,
= 3.26
unit of
u equal
defined as
second of arc.
discover
pc =
Whether one
As we begin our study of the

our own
uses light years or parsecs
is
a matter of
taste.
star,
stars in
the Sun. In the 1920s
Chapters
and 1930s,
1 1
and
12,
we examine
physicists figured out
how the Sun shines. At its center, thermonuclear reactions convert hydrogen into helium. This violent process releases a vast amount of energy that eventually makes its way to the Sun's surface and escapes as
sunlight. By 1950, physicists had learned how to reproduce this thermonuclear reaction here on Earth. Hydrogen bombs operate on the same
basic principles as the energy production at the Sim's center. Thermonuclear weapons stockpiled around the world have a profound effect on
international politics and could dramatically influence the future of life
on our planet. Once again, we see the surprising impact of astronomy on
the course of civilization.
As we look deeper into space, we find star clusters and clouds of

glowing gas, called nebulae, scattered across the sky. In Chapter 13 we
find that these beautiful objects tell us much about the lives of stars. We
discover that stars are born in huge clouds of interstellar gas and dust
such as the Orion Nebula shown in Figure 1-7. After billions of vears.
Astronomy ami the universe
Figure 1-7
The Orion Nebula
[left]
beautiful nebula (also called
1976)
where
the
is
M42
or
This
NGC
a fine example of a stellar "nurser\"
bom. Intense radiation from
stars are
newly formed stars causes the surrounding
Many
gases to glow.
this
nebula are
less
The Onon Nebula
and the
about 23 light
Earth,
of the stars embedded in

than 1 million years old.
is
1600
light years
from
distance across the nebula

years. (U.S.
is
Naval
Obseniatory)
The Crab Nebula
Figure 1-8 [right]

nebula (also called
Ml
supernova remnant.
and
this beautiful
or
NGC
1952)
is
at
about
light years
the nebula
from Earth,
is
still
moving
out-
km/sec (roughly 2 million
The Crab Nebula
is
6300
atid the distance across
about 6 light years.
therefore arrive at the surprising realization that
virtuallv everything, including the
funeral shroud was caused
1000
miles per hour).
We
interstellar material.
by the gases that were blasted violently into
ward
called a
This
dying star exploded,
space. In fact, these gases are
Some end their lives with a spectacular detonation

supernova that blows the star apart, producing objects such as
the Crab Nebula seen in Figure 1-8.
During their death throes, stars return gas to interstellar space. We
shall learn that this gas contains many heavy elements created by thermonuclear reactions in the stars" interiors. Interstellar space thus becomes enriched with chemicals that did not exist in earlier times, as in a
forest where decaying leaves and logs enrich the soil for future generations of trees. The Sun and its planets were formed from enriched
Stars eventually die.
atoms
in
our bodies, was originally
created deep inside ancient, now-dead stars.

In Chapters 14 and 15 we find that dying stars can produce some of
the strangest objects in the sky. Some dead stars become pulsars that
emit pulses of radio waves or bursters that emit powerful bursts of X
rays. Massive dead stars become black holes, surrounded by incredibly
powerful gravity from which nothing (not even light) can escape. Many
of these bizarre stellar corpses have been discovered in recent years with
(Lick Obseniatory)
Earth-orbiting telescopes that detect nonvisible
By observing galaxies,
astronomers learn about
the creation and fate
Stars are not spread uniformly across the universe but are grouped together in huge assemblages called galaxies, the largest individual objects
of the universe
several
in the universe.
hundred
typical large galaxy, like
light.
our own Milky Way. contains
billion stars.
In Chapter 16 we begin with a tour of the Milky Way Galaxy. We discover that our galaxy has beautiful, arching spiral arms (like those of
M83 in Figure 1-9) that are active sites of star formation. We are surprised to learn that the center of our galaxy is emitting vast quantities of
energy.
Exploring other galaxies (Chapter 17), we find that they come in a
wide range of shapes and sizes. Some galaxies are quite small and con-
few hundred million stars. Others are veritable monstrosities

devour neighboring galaxies in a process of "galactic cannibalism."
Some of the most intriguing galaxies appear to be in the throes ol
tain only a
that
violent convulsions.
The
centers of these strange, distorted galaxies are
Aslniii(iiii\ aiifl
iilai
giilaxy
ahiiul
200
M83
The galaxy
Figure 1-9
(ahu culled
XCC
hillien stars.
The
This \f)n
5236)
till'
universe
tin
loiilaiif.
galaxy's spiral
arms are oulUned by numerous nebulae thai

are the
sites
of active star formation. This gal-
axy has a diameter uj about 35.000 light years

(iiiit
IS
at
II
distance uj 12 million light \ears
jrum Earth. (Courtesy
oj
R.J. Dufmir)
often ijovverful sources of X rays and radio waxes. In many cases,

looks as tiiough the entire galaxy is being blown apart.
it
Even more dramatic sources of energy are found still deeper in space.
As described in Cha|3ter 18, at distances of billions of light years from
Earth we find the mysterious quasars. Although tjuasars look like stars
(see Figure 1-10), they are probably the most distant and most luminous
objects in the sky. A typical quasar shines with the brilliance of a hundred galaxies. We shall examine data suggesting that quasars draw their
awesome energy from enormous black holes.
Finally, in Chapter 19, we turn to the most fundamental questions
about the creation and fate of the universe. VVc shall see how the motions of the galaxies reveal that we live in an expanding universe. Extrapolating backward, we learn that the universe was probably born from
an infinitely dense state nearly 20 billion years ago.
Most astronomers believe that the universe began with a cosmic explosion, called the Big Bang, that occurred throughout all space at the beginning of time. During the Big Bang, events happened that dictated the
present nature of the universe. We shall learn how astronomers are making significant progress in understanding these cosmic events. Indeed,
we may be about to discover the origin of some of the most basic properties of the universe. We shall see, finally, how the motions of the most
distant galaxies tell us the ultimate fate of the imiverse: whether it will
expand forever or someday sto]) and collapse back on itself.
An underlying theme throughout is the idea that realit) is rational.
The
hodgepodge of unrelated things beha\ ing in unwe find strong evidence for the existence of
fundamental laws of physics that govern the nature and behavior of
universe
is
not a
predictable ways. Rather,

the
Figure
are the
110
The quasar 3C48 Q_uasurs

most distant and most luminous objects
that astronomers
glance, a quasar
star.
This quasar
have ever
is
is
seen.
At
easily mistaken
thought
first
fur a faint
to be at
of 4.S billion light years from Earth.

(I'atniiiar
Observatory)
a distance
e\erything in the universe. This powerful unifying concept enables us to

explore realms far removed from our earthly experience. Thus a scientist can do laboratory experiments to determine the properties of light
or the behavior of atoms and then use this knowledge to discover the
life cycles of stars and the structure of the universe.
These discoveries have a direct, profound infiuence on humanitv. The
past four centuries of civilization clearly show that major scientific ad-
Aslri)om\
An
Figure l-ll
complex
space. Exvtic atln\s.
among
the universe
Earth-orbiting industrial
lu-u' f!;i)ie>ation oj
high-teihiidhi^
maleriah could easih be manufactured
semiconducting
and
foam
in
metals, ullrapure
crystals. a)id rare vaccines
are
the obvious practical applications nj
zero-gravil\ indiL^lry. This artist's conception
shows an industrial space station under

construction.
(\.\SA)
sooner or later make their way into our lives. In the near future
look forward to the benefits of space technologv Weightlessness
and the near-perfect vacuum of space will enable us to manufacture a
wide range of exceptional substances, from exotic allovs to ultrapurc
varices
we can
medicines.
\'erne and H. G. Wells pale in comparison to the
of today. Ours is an age of exploration and discovery more profound than anv since Columbus and .Magellan set sail across uncharted
seas. We have walked on the moon, dug in the Martian soil, probed the
poisonous clouds of Venus, and seen the craters on Mercury. We have
discovered active volcanoes and barren ice fields on the satellites of Jupiter, and have visited the shimmering rings of Saturn. Never before has
The dreams of Jules
reality
much been revealed in such a short time.

Some people mistakenly see astronomy as a sad and depressing subject telling us that we are insignificant creatures living a brief and meaningless existence on a tiny rock we call the Earth. One of the great lessons of modern astronomv is the awesome power of the human mind to
so
reach out. to explore, to observe, and to comprehend, therein transcending the limitations of our bodies and the brevity of liuman life.
SummarY
Important contributions to astronomical knowledge were

many cultures over the centuries.
made bv
nianv
individuals in
.
The
universe
is
comprehensible.
.\ngular measure and powers-ot-ien notation arc important tools lor the
study of astronomy.
Observation of the heavens has led

menlal laws of nature.
to tlisco\erv
of some of
ilic
limda-
Aslronumy uml
variety of distance units, including
tlie
AU,
the
iDiivme
the parsec and the light
year, are used bv astronomers.
Study of the planets provides information about the Earth's historv and
resources.
Study of the stars and nebulae provides information about the origin and
history of the Sun.
Study of the galaxies provides information about the origin and history of
the universe.
Review questions
With the aid
diagram, explain what
of a
subtends an angle of
2
What
is
it
means
to say that the
Moon
5.
the relationship between degrees, minutes of arc, and seconds of
arc?
numbers using the powers-of-ten notation: (a) ten milfour hinulred thousand, (c) six one-himdredths, (d) seventeen bil-
* 3 Write the following

lion, (b)
lion.
AU?
4 What
is
an
What
is
a parsec?
6 What is the advantage to the astronomer of using the light year as a unit
of distance?
Advanced questions
The speed
from the Sun

*
Discussion questions
of light is 3 x lO'" cm/sec.

to the Earth?
How
long does
it
take light to get
8 The diameter of the Sun is 1.4 x lO" cm and the distance to the nearest
star, Proxima Centain i, is 4.3 light years. If the Sun were reduced to the size
of a basketball (about 30 cm in diameter), at what distance would Proxima
Centauri be from the Sim on this reduced scale?
9 How do astronomical observations and experiments differ from those of

other sciences?
10 Discuss the meaning and justification of the assumption that "reality
is
ra-
"
tional.
For further reading
I. The Measure of the Universe. Harper 8c

journey through the cosmos in half powers 0/
Chaisson, E. Cosmic Dawn. Little, Brown, 1981.
Asimov,
Row, 1983. An
extensive
ten.
Jastrow, R. Red Giants and White Dwarfs. 2nd ed. Norton, 1979.
King, I. "Man in the Universe." Mercun, Nov. /Dec. 1976, p. 7.
Morrison, P., Morrison, P., and The Office of C. and R. Eames. Powers of
Ten. Scientific American Books, 1982. A tour of the universe where
each step corresponds to a power of ten.
Seielstad, G. Cosmic Ecology. University of California Press, 1983.
,
"Cosmic Ecology:
Dec. 1978,
p.
A View from
119.
the Outside In." Mercury, Nov./
Circumpotar star trails This long exposure

M aimed al the south celestial pole and shows
the rotation of the sk\.
ing
tory
is
The foreground
build-
part of the Anglo-Australian Observa-
housing one of the largest telescopes in the
Many
southern hemisphere.
in this book
were taken with
whose primary mirror

in diameter.
During
is
of the photogmphs
this telescope,
3.9 meters (12.H
the exposure,
feet)
someone
carrying a flashlight walked along the dome's

outside catwalk. Another flashlight
the
wavy
trail at
ground
made
level.
(Anglo-Australian Obsematory)
Ancient cultures made many important astronomical obseri'atiom. Modem

astronomers still use constellations described by the ancient Babylonians and
Greeks. In this chapter,
we
leant to find our
conveniently described as the celestial sphere.

related to the
tilt
slowly changiJig
and
of the Earth's
its
orientation.
their relationship to the
motion about the Sun.
We
cuxis
We
way around
We
of rotation
how
which
and
that the CLxis itself
also learn about the phases of the
Moon's motion about
see
the sky,
is
discover that the seasons are
the
Earth and
is
Moon
the Earth's
anciejit astronomers attempted to
measure
and the distances from Earth to the Sun and the Moon.
Eclipses of the Sun and the Aloon played important roles in many of these
early theories and mea.surements. We learn about the conditions under which
eclipses occur and about their physical details.
the size of the Earth
Discovering Ihv heavt
Figure 2-1
munumenl
Stonehenge
w(ti
This astronuniicul
comtruclcd
nfiirly
4000
years
ago on Salisbury Plain in southern England.

Originally, the
monument
consisted of thirty
blocks of gray sandstone, each standing

ers
(13i feet) high,
set in
circle
30
4 met-
meters
(97 feet) in diameter. These stones were topped

unth a continuous circle of smaller stones. Inside the circle are geometrical
arrangements of
other stones, most notably a horseshoe-shaped

set
of larger stones opening toward the north(Courtesy of the British government)
east.
The beauty of the star-filled night sky or the drama of an eclipse would
suffice to make astronomy fascinating. But there are practical reasons as
well for an interest in the universe. The ancient Greeks knew the connection between the seasons and the relative orientation of the Sun and
the Earth. Many early seafaring cultures were aware that the tides are
influenced by the position of the Moon.
Ancient
placed great emphasis on careful astronomical

Hundreds of impressive monuments, such as Stonehenge
civilizations
observation.
(Figure 2-1), that dot the British Isles provide evidence of this preoccupation with astronomy. Alignments of the stones point to the rising and
setting locations of the
ilar
Sun and Moon
at
key times during the year. Sim-
monuments are foimd in the Americas. The

Wyoming, constructed high atop a windswept plateau
astronomically oriented
Medicine Wheel in
by the Plains Indians, has stones and markers aligned with the rising
points of several bright stars, as well as that of the Sun.
Architects of the Mayan city of Chichen Itza on the Yucatan Peninsula
of Mexico built an astronomical observatory, the Caracol, nearly a thousand years ago. The (>aracol has a cylindrical tower that contains windows aligned with the northernmost and southernmost rising and setting
points of both the Sun and the planet Venus. A similar four-story adobe
building, probably constructed during the fourteenth century, is located
at the Casa Grande site in Arizona. And in the ruined city of
Tiahuanaco in Bolivia, ancient engineers built the Temple of the Sun
with walls aligned north south and east west with an accuracy better
than one degree. All of these structures bear witness to careful and
patient astronomical observations by the people of many ancient
civilizations.
Eighty-eight constellations
The
cover the entire sky
Perhaps
ancient roots of astronomy are most apparent in the constellations.

it was shepherds tending their fiocks at night or priests studying
the starry heavens
who
You may already be
first
imagined pictures among groupings
familiar with
such as the Big Dipper, which
is
some of
of stars.
these patterns in the sky
actually part of a large constellation
Major (the Great Bear). Many o( tiiese constellations, such as

Orion in Figure 2-2, have names from ancient nnths and legends. .Although some star groupings vaguely resemble the figures the\ are supposed to represent, most do not.
tailed Ursa
On modern
Figure 2-2
IS
Omm
Orion
winter conslellalinn.
Orion
easily seen
From
Some
is
a promimnl
the L'nileil Slates,
high above the suulheni
horizon from December through March. Be-
cause of the time exposure
four minutes on
Kodak Ekiachrome the colors of the stars are

very rwticeahle. The fanciful drawing of Orion
is from an 1835 star atlas. (Courtesy of
Robert Mitchell and fanus Publications)
star charts, the entire sky is divided into 88 constellations.

constellations cover vcrv large areas (Ursa Major is one of the big-
The constellations are used to specifv certain regions

Thus, for example, we might speak of "the galaxy M31 in
Andromeda" much as we would refer to "the Ural .Mountains in the Soviet Union."
People who spend time outdoors at night are familiar with the apparent motions of the constellations. If you are a city dweller, vou should
lake the time to obser\e the basic facts of astronomy yourself. Go outd(x)rs soon after dark, find a spot away from bright lights, and note the
gest),
others small.
of the sky.
patterns of stars in the skv.
.A
few hours
later,
check again. Vou
that the entire pattern of stars (including the .Moon,
it
it
is
will find
visible) will
have shifted its position. New constellations will ha\e risen above the
eastern horizon, while some will have disappeared below the western
horizon. If you check again just before dawn, vou will find low in the
western sky the stars that were just rising when the night began.
The constellations that you can see in the sky change slowh from one
night to the next. Ihis shift occurs as the Eanh orbits the Sun. as shown
in Figure 2-3. The Earth takes a full year to go once around the Sun,
and thus the darkened, nightiime side of the Karth is gradualh turned
toward different pans of the heavens. Specifically, if you follow a particular star on successive evenings, you find that ii rises approximately 4
minutes earlier each night.
selected
months
of the year
A
is
set of siar
charts for
llie
included at the end of
evening hours of
this
book.
Discuvrriiii^ the
Our changing view of the
Figure 2-3
Ai we
night sky
Sun, the nighttime
orbil the
Earth gradually turns toward
sky of the
differ-
ent parts of the sky. Thus, the constellations

Ihat
henvem
we ran
see
change slowly from
(* "^
Earth
one nighl
)ur
to the next.
wffks
later
^
\
(*-
Earth
two weeks
later
Earth
,^^ //
is
It
often convenient
As you gaze at the heavens on a clear, dark night, you miglil think that
you can see milhons of stars. Actually, the unaided human eye can detect
only about 6000 stars over the entire sky. At any one time, you can see
roughly 3000 stars because only half of the sky is above the horizon. Of
course, the Earth rotates once every 24 hours (that is why we have day
and night), and hence the stars rise in the east and set in the west, as do
the Sun and Moon. This daily, or diurnal, motion of the stars is apparent in time exposure photographs such as Figure 2-4.
to imagine that the stars
are located on the

celestial
sphere
Many
ancient societies believed the Earth to be at the center of the

They also imagined that the stars were attached to the surface
of a huge sphere centered on the Earth. This imaginarv sphere, called
universe.
the celestial sphere,
Of
is still
a useful concept.
course, the stars actually are scattered at various distances
from
Many
of the brightest stars you can see in the sky are 10 to 1000
light years away. These distances are so immense, however, that all the
stars appear to be equally remote, fixed to a spherical backdrop. We can
use this backdrop as a reference to specify the directions to objects in
the sky.
Earth.
As shown
in Figure 2-5, the Earth is at the center of the celestial

can project key geographic features out into space to establish directions and bearings on the celestial sphere. If we project the
Earth's equator onto the celestial sphere, we obtain the celestial equator.
The celestial equator divides the sky into northern and southern hemispheres, just as the Earth's equator divides the Earth into two hemi-
sphere.
We
spheres.
We
can also imagine extending the Earth's north and south poles out
into space along the Earth's axis of rotation. This gives us the north ce-
pole and the south celestial pole, also shown in Figure 2-5.
To denote positions of objects in the skv, astronomers use a svstem
that is very similar to longitude and latitude: "right ascension" and
lestial
Figure 2-4
Star
celestial pole
24
hours,
trails
around the north
The Earth
and hence
rotates
the stars
once
appear
ei'eiy
to
move
across the sky. This time-exposure photogiaph

IS
centered on the yiorth celestial pole, which n
slightly less
star Polaris.
than 1 from the moderately bright
The north
celestial pole, directly
over the Earth's north pole,
which the heavens appear

Nai'al Observatory)
is
a point about
to revolve.
(U.S.
The declination of
angular distance north or south of the celestial equator,
measured along a circle passing through both celestial poles, as shown in
Figure 2-5.
Right ascension corresponds to longitude. Astronomers measure right
ascension from a specific point on the celestial equator called the vernal
equinox, which is one of two locations where the Sim crosses the celestial
equator during the year. The right ascension of an object in the sky is
"declination." Declination corresponds to latitude.
an object
is its
Di.scm'i'inig ihe
Figure 2-5
A\lrimumn\
Celestial coordinates
(ti'iiotv
positions oj iibjech in ihr
and declination,
and latitude on
ski by right ascension
lem similar
hmvetts
to
ii
longitude
Earth. Right a.'.cension
is
measured
\\\-
lln
unil\ /
time (hours, minutes, seconds) eastward along

the celestial equator starting
equinox. Declination
is
from
the vernal
the angle (in degiees.
minutes of arc. secondi of arc) north or south

of Ihe celestial equator.
the angular distance from the vernal equinox eastward along the celestial
measuring its declination (see Figure 2-5).
ecjuator to the circle used in
Following traditional practice, astronomers measure this angular distance

time units (hours, minutes, seconds) corresponding to the time required for the celestial sphere to rotate through this angle.
In catalogues of faint stars, galaxies, and nebulae, the positions of objects are given by their right ascension and declination. This \aluable information tells the astronomer precisely where to point a telescope.
in
The seasons
by the
tilt
are caused
of the Earth's
axis of rotation
In addition to rotating
on
its
axis every 24 hours, the Earth revolves
The seasonal changes we experience on

Earth during a year result from the way the Earth's axis of rotation is
tilted with respect to the plane of the Earth's orbit aroinid the Sun.
The Earth's axis of rotation is not perpendicular to the Earth's orbit.
Instead, the Earth's axis is tilted by 232 away from the perpendicular, as
shown in Figure 2-6. The Earth maintains this tilted orientation as it
orbits the Sun. Thus, during part of the year we Unci the northern
hemisphere tilted toward the Sun and the southern hemisphere tilted
away, producing summer in the north and winter in the south. Half a
year later the situation is reversed, with winter in the northern hemisphere (now tilted away from the Sun) while the southern hemisphere
experiences summer. During March and September, when spring and
fall begin, both hemispheres receive roughly equal amounts of illumination from the Sun.
We can relate this seasonal phenomenon to the celestial sphere by
examining the .Sim's apparent motion against the background stars. As
the Eaith mo\es along its orbit, the Sun appears to shift its position
graduallv from dav to day, tracing out a path in the sky, called the
ecliptic (see Figure 2-7). The Sun takes one year to complete a trip
around the ecliptic. There are about 365 days in a year and 360 in a
aroimd the Sun everv
?i65i days.
DLVuvcriiifr Ihe heiwein
Figure 2-6
of rolalwn
is
The seasons The Earth's a\i\

23 f away from the jii-t-
inclined
jjnnlinilnr tu Ihe plane of the Earth's
The Earth maintains

axis oj rotation
this
aimed
orliit.
orientation (with
its
at the celestial north
pole near the star called Polaris) throughout

Ihe
year as the Earth orbits the Sun. Con,\e-
amount of solar illumination and

number of daylight hours at any location
ijuently. the
Ihe
on Earth varies
a regular fashion through-
out the year.
Figure 2-7
ecliptic
on Ihe
IS
Ihe
Equinoxes and
the celestial equator by
This path
is
Ihe celestial equator intersect at
ini the ecliptic is
oj the
ecliptic
tilt
and
two points
The northernmost point
called the
summer
solstice.
The corresponding southernmost point

Ihe winter solstice.
Sun
inclined to
23i because of the
of the Earth's axis of rotation. The
called the equinoxes.
The
solstices
apparent annual path
celestial sphere.
is
called
the Sun appears to move along tiie ecliptic at a rate of approximately 1 per day.
Because of the tilt of the Earth's axis of rotation, the ecliptic and the
celestial equator are inclined to each other by 235, as shown in Figure
2-7. These two circles intersect at only two points, which are exactly opposite each other on the celestial sphere. Both points are called equinoxes (from the Latin words meaning "equal night") because, when the
Sun appears at either point, daytime and nighttime are each 12 hours
long at all locations on Earth.
The vernal equinox marks the beginning of spring in the northern
hemisphere, as the Sun moves northward across the celestial equator in
circle, so
Discuveiiiif; the
heavem
March. Tlie autumnal equinox marks the moment when lall hegW
northern heniispheic (.ihout September 22), as the Sun moves
southward across the ce.estial equator.
We should always ienieml)er liiat tiie northern and southern hemisjiheres experience ojjposite seasons at an\ given point in time. I-'or example, Marcii 21 marks the beginning of autumn lor people in Australia. These terms come irom a time when \irtuall\ all asironomers li\ed
late
in the
north of the equator.
Between the vernal and auuunnal e(ininoxes are t\vo other signifkani
The poini on the ecliptic farthest north of
locations along the ecliptic.
the celestial equator
is
called the
summer
solstice.
It is
the location of
Sun at the moment summer begins in the northern hemispheie.

about June 21. At the beginning of winter, about December 21. the Sun
the
Sept. 22
The Sun's daily path On llie

fini clay a/ sfmiit^ mid thr fii^l ilin / //ill. ihc
Sun riifi pieci.sch in Ihc cm/ und sch Imtuch
in the west. During summer, the Sun run in
the northeast and sets in the northumt. The
maximum uorthert\ e.xcursion of the Sun occurs at the summer solstice. In the winter, the
Sun rises in the southeast and sets in the
Figure 2-8
southwest, with
sion
occumng
Precession
motion
its
ina.ximnm simthcrh excur-
at the winter solstice.
is
a slow, conical
of the Earth's
axis of rotation
is
farthest south of the celestial equator at a point called the winter sol-
stice.
Seasonal changes in the Sim's dailv path across the sk\ aie diagramed
Figure 2-8. On the first day of spring or fall (when the Sun is at one
of the equinoxes), the Sun rises directly in the east and sets directiv in
the west. Daytime and nighttime are of equal duration.
in
During the sumiuer months, when the northern hemisphere is tilted

toward the Sun, sunrise occurs in the northeast and sunset occurs in the
northwest. The Sun spends more than 12 hours above the horizon and
passes high in the sky at noontime. Incidentally, the point in the sky directly overhead is called the zenith, as shown in Figure 2-8. At the summer solstice, the Sun is as far north as it gets, giving the greatest number of da\ light hours. Indeed, north of the Arctic Circle the Sun does
not set at all during some summer nights.
During the winter months, when the northern hemisphere is tilted
awav from the Sun, sunrise occurs in the southeast. Davlight lasts for less
than 12 hours as the Sun skims U)w over the southern horizon and sets
in the southwest. Night is longest when the Siui is at the winter solstice.
In fad, north of the Arctic C.ircle a winter night can last for 24 hours.
Ancient astronomers realized thai the Moon orbits the Earth. IhcN knew
that the Moon takes roughly four weeks to go once around the Earth.
Indeed, the word "month" tomes frt)m the same Old English rt)t)t as the
word "moon."
As seen from the Earth, the Moon is never far from the ecliptic. In
other words, the Moon's path among the constellations is close to the
Sim's paih. The Moon's path remains within a band called the zodiac
that extends about 8 on either side of the ecliptic. fweUe famous constellations lie alt)ng the zodiac, and the Mt)t)n is generalU found in t)ne
of
them.
lestial
.\s
the
Moon
int)ves alt)ng
celestial
equator for
its
orbit,
it
appears ntirth of the ce-
and then
appeals
about the nexi two weeks.
eciuator for about iwo weeks,
Both the Sun and the
il
soiiih
t)f
the
Moon
exert a gravitational pull on the Earth.

greater detail in Chapter $. For now, it is sufficient to realize that gravity is the universal attraction of matter for tither
matter.
We
will di.scuss gravity in
The gravitational pull of the Sun and .Moon alUt the Earth's rotation
because the Earth is nt)l a perfect sphere. Our planet is slighilv lalier, by
about 4?> kilometers (27 miles), across the equator than it is fn)m pole to
pole. I'he Earth is therefore said to ha\e an "equatorial bulge." The
gravitational pull of the Moon an<f ilic Sun on this equatorial bulge
I
Discoveruig Ihf hfavfus
Fi^re 2-9
rquatorial bulge
the
came
Earth ^recesses,
traces out a circle
analogous
The gravitational
Precession
Moon and
pills of the
to a
its
the
the
Sun on
Earth
the Earth's
As
to precess.
axis of rotation slowly
the sky.
spinning
top.
The
As
situation
is
the top spins,
the Earth's gravitational pull causes the top's
axis of rolaiiiin to
move
in
circle.
fr-
/'/'// W
'
'/
'
gradually changes the orientation of the Earth's axis of rotation, producing a phenomenon called precession.
Imagine a spinning toy top, as illustrated in Figure 2-9. If the top
CvKMUs
were not spinning, gravity would pull it over on its side. But when it is
spinning, the combined actions of gravity and rotation cause the top's
axis of rotation to trace a circle, a motion called precession.
As the Svm and Moon mo\c along the zodiac, each spends half the
time north of the Earth's ec|uat()rial bulge and half the time south of it.
The gravitational pull of the Sini and Moon tugging on the equatorial
bulge tries to "straighten up" the Earth. In other words, as sketched in
Figure 2-9, the gravity of the Sun and Moon tries to pull the Earth's axis
of rotation toward a position perpendicular to the plane of the ecliptic.
But the Earth is spinning. As with the toy top. the combined actions of
gravity and rotation cause the Earth's axis to trace out a circle in the sky
while remaining tilted about 23i to the perpendicular.
The Earth's rate of precession is fairly slow. It takes 26,000 years for
the north celestial pole to complete one full precessional circle around
the sky (see Figure 2-10). Al the present time, the Earth's axis of rotation points within 1 of the star Polaris. In 'MHH) bc, it was pointing near
Thuban
of Draco (the dragon). In ad

Lyra (the harp). Of course, the
south celestial pole executes a similar circle in the southern sky.
As the Earth's axis of rotation precesses, the Earth's equatorial plane
also moves. Because the Earth's equatorial plane defines the location of
the celestial equator in the sky, the celestial equator also precesses. The
intersections of the celestial equator and the ecliptic define the equinoxes, so these key locations in the sky also shift slowly from year to
year. The entire phenomenon is often called the precession of the equinoxes. Today the \ernal equinox is located in the constellation Pisces
(the fishes). Two thousand years ago, it was in Aries (the ram). Around
the year ad 3000, the vernal equinox will move into Aquarius (the water
the star
in the constellation
14,000, the "pole star" will be
Ursa Maji
Path
ol
north
ti-kstial polf
Figure 2-10
celestial pole
The path of
.As the
the north
Earth precesses. the
north celestial pole slowh traces out a circle
among
the northern constellations in the
At the present time, the north
sh.
celestial pole
near the moderately bright star Polaris,
is
u'hicli
Vega
in
serves as the "pole star."
bearer).
Lunar phases are due to the

orbital motion
Ancient astronomers knew that the Moon shines by reflected sunlight.

They also understood that the Moon takes roughly four weeks to orbit
the Earth. This knowledge was deduced from observations of the chang
ing phases of the Moon as varying amounts of its illuminated hemisphere are exposed to observers on the Earth.
Moon's
Di.sroveniig the
heavem
Figure 2-11
19
an extraordinary photograph of both the Earth and

The Moon orbits the Earth every 27.3 days, so it
takes approximately foin weeics for the Moon to complete its cycle ol
phases, as shown in Figure 2-12.
The phase called new moon occurs when the unlit hemisphere of the
Mcjon faces the Earth. The Moon is not visible in this phase because it is
in the same part of the sky as the Sun.
During the next seven days. Earth-based observers see a phase called
waxing crescent moon, in which more and more of the illuminated
hemisphere is progressively exposed to our view. At first quarter moon,
the angle between the Sun, Moon, and Earth is 90, so we see half of the
Moon from
moon
is
space.
illuminated.
Dining the next week, still more of the ilknninated hemisphere is

seen from Earth, in a phase called waxing gibbous moon. When the
Moon
Figure 2-11
The Moon
The Earth and
circles the
the
Moon
Earth every 27 3 days at
an average distance of 384,400 kilometers

(238,900 miles). This picture was taken in
1977 from the Voyager 1 spacecraft after it
was launched toward Jupiter and
Saturn.
Figure
(NASA)
21 2
Light jrom the
Moon,
Moon
orbits the
amounts of
It
takes
all
its
The phases of the Moon

Sun dhimmalis iinr half of
while the other halj n dark.
Earth,
we
see
the
.\s the
varying
the Moon's illuminated hemisphere.

29i days fur the Mimn to go through
pha.ses.
stands opposite the
Sun
in the sky,
we
see the fully illuminated
hemisphere, producing the phase called full moon. Moonrise always occurs at sunset during full moon.
Over the subsequent two weeks, we see less and less of the illuminated
hemisphere as the Moon continues along its orbit. This progression produces the phases called waning gibbous moon, last quarter moon, and
waning crescent moon, as diagramed in Figure 2-12.
Because the position of the Sun in the sky determines the local time,
we can correlate the Moon's phase and location with the time of day. For
^pn^ I
^^ÊSE^
'
^^'îiito
yl
^1
^^^P KI!I!^^^^^^H
If^îw 1
^r^mI^ mf
'.*^-^^^
kvl
HHf^HK^ >'
Waxing crescent
Kirst
Figure 2-13
Moon
The Moon's appearance
The
aluKiM keeps the same side faring the
Earth. Earth-based observers therefore always

see the
same
craters
and lunar mountains,
re-
gardless of the phase. (Lick Observatory)
months
The
lakes the
Moon
respect to the
sidereal
background
cause the Earth

orbit
month
is
slightly
one new moon
month
is
However,
stars.
Moon
u'ilh
be-
slightly longer
its
must travel
more than 360
to the next.
Thus
to get
from
the synodic
than the
sidereal monlh.
monlh completed
when Moon was here
^Sidereal
(age: 10 days)
Full
(age:
Mdon
14 days)
example, during first quarter moon, the Moon is approximately 90 east

of the Sun in the sky: hence, moonrise occurs approximately at noon.
Figure 2-13 is a series of photographs showing the various phases of the
moon.
relative to the stars or to the Sun. Neither interval
it
constanth moving along
about the Sun, the
through
the time
is
complete one revolution
to
i1
gibbotis
It takes about one month for the Moon to complete an orbit around
the Earth. However, astronomers are careful to distinguish between two
types of months, depending on whether the Moon's motion is measured
The sidereal and synodic
Figure 2-14
jî^^l
Waxing
^^^-
kS
^PP?^^''''''
(age: 7 days)
(age: 4 days)
K m KwHy
rj m
quarter
the
corresponds exactly to
months of our calendar.
The
month is the time it takes for the Moon to complete one

of the Earth, measured iinlh respect to the stars. This interval is
the Moons' true orbital period, equal to 27.3 days.
The synodic month is the time it takes lor the Moon to complete one
cycle of phases, as for example, from one new moon to the next. Consequently, the synodic month is measined with respect lo the Sun and is
equal to about 29.5 days.
Of course, the Earth is orbiting the Sun while the Moon is going
sidereal
full orbit
through its phases. Thus, to get from one new moon to the next, the
Moon must travel more than 360 along its orbit, as shown in Figure
2-14. The synodic month is thus approximately two days longer than the
sidereal month.
The Moon stays in orbit about the Earth because of the gravitational
attraction between these two bodies. (We shall discuss gravity and orbits
in greater detail in the next chapter.) In the absence of a third body, the
Moon's orbit would remain forever imchanged. In addition to the Earth,
however, the Sun also pulls on the Moon. The Sun's gravitational pull is
continually producing small changes in the Moon's path around the
Earth. The final result is that both the sidereal and the synodic months
are variable. The sidereal month (average length = 2T^ 7 43"' IP) can
vary by as inuch as seven hours. The synodic month (average length =
29d 12'' 44'" 3^)
can vary by as much as half a day.
Tiiig ihe heavt
^^^^^K^/f^^Bi
1
B^^^Hi^B
^^!9i91
^^^ ^HkBhR
^^^^- -^^
'fi
:''.^^^|
>.!.'
'i^^H
'/:?I^^^^H
,'
.:
''-'^^^^^1
Bitaiifl
Waning gibbous
20 days)
(age:
Ancient astronomers
measured the size of the
Earth and attempted to
determine distances to the
Sun and Moon
Last quarter
Waning
(age: 22 days)
(age:
crescent
26 days)
More than two thousand years ago. Creek astronomers were fully aware
of the Earth's spherical shape. Eclipses of the Moon pro\ ided the convincing observations. During a lunar eclipse, the Moon passes through
the Earth's shadow. Ancient astronomers noticed that the edge of the
Earth's shadow is always circular. A sphere is the only shape that casts a
circular shadow from any angle, so the ancient astronomers concluded
that the Earth is spherical.
Around 200 bc, the Greek astronomer Eratosthenes devised a way to

measure the circumference of the Earth. He was intrigued by reliable
reports from the town of Syene in Egypt (near the modern Aswan) that
the Sun shone directly down vertical wells on the first day of summer.
Eratosthenes knew that the Sun never appeared at the zenith from his
home in Alexandria, on the Mediterranean Sea almost due north of
Syene. He measured the position of the Sun at local noon on the summer solstice in Alexandria and found it to be about 7 south of the zenith, as in Figure 2-15. This angle is one-fiftieth of a circle, so Eratosthenes concluded that the distance from Alexandria to Syene was
one-fiftieth the Earth's circumference.
In Eratosthenes' day, the distance from Alexandria to Syene was said

5000 stades. (The stade was a Greek unit of length.) Eratosthenes
found the Earth's circumference to be 50 x 5000 stades = 250,000
to be
no one today is sure of the exact length of the

guess is that it was about k kilometer, which would mean that
Eratosthenes obtained a circumference of about 42.000 kilometers
within 5 percent of the modern value of about lO.OOO kilometers.
Eratosthenes was only one of several brilliant astronomers to emerge
from the distinguished Alexandrian school. .Around 280 bc, one of the
stades. Unfortunately,
slade.
Figure 2-15
Eratosthenes' method of
determining the Earth's size Emliisllwnes
milind llidt Ihr Sun is iihoiit 7 saiilh aj the
zenith
head
of
III
Alfxiitidim
at Syene.
when
The angle
circle, so the
and Syene must
it
is
is
directly iivn-
about one-fiftieth
distance between Alexandria
be
rilwiil mie-jilttrlh of the
Earth's circumference.
One
first Alexandi ian astronomers, Aristarchus of Sanios, devised a method

of determining the relative distances to the Sun and Moon.
Aristarchus reasoned that the angle between the Sun, Moon, and
Earth is exactly 90 at the moment of f list or last quarter moon, as dia-
First
quarter
in Figure 2-16. He also argued that the Moon takes longer to go

from first quarter to last quarter than it does to go from last quarter
around to first quarter. Observations led Aristarchus to believe that the
travel time from first to last quarter is one day longer that the interval
from last to first quarter. He used this difference to determine the rest
of the angles in the triangles in Figure 2-16. With additional geometric
arguments, Aristarchus then deduced the relative lengths of the sides of
gramed
moon
the triangles.
Last quarter
Figure 2-16
Aristarchus's method of
determining distances
Moon
to the
quarter tlian
measuring
this
to
it
go from
does
from
Earth,
and Moon
He was
tive lengths
first
at first
formed
and
last
by the
to
By
Sun.
quarter
then able to calculate the rela-
of the sides of these triangles and
size
Sun and
of the Moon's orbit
is
greatly
exaggerated in the diagram.
Eclipses occur only
when
the
Sun and Moon are both on

the line of nodes
main errors arose from the
inability to
instant of first or last quarter visually. Nevertheless,
quarter
last to first.
thereby obtain the distances to the
Moon. The
Moon
lime difference, he determined
the angles in the triangles
phases.
Aristarchus's
Sun and
Anstarchus argued that the
should take longer

last
Aristarchus concluded that the Sun is only 20 times farther from us

than the Moon is. We now know that the average distance to the Sun is
about 390 times longer than the average distance to the Moon. One of
moon
determine the exact
is impressive that
people were logically trying to measure distances across the solar system
more than two thousand years ago.
Aristarchus also used hmar eclipses in an equally bold attempt to determine the relative sizes of the Earth, Moon, and Sun. Observing how
long it takes for the Moon to move through the Earth's shadow, Aristarchus estimated that the diameter of the Earth is about three times the
diameter of the Moon. To determine the diameter of the Sim. he simply
pointed out that the Sun and the Moon have the same angular size in
the sky, so their diameters must be in the same ratio as their distances.
In other words, because Aristarchus believed that the Sun is 20 times
farther from the Earth than the Moon, he concluded that the Sun must
be 20 times larger than the Moon.
You can now appreciate the significance of Eratosthenes' measurement of the Earth's circumference. The Greeks knew that the Earth's
diameter is equal to its circumference divided by the constant called tt
(pi). Knowing the Earth's diameter, Alexandrian astronomers could calculate the diameters of the Sun and Moon as well as their distances from
Earth. Although some of these ancient measurements are far from their
modern values, our ancestors' achievements stand as an impressive exercise in observation and reasoning.
it
among the most spectacular of nature's phenomena. In only

few minutes, the Sun can seemingly be blotted from the sky as broad
davlight is suddenlv steeped in eerie twilight, or the brilliant full moon
on a cloud-free night can gradually grow dark and red.
A lunar eclipse occurs when the Moon passes through the Earth's
shadow. This can happen only when the Sun, Earth, and Moon are in a
straight line at full moon. A solar eclipse occurs when the Earth passes
through the Moon's shadow. As seen from Earth, the Moon moves in
front of the Sun. This can happen onlv when the Sun, Moon, and Earth
Eclipses are
are aligned at
new moon.
New moon and
full
moon
both occur
at intervals
of 29? days, but
and lunar eclipses happen much less frequendy. This scarcity of

eclipses is due to the Moon's orbit being tilted slightly (5) out of the
plane of the Earth's orbit, as shown in Figure 2-17. Because of this tilt,
new moon and full moon usually occur when the Moon is either above
solar
or below the plane of the Earth's orbit. In such positions, a perfect alignment between the Sun, Moon, and Earth is not possible and an eclipse
cannot occur.
The plane of the Earth's orbit and the plane of the Moon's orbit intersect along a line called the line of nodes, which passes through the
Earth and is pointed in a particular direction in space, as shown in Fig-
Figure 2-17
of
lite
to the
Moon's
The
orbit
The plane
line of nodes
tl
tilled slightly
plane oj the Earth's
orbit.
with respect
These two
planes intersect along a line called the line
of nodes.
lire 2-18. Eclipses can occur onlv when both the Sun and Moon are on
or very near the line of nodes, because onh then do the Sun. Earth, and
Moon lie along a straight line.
Knowing the orientation of the line of nodes is clearlv important to
anyone who wants to predict upcoming eclipses. However, predicting
eclipses is complicated b\ the fact that the line of nodes graduallv
changes its direction in space. The constant gra\ itational pull of the Sun
on the Moon causes the Moon's orbit gradualh' to shift its orientation in
space. The resulting slow westward movement of the line of nodes is one
of several details that astronomers must include in their calculations for
times of upcoming eclipses.
At least two and not more than five solar eclipses occur each vear.
(The last year in which five solar eclipses occurred was 1935.) Lunar
eclipses occur just about as frequentlv as solar eclipses, but, the maximum number of eclipses (both solar and lunar) possible in a vear is
seven.
Figure 2-18
Conditions for eclipses
solar eclipse occurs only
ij
the
Moon
near the line of nodes at new moon.

eclipse occurs only if the
line of nodes at full
Moon
moon.
is
is
.1
very
lunar
very near the
Diicuverinii thr lieav
Lunar and solar eclipses can

be either partial or total,
depending on the alignment
of the Sun, Earth, and Moon
During the few hours of a lunar
move
across the
Moon. The
eclipse,
Earth's
you can see the Earth's shadow
shadow has two
distinct parts, as dia-
irramed in Figure 2-19. The umbra is the darkest part of the shadow,
from which no portion of the Sun's surface can be seen. In the penumbra, only part of the Sun's surface is blocked out.
Three kinds of lunar eclipses can occur, depending on exactly how
the
Moon
travels
through the Earth's shadow. First, the Moon miglit

penumbra, creating a penumbral eclipse.
pass through only the Earth's
During this type of eclipse, none of the lunar sinface is completely

shaded by the Earth because only part of the Sun is covered by the
Earth. At mid-eclipse, the Moon merely looks a little dimmer than usual
from Earth and there is no "bite" taken out of the Moon by the Earth's
umbra. The Moon still looks full, and thus it is easy to miss a penimibral
eclipse.
The geometry of a lunar
Figure 2-19
eclipse
People on the iiighllime side
Earth see a lunar eclipse when the
tij
llie
Moon
mmies through the Earth's shadow. The umbra

is
the darkest part
numbra, only part
of the shadow, hi the peof the Sun is covered b\
the Earth.
Most people notice a limar eclipse onl\ if the Moon passes into the
umbra. During the umbral phase of such an eclipse, a "bite"
seems to be taken out of the Moon. It the .Moon's orbit is oriented so
that only part of the lunar surface passes through the umbra, then we
see a partial eclipse. When the Moon traxels completely into the umbra,
Earth's
we see a total eclipse of the Moon. The

duration of totality occurs when the Moon travels directly
through the center of the umbra. The Moon's speed through the Earth's
shadow is roughly 1 kilometer per second (2300 miles per hour), which
as sketched in Figure 2-20,
maximum
that totality can last for as much as

hour 42 minutes.
As an example of the frequency of lunar eclipses, Table 2-1 lists
the total and partial eclipses from 1985 through 1990. Penumbral
means
all
eclipses are not included in this listing.
The Moon does
small
amount of
not completeh disappear, even during a total eclipse.

sunlight passing through the Earth's atmosphere is
deflected into the Earth's umbra. Most of this deflected
To penumbral
Figure 2-20
Various lunar eclipses
diagram shows
the Earth's
numbra
umbra and
at the distance of the
Moon's
To
total eclipse
To
partial eclipse
Thr
pe-
orbit.
Different kinds of lunar eclipses are seen,
depending on the Moon's path through

Earth's shadow.
eclipse
tin-
lisjht is
red, so
TABLE
2-1
Lunar
eclipses
Date
100
1985
100
24 April
1986
100
100
14
October
1986
October
1987
of the
Moon
tronomer during the lunar eclipse of September

6, 1979. Notice the dtstincth reddish color of
the
30
20 February 1989
hr 10 min
42
1988
100
16
1989
100
38
9 February 1990
100
46
68
17
total eclipse
totality
total)
1985
27 August
Duration of
(100% =
May
Thi\ photograph was taken b\ an amateur as-
Percentage eclipsed
28 October
Figure 2-21
1985-1990
August
6 August
1990
Moon. (Courtesy of Mike Harms)
the darkened Moon glows lainth in reddish hues during totality, as

shown in Figure 2-2
As with the Earth's shadow, the darkest part of the Moon's sliadow is
also called tiie umbra. Voii must be inside the Moon's umbra in order to
see a total solar eclipse, for that is the only region from which the
Moon completely coyers the Sun. Because the Sun and Moon have
nearly the same angular diameter as seen from Earth
about i the
Moon "fits" oyer the Sun during a total solar eclipse. This is important
1
to astronomers because hot gases, called the solar corona, that surround
the Sun can be photographed and studied in detail during the few precious moments when the eclipse is total (see Figure 2-22).
Figure 2-22
During a
total eclipse of the
total solar eclipse, the
pleteh covers the Sun's disk,
Sun
Moon
and
com-
the solar co-
rona can be seen. This halo of hot gases exlends for thousands upon thousands of
kilometers into space.
Only the
hrighlesl. inner
portions of the solar corona are seen in this
photograph of the solar eclipse of .March
1970.
\'AS A
7.
Immediately surrounding the Moon's umbra is the region called the

penumbra, from which the Sun's surface appears only partly coyered by
the Mi)on. During a solar eclipse, the Moon's penumbra extends oyer a
large portion of the Earth's surface, .\nyone standing inside the penumbra sees a partial eclipse of the Sun.
Only the tip of the Moon's umbra reaches the Earth's surface, as in
Figure 2-23. As the Earth turns, the tip of the imibra traces an eclipse
path across its surface. Only those people inside this path are treated to
the spectacle of a total solar eclipse. (Figure 2-24 shows the dark spot
produced by the Moon's umbra on the Earth's surface.)
During
a total eclipse, the Earth's rotation and the orbital motion of

cause the imibra to race along the eclipse path at speeds in
excess of 1700 kilometers per hour (1050 miles per hour), so most people
along the eclipse path obserye totality for only a few moments. Totality
neyer lasts for more than 7A minutes at any one location on the eclipse
path. In a typical total solar eclipse, the angle of the Sun and the EarthMoon flistance together limit the a duration of totality to much less than
the maximum 7i mimiies.
The width of the eclipse path depends primarily on the Earth-Moon
distance dining an eclipse. The eclipse path is widest
up to 270 kilometers (170 miles)
if the Moon happens to be at the point in its orbit
nearest the Earth. Usually it is much narrower. In fact, sometimes the
.Moon's umbra docs not eyen reach down to the Earth's surface. If the
alignment for a solar eclipse occurs when the Moon is farthest from the
Earth, then the Moon's umbra falls short of the Earth, and no one sees a
the
Moon
Figure 2-23
The geometry of a total solar

During a total solar eclipse, the lip of
Moon's umbra traces an eclipse path across
eclipse
the
the Earth's surface. People inside the eclipse
path see a
inside the
total solar eclipse,
penumbra
see only
whereas people
partial eclipse.
The Moon's shadow on
Figure 2-24
Earth
the
This photograph was taken from an
Earlh-orbiting satellite during the total eclipse
of
March
7.
1970. The Moon's umbra ap-
pears as a dark spot on the eastern coast of thr
United
States.
(NASA)
the Earth's surface, the Moon appears too small

a thin ring of light is seen around the
edge of the Moon at mid-eclipse. An eclipse of this type is called an annular eclipse. The length of the Moon's imrbra is nearly 5000 kilometers
truly total eclipse.
to cover the
From
Sun completely, and
(3100 miles) shorter than the average distance between the Moon and
shadow often fails to reach the
Earth, making annular eclipses more common than total eclipses.
As an example of the frequency of solar eclipses, Table 2-2 lists all
the total, annular, and partial eclipses from 1985 through 1990. All of
the details of a solar eclipse are calculated well in advance and are published in reference books such as The Astronomical Almanac.
the Earth's surface. Thus, the Moon's
TABLE
2-2
Solar eclipses 1985-1990
Date
Area
Type
Notes
1985 May 19
Arctic
Partial
84%
1985 November 12
S. Pacific,
Total
Max. length
1986 April 9
Antarctic
Partial
82%
1986 October 3
N.
Total
Max. length Om Is
Annular along most of track
1987 March 29
Argentina, Atlantic, Central Africa
Total
Max. length 56s in Atlantic

Annular along most of track
1987 September 23
Russia, China, Pacific
Annular
1988 March 18
Indian Ocean, E. Indies. Pacific
Total
Max. length
1989 March 7
Arctic
Partial
83%
eclipsed
1989 August 31
Antarctic
Partial
63%
eclipsed
1990 January 26
Antarctic
Annular
1990 July 22
Finland, Russia, Pacific
Total
Antarctica
.i^tlantic
eclipsed
Ini 55s
eclipsed
Max. length
3m
46s
2m
33s
Diicuvering
Ancient astronomers
achieved limited ability to
llie
heavem
is a dramatic event. Ihe sk\ begins to darken, the

temperature falls, and the winds increase as the Moon's umbra races
toward \ou. All nature losptjnds: birds go to roost, flowers close their
petals, and crickets begin to chirp as if evening had arrived. As totality
approaches, the landscape is bathed in shimmering bands of light and
dark as the last few rays of sunlight peek out from behind the edge of
the Moon. And finally the corona blazes forth in a star-studded midday
sky. It is an awesome sight.
total solar eclipse
air
predict eclipses
In ancient times, the ability to predict eclipses must have seemed very
desirable. Archaeological evidence suggests that astronomers in many civ-
struggled to predict eclipses, with varving degrees of success.

certain holes in the ground around
Stonehenge suggest some ability to predict eclipses more than 4000 years
ago. One of three priceless manuscripts to survive the devastating Spanilizations
The number and placement of
Mayan astronomers in Mexico and Guatemala

method of eclipse prediction. And there are many
such as the one about the great Greek astronomer
conquests shows that
ish
had
a fairly reliable
apocryphal stories
Thales of Miletus, who is said to have predicted the eclipse of 585 bc,
which occurred during the middle of a war. The sight was so awesome
and unexpected that the soldiers put down their arms and declared
peace.
In retrospect, it seems that ancient astronomers actually produced
eclipse "warnings" of various degrees of reliability rather than actual predictions. Working with historical records, these astronomers generally
searched for cycles and regularities from which to anticipate future

eclipses.
Ancient astronomers, particularly those of Greece, gave humanity a

thinking about the world. They gave the first
clear demonstration that the tools of logic, reason, and mathematics can
be used to discover and understand the workings of the universe. This
new and powerful way of
general approach to reality underlies all modern science. Aristarchus

even went so far as to suggest that the motions of the planets in the sky
could be simply explained if all the planets, including Earth, orbit the
Sun. As we shall see in the next chapter, he was nearly 2000 years ahead
of his time.
Summary
It is convenient to imagine the stars fixed to the

Earth at its center.
rhe surface of the
celestial
sphere
is
di\
celestial
sphere with the
ided into 88 regions, or constella-
tions.
The
celestial
and
night; in fact, of course,
in
each dav
The
poles and equator of the celestial sphere are determined by extending the axis of rotation and the equatorial plane of the Earth to the celestial sphere.
The position of an object on the

its
sphere appears to rotate around the Earth once

it is the Earth that is rotating.
right ascension
and
Earth's axis of rotation

to the
tips
is
described b\ specif\ing
is tilted at an angle of 23i from the perpendicular

plane of the Earth's orbit. The seasons are caused by this lilt, which
oi the other toward the .Sun at certain times of the
one hemisphere
sear.
celestial sjihcre
declination.
Discui'niiii; llw
lii'~
Equinoxes and solstices are significant points along the Earth's orbit, determined by the relationship between the Sun's path on the celestial
sphere (the ecliptic) and the celestial equator.
The Earth's axis of rotation moves slowly in a conical fashion, in a phenomenon called precession. Precession is caused bv the gravitational pull
of the Sun and Moon on the Earth's equatorial bulge.
The
phases of the
Moon
are caused b\ the relati\e positions of the Earth,
Moon, and Sun.
The Moon
rhe
Moon
one orbit arotnid the Earth

month averaging 27.15 days.
coinpletes
in a sidereal
completes one cvcle
of
phases
in a s\
with respect to the stars
nodic
month averaging
29.3 days.
Ancient astronomers made great progress in determining the

relative distances of the Earth. Moon, and Sun.
sizes
and
when the Sun and Moon arc both on the line of

at full moon. A solar eclipse occurs when the Sun and Moon are
both on the line of nodes at new moon. The line of nodes is the line
where the planes of the Earth's orbit and the Moon's orbit intersect.
lunar eclipse occurs
nodes
Detailed knowledge of the Moon's orbit around the Earth

predict eclipses.
is
needed
to
The
shadow^ of an object has two parts: the luubra, where the light source
completely blocked; and the penumbra, where the light source is only
partially obscured.
is
Depending on the
relative positions of the Sun, Moon, and Earth, liuiar

be penumbral, partial, or total, and solar eclipses may be
partial, annular, or total.
eclipses
Review questions
* 2
win
is
may
the ancient idea of the celestial sphere
At what location on Earth
Where do you have
to
is
be on the Earth
nith? If you stay at that location for a

Sun pass through the zenith?
*
still
a useful concept today?
the north celestial pole on the horizon?
fiifl
in
order to see the Siui at the zeon how man\ da\s will the
vear,
4 Where do you have to be on Earth in order to see the south celestial pole
overhead? What is the maxiinum possible elevation of the Sun above
the horizon at that location? On what dale is this maximum ele\ation obdirectiv
served ?
* 5
At what point on the horizon does the vernal equinox
rise?
6 Sketch a diagram of the relative positions of the Earth, Moon, Sun, and
Voyager 1 when the photograph in Figure 2-fI was taken.
7 Which type of eclipse, lunar or solar, do vou think most people have
seen? Why?
8
Is
it
possible for a total eclipse ot the

lunar eclipse? Why?
Sun
to
be folltjwed three months
later In a
9 Can one ever observe an annular eclipse of the
Moon? Why?
DiMuvi'iiiii; ihr
Advanced questions
heavens
10 Consult a star map of the Southern Hemisphere and determine which, ii'
any, bright southern stars could some day become south celestial "pole" stars.
11 Using a star map determine which, if any, bright stars could some day
the location of the vernal equinox. Give the approximate years when
that would happen.
mark
12 During a lunar eclipse, does the Moon enter the Earth's

from the west? Explain why.
shadow from the
east or
13 Examine a list of the 88 constellations. Are there any that obviously date
from modern times? Where are they located? Why do you suppose they do
not have archaic names?
14 Describe the seasons

to
its
if
the Earth's axis of rotation were tilted 0
and 90
orbital plane.
15 Describe the cycle of lunar phases that would be obserx ed if the Moon
the Earth in an orbit perpendicular to the plane of the Earth's
orbit. Is it possible for solar and lunar eclipses to occur under these circumstances?
moved about
For further reading
Allen, R. Star Names: Their Lore and Meaning. 1899.
Dover
reprint, 1963.
Chartrand, M. Skygiiide. Western Publishing, 1982.

Cornell, J. The First Stargazers. Scribners, 1981.
Gallant, R. The Cunstellations: How They Came To Be. Four Winds Press,
1979.
W. The Stargazers Bible. Doubleday. 1980.

Krupp, E. Echoes of the Ancient Skies. Harper & Row. 1983.
Kundu, M. "Observing the Sun during Eclipses." Mercun\ July/Aug.
Kals,
1981, p. 108.
Menzel, D., and Pasachoff,

ed.
Houghton
J.
Field Guide
to the
Stars
and
Planets.
2nd
Mifflin, 1983.
"Solar Eclipse: Nature's Superspectacular." Xalional Geographic,
Aug. 1970.
Sagan, C. "The Shores of the Cosmic Ocean." In Sagan, C. Cosmos. Ran-
dom
to
House, 1980. An
measure the
size
excellent recounting of Eratosthenes' experiment
of the Earth.
Gravitation and the motions of
the planets
Apollo 11 leaving the
Moon
module Eagle returns front

completing the
first successful
mission. This photograph
The lunar
the
Moon
after
manned lunar
was taken from

in which the
command module Columbia
the
aslro-
naiih irliino'd to Earth. All of the orbital

iiianru-i'ns to
and
dock the Eagle
and Columbia
then set course for Earth were based on
Newtonian mechanics and Newton's law of

gravity. These same physical principles are
used by astronomers
to
understand a wide
range of phenomena from the motions of

double stars
galaxy.
to the rotation of the entire
(NASA)
lo explain ihe molions of the planets led astronomers to an understanding of gravitation. Ancient astronomers believed that the heavens rotate
around a stationary Earth. We begin this chapter with a look at the complicated assumptions needed to explain the detailed motions of the planets in
such an Earth-centered view of the universe. Then we discuss Copemicm's
Sun-centered explanation of the universe, which provides simpler general ex-
Altempis
planations.
and
We
learn
how
Kepler's elliptical orbits for the planets improved
simplified the Sun-centered explanation
servations supported the idea that the Earth
move around
the
Sun. Next we
see
how
and
is
hoiv Galileo's telescopic ob-
one of the planets that
all
these descriptive successes led to
New-
of gravity and motion, which provide a compact explanation and

description of the motions of planets, comets, and satellites. Finally, we take a
brief look at the new ideas about space, time, and gy-avitalion introduced in
ton's laws
this
century by Einstein.
(iiuvitdtion
and
the
miitwm of
llir jilaitrls
It is not obvious that the Earth moves around tiie Sun. Indeed, oui- daily
experience strongly suggests that the opposite is true. The dailv rising
and setting of the Sun, Moon, and stars could lead us to believe that the
entire cosmos revolves about the Earth at the center of the universe.
That was just what most obsci\ers did bclie\ c for thousands of \cars.
Ancient astronomers
invented geocentric
cosmology to explain
planetary motions
.Ancient Cireek astronomers were among the first to leave a written record of their attempts to explain just how the universe works. Their universe consisted of the Earth and those objects that can be seen with the
naked eye: the Sun, the Moon, the stars, and five planets (Mercurv,
Venus, Mars, Jupiter, and Saturn). Most Greeks also assumed that the
unixerse revolves about the Earth and thus they developed a geocentric
cosmology, or Earth-centered theorv of the universe.
The planets are quite obvious in the night skv because thev slowlv
shift their positions from night to night with respect to the background
of the "fixed" stars in the ccjnstellations. Indeed, the word planet comes
from a Greek term meaning "wanderer." Furthermore, some of the
planets are very bright objects in the night sky. At its maxinuim brilliancy. Vênus is sixteen times brighter than the brightest star.
Simple observations of the planets' positions against the stars from
night to night soon make it clear that the planets do not revolve uniformly about the Earth in concentric circles. Explaining the motions of
the five known planets was one of the main challenges facing the astronomers of antiquity.
As seen from Earth, the planets vvander primarilv across the twehe
mentioned in Chapter 2, these constellations circle the sky in a continuous band centered on the ecliptic. If \ou
follow a planet as it travels across the zodiac from night to night, vou
find that the planet usuallv moves slowh eastward against the background stars. This is called direct motion. Occasionallv, however, the
planet seems to stop and then back up for several weeks or months. This
occasional westward movement is called retrograde motion. These moconstellations of the zodiac. ,As
tions are
much
slower than the dailv rotation ol the sk\ caused bv the

Both direct and retrograde motion are best detected bv
mapping the position of a planet against the background stars from
night to night over a long period of observation. An example is the path
of Mars from autunm 1977 through spring 1978 shown in Figure 3-1.
The Greeks devised various theories to account for retrograde motion
and the exact loops that the planets trace out against the backgrountl
stars. One of the most successful ideas was proposed bv Hipparchus ,uid
elaborated by the last of the great Greek astnmomers, Ptolemv, who
lived in Alexandria dining the second century An. The basic concepl is
sketched in Figure ,S-2. Each planet is assumed to revolve about a small
circle called an epicycle, which in turn revolves about a larger circle
called a deferent, which is approximately centered on the Earth. .As
viewed from Earth, the epicvcle moves eastward along the deferent, and
both circles lotate in the same direction (counterclockwise in Figure 3-2).
Earth's rotation.
Most of the time the motion of the ])lanet on its epicvcle adds to the
eastward motion of the epicvcle on the deferent. Thus the planet is seen
to be in direct (eastward) motion against the background .stars throughout most of the year. However, when the planet is on the part of its epicycle nearest the Earth,
motion
of the epicycle
its motion along the epicvcle subtracts from the

along the deferent.
he planet thus appears to
1
Gravitiilioti
and
the nwlioi)^ uj the plunets
The path of Mars, 19771978 From fall 1977 Ihruugh spring I97S.
Mars moved across the constellations of
Gemini and Cancer. From mid-December
Figure 3-1
through the end of February,

was retrograde.
Alars's motion
CANIS MINOR
Procyon
its usual eastward movement among the constellaeven seeming to go backward for a few weeks or months. This
concept of epicycles and deferents provides a general explanation of the
slow and then halt

tions,
retrograde loops that the planets execute.

Using the wealth of astronoinical data in the library at Alexandria,
including records of planetary positions covering hundreds of years,
Ptolemy deduced the sizes of the epicycles and deferents and the rates of
rotation needed to produce the recorded paths of the planets. After
years of arduous work, Ptolemy assembled his calculations in thirteen
volumes collectively called the Almagest. The positions and paths of the
Sun, Moon, and planets were described with unprecedented accuracy.
The
Figure 3-2
geocentric explanation of
planetary motion
Each planet
about an epicycle, which
revolves
turn levolves
about a deferent centered approximately on the

Earth. As seen from Earth, the speed of the
planet on the epicycle alternately adds
subtracts
from
deferent, thus
direct
and
to
or
the speed of the epicycle on the
producing alternating periods of
retrograde motion.
Nicolaus Copernicus devised

the first comprehensive
heliocentric
cosmology
Almagest
became the astronoiner's
bible.
For over a thousand years, Ptolemy's cosmology endured as a useful

description of the workings of the heavens. Eventually, however, things
began going awry. Tiny errors and inaccuracies that were unnoticeable
in Ptolemy's day compounded and multiplied over the years, especially
with regard to precession. Fifteenth-century astronomers made some cosmetic adjustments to the Ptolemaic system. However, the system became
less and less satisfactory as more complicated and arbitrary details were
added to keep it consistent with the observed motions of the planets.
Imagine driving on a freeway at high speed. As you pass a slowly movit appears to move backward even though it is traveling in the
same direction as your car. This sort of observation inspired the ancient
Greek astronomer Aristarchus to suggest a more straightforward explanation of retrograde motion
one in which all the planets, including the
Earth, revolve about the Sun. The retrograde motion of Mars occurs, for
example, when the Earth overtakes and passes Mars, as shown in Figure
3-3. In Aristarchus's day, however, the idea of a moving Earth seemed
ing car,
incompatible with explanations of other phenomena at the Earth's surface. Almost 2000 years elapsed before someone had the insight and deterinination to work out the details of a heliocentric (Sun-centered) cosmology. That person was a sixteenth-century Polish lawyer, physician.
uravitaliuii
Figure 3-3
Sun more
ifuenlly, as the
the
motwm
of the planeis
heliocentric explanation of
planetary motion
the
and
Tin-
ral)iith
Earth
Iravi-is
around
than does Mars. Cutue-
Earth overtakes and passes thh
slower-moving planet. Mars appears
to
move
backward for a few months.
economist, monk, and artist named Nicolaus Copernicus. Especially

gifted in mathematics. Copernicus turned his attention to astronomy.
Copernicus realized that, with a heliocentric perspective, he could deter-
mine which planets are closer to the Sun than is the Earth and which
are farther away. Because Mercury and Venus are always observed fairlv
near the Sun. Copernicus concluded that their orbits must be smaller
than the Earth's. The other visible planets

Mars, Jupiter, and Saturn
can be seen in the middle of the night, when the Sun is far below the
horizon, which can occur only if the Earth comes between the Sun and a
planet. Copernicus concluded that the orbits of Mars. Jupiter, and Saturn are larger than the Earth's orbit.
It is
often useful to specify various points on a planet's orbit, as in
Figure 3-4. These points help us identify certain geometrical arrange-
Quadrature q'
Figure 3-4
Planetary configurations
It is
often usejul to specify key points along a planet's
orbit
cLs
shown
in this
diagram. These
points identify specific geometric arrangements
between the Earth, a planet,
and
the Sun.
Quadrature
GmvUatwn and
llie
moliims / Ihr
pitiiii'ls
ments or configurations between the Earth, another planet, and the

Sun. For example, when Mercury or Venus is between the Earth and the
Sun. we say the planet is at inferior conjunction. When the\ are on the
opposite side of the Sun, thev are at superior conjunction.
The angle between the Sun and a phuiet as \iewed honi the Earth is
called the planet's elongation. At maximum eastern elongation, Mercury
or Venus is as far east (A the Sini as it can be. ,At siitli limes, the planet
appears above the western hori/on alter siuiset and is often called an
"evening star." Similarly, at maximum western elongation, Mercurv or
Venus is as far west of the Sun as it can possibly be and rises before the
Sun, gracing the eastern predawn sky as a "morning star."
When one of the outer planets is behind the Sun, we sav that the
planet is at conjunction. When it is exactly opposite the Sun in the sky,
it is at opposition. It is not difficult to determine when a planet happens
to be located at one of the key positions in Figure 3-4. For example,
when Mars is at opposition, it appears high in the sky at midnight.
.'Vlthough it is easy to follow a planet as it moves from one configuration to another, these observations in themselves do not immediately
provide relevant data about the planet's actual orbit around the Sun.
Copernicus realized that the Earth, from which we make the observations, is also moving. He was therefore careful to distinguish between
two characteristic time intervals, or periods, of each planet. The synodic
period is the time that elapses between two successive identical configurations (as seen from the Earth)
from one opposition to the next, for
example, or from one conjunction to the next. The sidereal period is
the true orbital period of a planet, the time it takes the planet to complete one orbit of the Sun.
The synodic period of a planet can be determined bv observing the
skv. but the sidereal period must be calculated. Copernicus figined out
how to do this and obtained the results shown in Table 3-1.
Knowing the sidereal periods of the planets. Copernicus was able to
devise a straightforward geometric method of determining the distances
of the planets from the Sun. His answers turned out to be remarkably
close to the modern values, as shown in Table 3-2.
From these two tables it is apparent that, the farther a planet is from
the Sun, the longer it takes to travel around its orbit.
Copernicus compiled his ideas and calculations into a book entitled Dc
Revolutionibus Orbium Cek'slium (On Ike Revolutions of the Celestial Spheres)
that was published in 1543, just in time to reach Copernicus on his
deathbed. Although he assumed that the Earth travels around the Sun
along a circular path, he foiuid that perfectlv circular orbits cannot accurately describe the paths of the other planets. Copernicus had to add an
epicvcle to each planet to account for the slight variation in speed along
TABLE
3-1
The synodic and sidereal
periods of the planets
Sidereal period
Planet
Synodic period
Mercury
116 days
88
Venus
584 days
225
Earth
Mars
780 days
1.9 years
Jupiter
399 days
11.9 years
Saturn
378 days
29.5 years
days
days
.0
year
Grmtlation a>id the motions of the planets
TABLE
3-2
planets
from
Averagie distances of the

the
Sun
tin
astronomical
units)
Planet
Copernicus
Modem
Merciin
0.3S
0.39
\enus
0.72
0.72
00
1.00
Earth
1.
Mars
1.52
1.52
Jupiter
5.22
5.20
Saturn
9.07
9.54
its
orbit.
Thus, according to Copernicus, each planet revolves around a
Sun ;ilong a circular path.

Sun at the center of the universe was such a revolutionar\one Renaissance astronomer. T\cho Brahe. tried to test
small epicvcle. which in turn orbits the

Placing the
proposal that
Copemicuss ideas uith detailed observations of the skv. Brahe spent his
lifetime making accurate observations of the positions of the stars and
planets. achie\ing
We know
that,
an imprecedented
level
when we walk from one
of precision.
place to another, nearbv
ot)-
background of more dbtant objects. Brahe argjued that, if Copernicus was correct, nearbv stars
should shift slightlv against the background stars as the Earth orbits the
Sun. In spite of Brahe's accurate observations, he could not detect anv
shifting of star positions. He therefore concluded that Copernicus was
wrong.
.\ctuallv. the stars are so far awav that naked-eve observations could
jects
appear
to shift their p>ositions against the
not possiblv detect the tinv shifting of star positions that has
now
b>een
confirmed with telescopic observations. Nevertheless. Tvcho Brahe's astronomical records were destined to plav an imf>onant role in the development of a heliocentric cosmologv. Upon his death in 1601. manv of
his charts and books fell into the hands of his gifted assistant. Johannes
Kepler.
Johannes Kepler proposed
Major
Minor
ftgurr 5-5
An
ellipse
axis
The emutrmctiam af am tUipst
can be droa-n
j.ith
a pencil, a loop
f string, and tiro thumbtacks, as shcam in

this diagram. If the stnng is kept taut, the
pencil traces out
an
ellipse.
Johannes Kepler's time at the beginning of the seventeenth cenastronomers had assumed that heavenlv objects move in circles.
Circles were considered the most perfect and harmonious of all geometric shapes. Since CkxI is perfect, he would use onh circles to control the
motions of the planets. Kepler doubted such arguments. His first major
contribution to astronomv was the suggestion that noncircular curves
might fit planetarv orbits.
Kepler turned from circles to ovals, but he was unable to fit ovab to
the orbits of planets about the Sun. Then he began working with a
slightlv different curve called an ellipse.
.\n ellipse can be drawn with a loop of siring, two thumbtacks, and a
fjencil. as shown in Figure 3-5. Each thumbtack is at a focus (plural.
foci). The longest diameter across an ellipse, called the major axis,
passes through both foci. Half of that distance is called the semimajor
axis, whose length is usuallv designated bv the letter a.
To Kepler's delight, the ellipse turned out to be the curve he had
been searching for. He published this discoverv in 1609 in a book known
todav as S'ru- .\<tronom\. This important discoverv is now called Kepler's
first law and is stated as follows:
Until
paths of the planets

about the Sun
elliptical
turv.
The too thumb-
tacks are located at the tiro fact of
The
the ellipse.
focus.
orbit of a planet about the
Sun
is
an ellipse with the Sun
at
one
Gravitation
Figure 3-6
laws
Kepler's first
According
every planet travels

elliptical orbit
and second
to Kepler's first
around
with the
Sun
the
at
two laws,
Sun along an
one focus in
such a way that the line joining the planet and

the
Sun sweeps
and
the motions uf the planets
Kepler also realized that planets do not move at a uniform speed

along their orbits. A planet moves most rapidly when it is nearest the
Sun, at a point on its orbit called the perihelion. A planet moves most
slowly when it is farthest from the Sun, at a point called the aphelion.
After much trial and error, Kepler discovered a way to describe how
fast a planet moves along its orbit. This discovery, now called the law of
equal areas or Kepler's second law, is illustrated in Figure 3-6. Suppose
that it takes 30 days for a planet to go from point A to point B. During
that time, the line joining the Sim and the planet sweeps out a nearly
triangular area. Kepler discovered that the line joining the Sun and the
planet sweeps out an equal area during any other 30-day interval. In
other words, if the planet also takes a month to go from point C to
point D. then the two shaded segments in Figure 3-6 are equal in area.
Kepler's second law, also published in New Astronomy, can thus be stated
as follows:
out equal areas in equal inter-
vals of time.
line joining a planet
and the Sun sweeps out equal areas
in equal
intervals of time.
One of Kepler's later discoveries, published in 1619, stands out because of its impact on future developments. Now called the harmonic
law or Kepler's third law, it states a relationship between the sidereal
period of a planet and the length of its semimajor axis:
The squares of
the sidereal periods of the planets are proportional to
the cubes of their semimajor axes.

If a planet's sidereal
its
semimajor
third law
is
axis (a)
period {P)
is
is
measured
in years and the length of

astronomical units, then Kepler's
measured
in
simply stated as
P- = a'
The length of the semimajor axis is actually the average distance between a planet and the Sun. Using data from Tables 3-1 and 3-2, we can
demonstrate Kepler's third law as shown in Table 3-3. This relationship
can also be displayed on a graph as in Figure 3-7.
testimony to Kepler's genius that his three laws are rigorously
in any situation where two objects orbit each other under the influence of their mutual gravitational attraction. Kepler's laws are obeyed
not only by planets circling the Sim, but also by artificial satellites orbiting the Earth and by two stars revolving about each other in a double
It is
obeyed
star system.
Throughout
this
we
book,
shall see that Kepler's laws
have a
wide range of practical applications.
TABLE
3-3
demonstration of Kepler's
third law
Planet
Sidereal period
Semimajor axis
a (in
(in years)
AU)
P^
a^
Mercury
0.24
0.39
0.06
0.06
Venus
0.61
0.72
0.37
0.37
Earth
1.00
1.00
1.00
1.00
Mars
1.88
1.52
3.53
3.51
Jupiter
11.86
5.20
140.7
140.6
Saturn
29.46
9.54
867.9
868.3
Grai'ilufioti
Figure 5-7 Kepler's third law On this

squans nj thi- pnwd\ iij the planets
grafih. the
and
tht motiim^ of the planets
I'lui.M
5(1,
000
(P') are plotted aiâinst the cubes uf their sem-
Ncplunr
imajor axes la'). The fact that the points fall
along a straight
discoi'ery that
line
is
P~ = a
venjication oj Kepler's
Ininus
y.
L_^
Square olorbilal period
Galileo's discoveries with a

telescope strongly supported
a heliocentric
cosmology
(yr")
While Kepler was making ra]jid progress in central Europe, an Italian

physicist was making equally dramatic obseryational discoyeries. Galileo
Cialilei did not inyent the telescope, but he was the first person to point
one of the new devices toward the sky and publish his obseryations. He
saw things that no one had eyer dreamed of. He saw mountains on the
Moon and sunspots on the Sun. He also discovered that Venus exhibits
phases (see Figure 3-8).
After only a few months of observation, Galileo noticed that the apparent size of Venus as seen through his telescope was related to the
planet's phase. The planet appears smallest at gibbous phase and largest
at crescent phase. There is also a correlation between the phases of
\'enus and the planet's angular distance from the Sun. These relationships clearly support the conclusion that Venus goes around the Sun (see
figure 3-9).
In IfilO, Galileo also discovered four
!')-]()
shows a portion
alized that these four
moons
orbiting Jupiter. (Figure
of his astronomical records.)
moons obey
Astronomers soon
re-
Kepler's third law: the cube of a
moon's distance from Jupiter
Figure 3-8
F.arth.
Crescent Venus
Venus
As seen froni
exhdiits phases that are corre-
lated with the planet's
angular
size.
The plant
appears largest at crescent phase
(maximum
angular diameter: 65.2 arc
as
here.
sec),
shown
Venus appears considerably smaller at
gibbous ptiase
9.5 arc
sec).
(minimum angular duimeter:
(Palomar Obserralory)
is proportional to the square of its orbital

period about the planet.
fhese telescopic observations constituted the first fresh influx of fundamentally new astronomical data in almost 2()()() years. In contradiction
to prevailing opinions, these discoveiies strongly suggested a heliocentric
view of the imiverse. I he Roman (Catholic church attacked his ideas because they were not reconcilable with certain passages in the Bible or
and Plat(}. Nevertheless, there was no turning back, (iaiileo was condennicd to spend his latter years under house
arrest "for vehement suspicion of heresy," but his revolutionary ideas
with the writings of Aristotle
soon inspired a sickly F.nglish boy born on Christmas day of 1642,

than a year after Galileo died. The boy's name was Isaac Newton.
less
Gravilalion
The
plutsei oj
Willi llie planet's
distance
miiliinn n/
llie
llic jilaiiils
The changing appearance of
Figure 3-9
Venus
and
from
the
Venus are coiielaled
and
its
angular
Sun, as sketched in
this diu-
angular
size
grain. These ohsen'ation,': clearly support the
idea that Venus orbits the Sun.
Figure 3-10
Jupiter's
In IblU. Galileo discover
four "stars" that move back and forth across

Jupiter from one night
to the next.
He
5o.w^
con-
Moon
orbits the
*o
O*"
a.-^.
Earth.
3-'-Ty^
(Yerkes Obsen'aton)
He.
"Ml.'
p.*.^^//., 7">
v.-^m'
>
11
14
Newton formulated
accounts for Kepler's laws

and explains the motions of
the planets
^-t
'
O
o
)-
-f- "Kr*-'
. '^:
'
-O
*-
-.o
.
^"1^^'
)^
3 /VV^'
'
*
*
Jf
it-
j(
4-o-nc''
^5"^:
O
Q ,
'>-r.
l7-'
Isaac
^4-^.'
>-'
* ^
l-u.W.û^;
description of gravity that
,^.
*o
*o
*'0
J.^
li-r'^H^
17.
cluded that these are four moons that orbit

Jupiter much as our
\r.
Galileo's observations of
moons
Until the mid-seventeenth century, virtually all mathematical astronomy

was entirely empirical, characterized by trial and error. From Ptolemv to
Kepler, essentially the same approach was used. Astronomers would
work
directl)'
from data and observations, adjusting ideas and
calcula-
came out with the right answers.

Isaac Newton introduced a new approach. He made three assumpUons, now called Newton's laws of motion, about the nature of reality.
Newton showed that Kepler's three laws follow logically from his own
laws of motion and a formula for the force of gravity that he derived.
Using this formula. Newton accurately described the observed orbits of
the Moon, comets, and other objects in the solar system.
tions until they finally
Newton's
first
law states that
body remains at rest, or moves in a straight

upon by an outside force.
line at a constant speed,
unless acted
there was no "outside" force acting on planets, they

and move away from the Sun along
straight-line paths at constant speeds. Since this does not happen, NewIn other words,
would leave
if
their curved orbits
Gravilalwn antl the motiom of the planets
ton concluded that the continuous action of this force confines the planets to their elliptical orbits.
Isaac Newton did not invent the idea of gravity. An educated seventeenth-century person had a vague appreciation of the fact that some
force pulls things down to the ground. What Newton did was to give us
a precise description of the action of gravity.
Using his first law. Newton

mathematicalh proved that the force acting on each of the planets is
aimed directly at the Sim. This discoverv led him to suspect that the
force pulling a falling apple straight down to the ground is the same as
the force on the planets that is always aimed at the Sun.
Newton's second assumption, called Newton's second law, describes
what a force does. Imagine an object floating in space. If you push on
Figure 3-11
modem
people
IS
Major contributors
gravitational theory
to
Each of
these
responsible Joi a major contribution
or breakthrough in our understanding of
and inwe now send spacecraft to the planets

and probe the geometry of the universe.
gravity. Because of their discoveries

sights,
(Yerkes Observatory)
Copernicus
(1473-1543)
Galileo
(1,564-1642)
Kepler
(1571-1630)

Gravilalioii intd the motions oj
the object,
speed
will
eration
is
tlir {jlfini'l^
it will begin to move. If you continue to push, the object's

continue to increase in other words, it will accelerate. Accelthe rate at which a velocity changes. Newton's second law says
that
The
acceleration of an object
is
proportional to the force acting on the
object.
Newton's
action
and
final
assumption, his third law,
is
the famous statement of
reaction:
Whenever one body

exerts an equal
exerts a force on a second body, the second body

and opposite force on the first body.
Using his three laws and Kepler's three laws, Newton succeeded in
formulating a general statement describing the nature of the force
gra\it\'
that keeps the planets in their orbits. Newton's universal law of
gravitation
is
stated as follows:
Two
bodies attract each other with a force that is directly proportional

product of their masses and inversely proportional to the square
of the distance between them.
to the
In other words, if two objects have masses m, and Wa and are separated
by a distance r, then the gravitatonal force F between these two masses is
F=G-
Figure 3-12
tion
fry
is
am
Conic sections
one oj
ii
jamih
dicing a cone unth a
this
conic iec-
cunvs obtained
plane, as shown in
oj
diagram. The orbit of one body about an-
other can be any one of these curves: a circle,
an
ellipse,
a parabola, or a hyperbola.
Ellip.se
In this formula, G is a number called the unwcnal conslant of gravilalioii

which has been determined from laiioratory experiiuents.
Using his law of gravity, Newton found that he could mathematically
prove the validity ot Kepler's three laws. For example, Kepler discovered
by trial and error that P- = a'\ Newton demonstrated mathematically
that this equation follows logically from his law of gravity anytime that
two objects revolve about each other.
Newton also discovered new features of orbits around the Sun. For
example, his equations soon led him to conclude that the orbit of an object around the Sim could be anv one of a family of curves called conic
Hypcrbol
Gravitatwu antl ihc
infitinn\ oj Ihr
planets
conic section is any curve that you get by cutting a cone with
shown in Figure 3-12. You can get circles and ellipses by
slicing all the wav througli the cone. You can also get two "open" curves
called parabolas and hyperbolas. Comets hurtling toward the Sun from
the depths of space sometimes follow parabolic orbits.
Newton's ideas and methods turned out to be incredibly successful in
a wide range of situations. The orbits of the planets and their satellites
could now be calculated with unprecedented precision. In addition, Newton's laws and mathematical techniques were capable of predicting new
phenomena. For example, one of Newton's friends, Edmund Halley, was
intrigued by historical records of a comet that was sighted about every
76 years. Using Newton's methods, Halley worked out the details of the
sections.
a plane, as
Figure 3-13
Halley's comet
HttUn's roiinl
Sun with an average period uf about

During the twentieth cetitury. the
once in
comet passed near the Sun ttvict
1 910 and again in 1986. This photograph
shows how the comet looked in 19N6.
orbiis the
76
years.
(Lumicon)
Figure 3-14
Neptune and Uranus
I lie
existence of
Neptune (with one moon) was
duced from
dei'iations in the predicted orbit
Uranus
de-
a major triumph for Newtonian mechanics.

Both planets are shown along with some oj
llmr
iij
(with three moons). This discoven teas
satellites.
(Lick
Ohsermton)
comet's orbit and predicted its return in 1758. It was first sighted on
Christmas night of 1757, and to this day the comet bears Halley's name.
Perhaps the most dramatic success of Newton's ideas involved the discovery of the eighth planet from the Sun. The seventh planet. Uranus,
was discovered accidentally by William Herschel in 1781 during a teleit was clear that Uranus was not following its predicted orbit. Two mathematicians, John
C:ouch Adams in England and U.J. Leverrier in France, independently
calculated that the deviations of Uranus from its orbit could be explained by the gravitational pull of a yet unknown, more distant planet.
They each predicted that the planet would be found at a certain location
in the constellation of Aquarius. A brief telescopic search on September
23, 1846, re\ealed Neptune less than 1 from the calculated position.
Although sighted with a telescope, Neptune was really discovered with
scopic survev of the sky. Fifty years later, however,
pencil
and paper.
GravilulKiii
ami
llii-
moliiiin
tif
l/ii-
phinets
Over the years, Newton's ideas were used in successful predictions

and explanations of many phenomena, and thus Newtonian mechanics
became the cornerstone of modern physical science. Even today, as we
send astronauts to the Moon and spacecraft to the outer planets, Newton's equations are used to calculate orbits and trajectories. In addition,
modern astronomers frequently use Newtonian mechanics in their studies of stars and galaxies. Indeed, basic information about the masses of
stars and galaxies comes directly from applications of Newton's laws.
There was, however, one instance in which Newtonian mechanics was
not quite in agreement with observations. During the mid- 1800s, Leverricr pointed out that Mercury was not following its predicted orbit. As
the planet moves along its elliptical orbit, the orbit itself rotates (or
precesses) as shown in Figure 3-15. Specificalh Mercury's perihelion
shifts by an amoimt that cannot be explained with Newtonian mechanics.
Although the effect is very small (the unexplained excess rotation of
Mercury's major axis is only 43 arc sec per century), all attempts to account for this phenomenon met with failure. Nevertheless, Newton's laws
proved so eminently successful in all other situations that most physicists
and astronomers simply ignored this tinv discrepancy.
It is a testament to Isaac Newton's genius that his three laws were
precisely the three basic ideas needed for a full understanding of the
motions of the planets. In this wav, Newton brought a new dimension of
elegance and sophistication to our understanding of the workings of the
universe. That is where things probably would have remained had it not
been for the insight, \ision, and genius of Albert Einstein.
.
Figure 3-15
perihelion
The advance of Mercury's

As Alercuiy moves along lis orbil,
Ihe orhil lulales. Coyisequenlly,
Mereuiy
Iraces
Old a loselle figure about Ihe Sun. greatly ex-
aggerated in
thus
diagram. The excess rotation
of Mercury's orbil (beyond that predicted by
Newtonian mechanics) umuiinis

sec
Id
only
43 arc
per centuiy.
Albert Einstein's theory

states that gravity affects the
shape of space and the flow
In the cosmologies of most ancient civilizations, the Earth and its inhabitants occupied a special place at the center of the uni\erse. Copernicuss
new cosmology made the Earth just one of a number of planets orbiting
of time
the Sun. Within less than a centurw Kepler's accurate description of

planetary orbits led directly to Newton's universal law of gravitation. As-
tronomers and philosophers then began to explore the possibility that

even the Sun might occupy no special location in the universe that our
Sun could be merely one of many stars scattered through space.
Albert Einstein introduced an even more powerful perspective. He
believed that the fundamental laws of the universe depend neither on
our location in space nor on our motion through space. In other words,
the laws of physics should be the same, whether we happen to be sitting
on the Earth or moving through space at 95 percent of the speed of
light.
Einstein began his

for the
work soon
phenomena of
electricity
after 1900 with a re\ised explanation
and magnetism. The basic properties of

and magnetism had been sununarized in four equations forelectricity
mulated in 1865 by the great Scottish ph\sicist James Clerk Maxwell.

These equations are the basis of electromagnetic theory, which today
has a wide range of practical applications from television sets to microwave ovens. Maxwell's electromagnetic theory predicts differing effects
depending upon the motion of electric charges and magnets. For example, a moving electric charge creates a magnetic "field," whereas a stationary electric charge does not. Such predictions imply some absolute or
fixed frainework of space within which an object is either stationary or
moving. Einstein's goal was to eliminate this assumption of absolute
space from electromagnetic theor\.
He succeeded
in
1903 and published
iiravitattoii (iiul
motinus nf
llif
iht'
phtni'ls
famous paper
his results in a
"On
entitled
the Electrod\ iianiics of
Mov-
ing Bodies."
work with electromagnetic theory introduced some revolu-
Einstein's
tionarv concepts to ph\sics. Einstein's calculations led to surprising conclusions: the length of an object depends upon its speed relati\e to the
observer
who measures
depends upon
its
it.
and the
speed relative
rale at
to the
which a clock
obser\er
who
licks similarly
times
it.
Ehe bod\
of knowleclge that describes all these effects is called the special theory
of relativity.
For most phenomena, Newtonian mechanics provides accurate predictions and explanations. The relati\ity of distances and time intervals is
easilv noticeable only at speeds close to the speed of light. For extremely
high speeds (such as those of subatomic particles in particle accelerators),
phvsicists found that special relativity is needed to explain and understand the results of experiments. Using very accurate "atomic clocks,"
phvsicists have even confirmed that a clock in a moving jet plane seems
to an Earth-based obser\er to tick slightly more slowlv than an identical
clock mounted on the Earth. Einstein's theorv has been supported by
everv experiment designed to test it.
Details of special relalivitv need not concern us here, but the newly
discovered relativitv of space and time was \ery important in Einstein's
quest to understand gravitv. In 1916, Albert Einstein formulated a new
theorv of gravity called the general theory of relativity. The basic idea is
that the mass of an object alters the properties ol space and time around
the object. According to the general theory of ielativit\, gra\it\ causes
space to become curved and time to slow down.
These concepts are even more difficult to understand than the concepts of special relativity. One useful analogy is to imagine that space
near a massive object (such as the Sun) becomes cur\ed like the surface
in Figure 3-16. Imagine a steel ball rolled along this surface. Far from
the "well" that represents the Sun,
the
flat
surface. If
it
it
would mo\e
along
curve toward
in a stiaight line
passes near the well, however,
it
will
moving at an appropriate speed with respect to the well,

an orbit around the sides of the well. The curvature of
the surface makes the ball follow a curved path toward the well.
It is hard to imagine the three-dimensional curvature of space, but its
effects are much the same as in the two-dimensional analogy in Figure
3-16. Far from the Sun, planets and comets should travel along nearly
straight-line paths. Near the Sun, planets and comets travel along curved
paths because space itself is cur\ed. The general theory of relativity predicts that onlv verv strong gra\itational fields have easily detectable effects upon space and time in their \icinit\-.
One of the first things tjnstcin did with his new theorv was to calculate the orbits of the planets. With only one minor exception, general
relativitv theorv ga\e almost exactly the same answers as Newtonian thethe well. If
it
it
may move
ory.
Mercurv
is
in
is
the only planet to pass so
clo.se to
the Sun's strong gra\
i-
produces an effect significantK

tlilTerent from that predicted by Newtonian mechanics. Indeed, througiiout most of the solar system the curvature of space is so slight that Einlational field that the curvature of space
equations are essentially equivalent to Newtonian calculations.

Einstein's theorv did predict that Mercury's orbit should be a
precessing ellipse, therebv explaining a phenomenon that had frustrated
astronomers for half a centur\
Fo test the validitv of his theorv Einstein predicted new phenomena.
He calculated that light ra\s passing near the surface of the Sun should
ajjpear to be dellecied Ironi iheir siraighi-line paths because the space
stein's
Figure 3-16
of space
(iry
The gravitational curvature
Arroiilini; Id Enislfin'\ neiieral Ihe-
of irhlivily. space hpcomes curwil
iieiir
massive object such as the Sun. This diagram
shows a two-dimensiimal
of space
atialofn/ of Ihe
around a massive
iihjecl.
\hape
However,
GrcivUatmn and
the miliums
iij
Ihe plani'ls
through which they are moving is curved. In other words, gravity should
bend light rays, an effect not predicted by Newtonian mechanics because
light has no mass.
Figure 3-17 shows a beam of light from a star passing by the Sun and
continuing down to the Earth. Because the light ray is bent, the star
appears shifted from its usual location. At most, the position of a star is
shifted by 1.75 arc sec for a light ray grazing the Sun's surface.
The gravitational
Figure 5-/7
deflected
ativistic
from
standards, the Sun's gravity
maximum
75 arc
for a
Sun
By
their straight-line paths.
weak. The
1
deflection
Light rayi passing near Ike
of light
sec
is
rel-
very
angle of deflection
light ray
are
is
only
grazing the Sun 's
surface.
This prediction was first tested during a total solar eclipse in 1919.
During the precious moments of totality, when the Moon blocked out
the blinding solar disk, astronomers succeeded in photographing the
stars around the Sun. Careful measurements revealed that the stars were
shifted from their usual positions by an amount consistent with Einstein's
theory. General relativity had passed another important test.
Einstein made a third prediction. Because gravity causes time to slow
down, he stated that clocks on the first floor of a building would tick
slightly more slowly than clocks in the attic that are farther from the
Earth, as sketched in Figure 3-18. This prediction was tested by analyzing light from very compact stars (whose surface gravity is very strong)
and by using extremely accurate clocks first developed in 1960. Again
Einstein was proven correct.
During the past several decades, these and similar tests have confirmed general relativity over and over again. This remarkable theory
now stands as our most precise and complete description of gravity.
It is important to emphasize that Einstein did not prove Newton
wrong. Rather, Einstein demonstrated that Newtonian mechanics is accurate only when applied to low speeds and weak gravity. If extremely
high speeds or powerful gravitational fields (such as those of neutron
stars and black holes) are involved, only a relativistic calculation will give
correct answers.
evoked a sensation when it was

proposed. Newtonian gravitation had been overthrown by a radi-
Einstein's general theory of relativity
J^l
first
approach that worked better.

excitement, however, interest in general relativity rapidly waned. The complex mathematics of relativity seemed burdensome,
and Newton's simpler approach gave reasonable precision in almost all
conceivable circumstances. No one could imagine places where the curvature of space might have a major effect. Relativistic theory offered only
cally different
After the
little
more
initial
accuracy, as in the case of Mercury's orbit.
In recent years there has been a reawakening of interest in general
Figure 3-18
The gravitational slowing of
time
ing
closer to the
is
ground floor of a buildEarth than a clock at a
clock on the
higher elevation. Therefore, according

eral relativity, the clock on the
should
tick
the roof.
more slowly than
to
gen-
ground floor
the clock on
relativity, primarily because of dramatic advances in our understanding

of the evolution of stars. As we shall see in later chapters, we now have
a reasonably complete picture of how stars are born, what happens to
them as they mature, and where they go when they die. In particular, it
appears that the most massive dying stars are doomed to collapse completely upon themselves, thereby producing some of the most bizarre
objects in the universe, black holes. Gravity around one of these massive
stellar corpses is so strong that it punches a hole in the fabric of space.
Because of this intense gravitational effect, black holes can be described
and discussed only in terms of general relativity.
As we set our sights on understanding the universe as a whole, we
must turn to general relativity. All the matter in all the stars and galaxies
is responsible for the overall curvature or shape of space. From the viewpoint of general relativity, it is therefore reasonable to ask about the actual shape of the universe. We shall see that the answer suggests the ultimate fate of the cosmos.
C,)(ivilalinii mill
Summary
till-
mnhiiii\ /
llir
liliniih
Ancient astronomers believed that tlic Earth is at the center of the universe and invented a complex system of epicycles and dctcrcnts to explain
direct and retrograde motions of the planets.
heliocentric (Sun-centered) theory simplifies the general explanation oi

planetary motions. In a heliocentric system, the Earth is one o( the plan-
ets.
The
sidereal period of a planet (measured with respect to the stars) is

true orbital period; its synodic period is measured with respect to the
Earth and Sun (for example, from one opposition to the next).
its
around the Sun much more acKepler's three laws give important details about
Ellipses describe the paths of the planets
curately than
do
circles.
elliptical orbits.
The
new
invention of the telescope led to
discoveries that supported a
heliocentric view of the universe.
Newton based his explanation of the universe on three assiuuplions or

laws of motion. Kepler's laws and extremely accurate descriptions of planetary motions can be deduced from Newton's laws and his imiversal law
of gravitation.
In general, the path of one object about another (such as a comet about
the Sun) is one of the curves called conic sections: a circle, an ellipse, a
parabola, or a hyperbola.
Although Newtonian mechanics accurately describes and predicts numerous phenomena, Einstein's relativistic theories are more accurate where
extremely high speeds or intense gravitational fields are involved.
The
special theory of relativity was designed to eliininate the notion of

absolute space and time. The theory leads to the conclusion that measurements of distance and time intervals are affected by the motion of the
observer.
The
general theorv of relativitv explains that gravity causes space to be

to slow down.
curved and time
Both the
special
and general theories have been supported bv the

designed to test hem.
of every ex|)eriment
Review questions
How
results
did ancient astronomers differentiate between a star and a ])lanet?
2 With the aid of a chawing, show how an epicycle can explain the retrograde motion of a planet in a geocentric cosmology. With a second drawing
show how a heliocentric cosmology explains the same retrograde motion.
3 At what configuration (superior conjunction, greatest eastern elongation,
would it be best to observe Mercury or Venus with an Earth-based telescope? At what configtnation would it be best to observe Mars, Jupiter, or
Saturn? Explain \our answers.
etc.)
4 In Noiir
own words,
stale Kepler's three laws ol pl.iiielar\ niolion.
5 In what ways did the aslrononiiial (>hsei\alions ol (iaiileo suppoii a heliocentric cosmology?
How
from
(lid
Nt-wlon's
a[)l)r()a( h to
ihal ol his predecessors?
undeistandini^ plani-lar\ motions diller
Gravtiation aitd the motions oj the f}lanets
What
are conic sections and in what wa\ are thev related lo the orbits of
objects in the solar system
Why
is
"-
Neptune considered
the discovery of
a major confirmation ot
Newtonian mechanics?
9
What new
theories
came from
F.insleins woi k
and what piienomena did
the\ deal with?
Advanced questions
10
Is
it
exactly
*11
possible for an object in the solar system

one year? Explain vour answer.
to
have
a synodic period
of
AU
line joining the Sun and an asteroid is found to sweep out 5.2 square
of space in 1983. How much area is swept out in 1984? In five vears?
*12 .A comet moves in a highly elongated orbit about the Sun with a period of
1000 years. What is the comet's average distance from the Sunr What is the
farthest it can get from the Sun?
*13
Is
it
Would
possible for a planet's sidereal period to equal

planet be closer to the Sun than the Earth
this
there a planet that nearly
fits this
its
is
synodic period?
or farther awav?
Is
description?
14 Look up the dates of various greatest eastern and western elongations for
Mercury in a year of your choice. Does it take longer to go from eastern to
western elongation or vice versa? Why do you suppose this is the case?
15 Which planet would vou expect to exhibit the greatest variation in apparent brightness as seen from Earth? Explain your answer.
16 Use two thumb tacks, a loop of string, and a pencil to draw several ellipDescribe how the shape of the ellipse varies as you change the distance
ses.
For further reading
between the thumb
tacks.
Banville. J. Kepler:
Novel. Godine, 1981.
Christianson, G. This Wild
Ab'\ss.
Free Press, 1978.
Cohen. O. "Newton's Discovery of Gravit\
."
Scientific
American.
March
1981.
Durham.
F.,
and Purrington.
sity Press,
R.
Frame
iif
the Universe.
Columlîa Univer-
1983.
Gardner. M. The Relativity Explosion. Vintage, 1976.

Gingerich. O. "Copernicus and Tvcho." Scientific .-{merican. Dec. 1973.
Hoffman. B.. and Dukas. H. .-Mhert Einstein: Creator and Rebel. \.\L
Plume Books, 1972.
Kaufmann, W. Relativity and Cosmolog\. Harper &: Row, 1977.
Kuhn, T. The Copernican Revolution. Harvard Uni\ersit\ Press. 1957.
Rogers. E. Astronomy for
the Inqiiirinir .Mind.
Princeton Universitv' Press,
1960.
Wilson. C. "How Did Kepler Discover His First 1

American. March 1972.
wo Laws?"
Scientific
4 Light, optics, and telescopes
Cerro Tololo Inter-American Obsematory

A^ldiiiiiiii'is
liili'd
the
prfjn
iiKiiiiilaiiilolii
an
i\
fay frum
iln. stable,
and
cil\ liglili.
Ohseivalon
(250
hiler-Ameri-
in Chile, about
miles) north oj Santiago.
houses Ihe 4-meter
CTIO
iio-
where
chiidfree. This aer-
ial I'leie choirs Ihe C.erro Tololo

caii
on
Id Iniild nh.sfii'iilorir.s
400 km
The large dome
telescope
shown
in
Figure 4-12. Sei'eral smaller telescopes also

share
(7200
this
is 2200
(NOAOl
muuntaintop. which
feet)
above sea
level.
meters
Lhitil very rcrt'ttlh.
our
ktiowlcdi:;!' (thoitt the luiivcrsc
frotn the visible light gatheied
li\
telescopes.
In
this
came
ehapler,
ctlitiosl
we
niliich
leant that
a form of eleclromagnette radialion and thai many other forms of such

exist. We di.scu.ss the two major types of optical telescopes: refractors
and reflectors. Then we learn aliont more exotic telescopes that detect radio
waves and other electromagnetic radialion arriving at the Earth, f'tnally. zee
examine the orbiting ohservaUnies that are pnidiicing a -wide range of neie
light
is
radiation
information about the universe.
Tclcscopt-s lia\e pla\c<l a major role in ri\calin;j;
ilic
uni\ersc since
Clali-
saw craters on the Moon four centuries ago. Tlie telescope lemains tfie single most important tool of astronomy. With a telescope, wc
can see cxtremeK iaint ohjeds in space lai more clearK than we can
with the naked e\c.
'I\aclitionall\. telescopes detect visible light. Kither lenses or mirrors
bring light from a distant object to a focus where the resulting image is
viewed oi photogiaphed. RecentU. lu)we\er. astronomers ha\e built tele.scopes that detect iiomisible torms of light such as X ia\s and ladio
waves. To ajjpieciate these developments and to understaiid how telescopes work, we must fust leaiii something about the basic pi()])erties
leo first
ol light.
Lighl. ufllKs.
Light
is
electromagnetic
and is characterized
radiation
by
its
wavelength
and
/<7(sr(y/;cs
and Newton made important contributions to our modern understanding of light as well as to our theories of gravity and mechanics. In
the early 1600s, Galileo made one of the first attempts to measure the
speed of light. At night he stood on one hilltop while an assistant stood
on another hilltop at a known distance, each holding a shuttered lantern.
First Galileo opened the shutter of his lantern. As soon as the assistant
saw the flash of light, he opened his own lantern. Using his pulsebeat,
Galileo attempted to measure the time between opening his shutter and
seeing the light from the assistant's lantern. With the distance and time
known, he could then compute the speed at which light traveled to the
distant hilltop and back again. He soon concluded that light travels so
rapidly that slow human reactions make it impossible to measure its
speed in this fashion.
Galileo
Successful laboratory experiments to measure the speed of light were

not performed until the mid-hSOOs, most of them using elaborate optical
equipment. From these experiments, we now know that the speed of
vacuum is about 3 x 10'' ni/sec (186,000 miles per second).

pioneering breakthrough in understanding light came from a simple experiment performed by Isaac Newton in the late 1600s. He passed
a beam of sunlight through a glass prism which spread the light out into
the colors of the rainbow as shown in Figure 4-1. This rainbow, called a
spectrum, suggested to Newton that white light is actually a mixture of
light in a
all
Figure 4-1
Whi'ii a
gluM
cvtvreil
prism and a spectrum
beam of while
/jiLsm,
band
llie
colors.
light
IS
light
pdues
Ihriiagh a
broken into a luinbini'
called a spectrum.
The numbers
on Ihe right side of the spectrum indicate

wai'elenglhs as described in the
text.
From his many experiments, Newton concluded that light is composed

of indetectably tiny particles. A rival explanation was proposed by the
Dutch astronomer Ghristian Huygens, who suggested that light travels in
the form of waves rather than particles.
The English physicist TlKjmas Young confirmed the wave nature of
light in 1801. Young demonstrated that the shadows of objects in light of
a single color are not crisp and sharp. Instead, the boundary between illuminated and shaded areas is overlaid with patterns of closely spaced
dark and light bands. These patterns are similar to the patterns produced by water waves passing the edge of a reef or barrier in the ocean.
Although some of these results could be explained by the particle theory
of light, experimenters soon obtained overwhelming evidence for the
wavelike behavior of light.
Further insight into the wave character of light came from calculations by the Scottish physicist James C^lerk Maxwell in the 1860s. As
mentioned in Chapter 3, Maxwell succeeded in describing all the basic
properties of electricity and magnetism in four equations. By combining
these equations. Maxwell demonstrated that electrical and magnetic ef-
Lii^/il. ajihfs.
ami
trlisuijir^
should travel through space in the form of waves, f'urthermore. his

proved that these waves should travel with a speed of about
3 X 10'* m/sec. Maxwell's suggestion that these waves do exist and are
observed as light was soon confirmed by a variety of experiments. Light
consists of oscillating electrical and magnetic fields, as shown in P'igure
4-1. Because of its electrical and magnetic properties, light is called electromagnetic radiation. The distance between two successive wave crests
is called the wavelength ot the light (see Figure 4-2).
fects
calculations
Figure 4-2
Electromagnetic radiation
Magnetic
All
field
jmms of light aiHiisI oj uscilliilnig elirtniul

and mugnt'lii JieUls that nwi'e through i'mpl\
space al a speed of 3
X 10"
misec.
tance between two successive crests
wavelength oj the
light,
niilcd h\ the lineeirase
which
Creek
is
The
dis-
called the
iisualh desig-
letter
llinnlidill.
The wavelength of visible light is extremely short, less than a thousandth of a millimeter. To express these tiny distances conveniently, scientists use a unit of length called the angstrom (abbreviated A), named
after the Swedish ph\sicist A. J. Angstrom), where 1 A = 10"'* cm. Experiments demonstrated that \isible light has wa\elengths covering the
range from about 4000 A tor \iolet light to about 7000 ,4 for red light.
Intermediate colors of the rainbow fall between these waxelengths (see
Figure 4-1).
More than a century elapsed between Newton's experiments with a
prism and the confirmation of the wave nature of light because its wavelength is so very short. Maxwell's equations, however, placed no restrictions on the wavelengths of electromagnetic radiation. In other words,
electromagnetic waves should exist with wavelengths both longer and
shorter than the 40007000 A range of visible light. Researchers began
to look for invisible forms of light, forms to which the cells of the
hinnan retina do not respond.
The British astronomer William Herschel fliscovered radiation Just
beyond the red end of the visible spectrum. In 1888, the German phvsicist Heinrich Hertz succeeded in producing light having wavelengths of
a few centimeters, now known as radio waves. In 1895, Wilhelm
Rontgen invented a machine that produces light with a wavelength
shorter than 100 A, now known as X rays. Modern versions of
Rontgen's machine are today found in medical and dental offices.
Over the years, radiation in many other wavelength ranges has been
discovered. Visible light is only a tinv fraction of the full extent of possible wavelengths collectively called the electromagnetic spectrum. As
shown in Figure 4-3. the clcctromagnetit speiirum strcitlus Irom the
longesi-waxelength radio waves to the shortesi-waxelengih gamma rays.
For i\ain|)le, at wavelengths slighllv longer than \isihle lighl, infrared
mm. From rougliU
radiation covers the range iron) aboui 7000 .\ to
to 10 cm is the range of microwaves, bevond which is the domain
1
1
mm
of radio waves.
At wavelengths shorter than that of visible light, ultraviolet radiation

extends from about 4000 .4 down to 100 A. W'axelenglhs between about
100 A and 0.1 A are \ ra\s, besond wiiidi is ilie domain of gannna
rays.
l.ifihl. vplK'..
The electromagnetic spectrum
Figure 4-3
The full array of

radiation
is
spectrum.
It
length
null trirsdijirs
all types
of electromugnelic
called the electromagnelic
extends
gamma
from
the shortest-wave-
rays to the longest-wavelength
radio waves. Visible light
forms only a
liny
portion nf the full eleclromngnetic specttiim.
4U00
Violet
Blue
Green
^^
bellow
Change
\
\,
I
IS.I)I.
IikIu
Red
000 A
11.VU- H
idial
on
Although these \arious types of electromagnetic radiation share many

example, they all travel at the speed of light), they
interact very differentlv with matter. \'our body is transparent to \ ravs
but not to \isible light; your eyes respond to \isible light but not to
gamma ravs; vour radio detects radio waxes but not ultraviolet light.
Consequently, astronomers use fundamentally different kinds of telescopes in these various wavelength ranges. For example, a radio telescope that detects radio waves from space is quite different from either
an X-rav telescope or an ordinary optical telescope. Because optical telescopes are the most common and familiar astronomical tool, we shall disbasic properties (tor
cuss
them
in detail
before turning to the exotic instruments that reveal
the non\isible skx.
refracting telescope uses a
.Although light travels
lens to concentrate incoming
more
starlight at a focus
of
at
about
;5
1()^
m/sec
in a
it moves
Ihe abrupt slowing
person walking from a
\atiuun,
slowly through a dense substance such as glass.
light entering a piece of glass is analogous to a

boardwalk onto a sandv beach: her pace suddenly slows as she steps
from the smooth pavement into the sand. And jtist as the person stepping back onto the boardwalk easilv resumes her original pace, light exiting a piece of glass resumes its original speed.
Furthermore, a light ray is bent as it passes from one transparent
medium into another at an angle oblit|iie to the surface between the
media. This phenomenon is called refraction and is caused by the
Light, optici.
and
change
the speed of light. Imagine driving a car from a smooth pavea sandy beach. If the car approaches the beach at an angle,
front wheels is slowed bv the sand before the other, causing
in
ment onto
one of the
telescopes,
the car to veer from
51
its
original direction.
To
describe the refraction of a light rav entering a piece of glass,

imagine drawing a perpendicular to the surface of the glass at the point
where the light strikes the glass, as shown in Figure 4-4. .\s a light rav
goes from a less-dense mediinn (such as air) into a more-dense medium
alwavs bent toward the perpendicular direcobliqueU onto a beach veers toward the direction perpendicular to the pavement sand boundarv. Upon emerging
from the other side of a piece of glass, light resumes its original high
speed, and the light ray is bent away from the perpendicular direction.
The exact amount of refraction depends on the speed of light in the
glass, which depends on the chemical composition of the glass.
Because of the refracting property of glass, a convex lens one that is
fatter in the middle than at the edges
causes incoming light ravs to
converge to a point called the focus, as shown in Figure 4-5. If the light
source is extreme!} far awav, then the incoming light ravs are parallel,
(such as glass), the light
is
tion, just as a car dri\ ing
Figure 4-4
Refraction
a piece of glass
is
bent
away from
light ray eitlenng
bent toward the perpendicu-
lar. .\s the ray leaves the
piece of glass
it is
the perpendicular.
and
the\
come
to a focus at a specific distance
from the
lens called the
focal length of the lens.

Stars
and planets are so
far
away
that light rays
from these
objects are
Consequentlv. a lens alwavs focuses light from an astronomical object as shown in Figure 4-5. An image of the object is
formed at the focus, and a second lens can be used to magnifv and examine this image. Such an arrangement of two lenses is called a refracting telescope, or refractor (see Figure 4-6). The large-diameter, longfoi al-iength lens at the front of the telescope is the objective lens. The
smaller, short-tocal-length lens at the rear of the telescope is the eyepiece lens.
The magnification, or magnifying power, of a refracting telescope is
ec]ual to the focal length of the objective lens divided bv the focal length
of the eyepiece lens. For exaniple. if the objective of a telescope has a
focal length of 100 cm and the evepiece has a focal length of A cm, then
the magnifving power of the telescope is 200 (usually written as 200X).
essentially parallel.
Figure 4-5
convex lens
convex lens
causes parallel light rays to converge to a

focus.
is
The distance from
the lens to the focus
the focal length of the lens.
you build a telescope using only the instructions given so far, vou
probablv be disappointed with the results. You will see stars surrounded bv fuzzv. rainbow-colored halos. This optical defect, called
chromatic aberration, exists because a lens bends different colors of light
through different angles, just as a prism does (recall Figure 4-1).
If
will
Objective lens
Eyepiece lens
Figure 4-6
refracting telescope
.4
re-
fracting telescope consists of a large, longfocal-length objective
lem and a
small, short-
focal-lenglh eyepiece tens that magnifies the
image formed
at the
focus of the objective
lens.
52
Chromatic aberration
Figure 4-7
gle lens suffers
from a
Light, optics,
and
telescupe:
The problem
sin-
defect called chromatic
aberration in which different colors of light
have
lem
slightly different
is
focal
lengtlis.
adding a second
corrected by
of a different kind of glass than
This problens
Ihiit of
made
the
first lens.
Focal point
Focal point
for red
liglit
for blue light
to a vat of molten
an optician can manufacture different kinds of glass. The speed of
a fact
light in glass depends on the chemical composition of the glass
that opticians use to correct for chromatic aberration. Specifically, a thin
lens can be mounted just behind the main objective lens of a telescope as
diagramed in Figure 4-7. By carefully choosing different kinds of glass
for these two lenses, the optician can ensure that different colors of light
By adding small amounts of various chemicals
glass,
come
to a focus at the
same
point.
Chromatic aberration is the most severe of a host of optical problems

that must be solved in designing a high-quality refracting telescope.
Nineteenth century master opticians devoted their lives to overcoming
these problems, and several magnificent refractors VkJere constructed in
the late 1800s. The largest refracting telescope, completed in 1897, is
located at the Yerkes Observatory near Chicago (see Figure 4-8) and has
an objective lens 102 cm (40 in.) in diameter. The second largest refracting telescope is located at the Lick Observatory near San Jose, California.
This refractor has an objective lens whose diameter is 91 cm (36 in.). All
refractors have extremely long focal lengths. For example, the Yerkes
Figure 4-8
large refracting telescope
This giant refracting telescope loas built in the
housed at Yerkes Obsema-
late
1800s and
tory
near Chicago. The objective lens
102 cm (40
tube
is
19i
is
in.) in
diameter,
and
is
the telescope
(63i ft) long.
reflecting telescope uses a

mirror to concentrate
incoming starlight at a focus
ufractor has a focal length of 19.35 m (631 ft).

Few major new refracting telescopes have been constructed in the
twentieth century. There are many reasons for the modern astronomer's
lack of interest in this type of telescope. First, because faint light must
pass readily through the objective lens, the glass from which the lens is
made must be totally free of defects such as bubbles that frequendy
form when molten glass is poured into a mold. Consequently, the glass
for the lens is extremely expensive. Second, glass is opaque to certain
kinds of light. Even visible light is dimmed substantially in passing
through the thick slab of glass at the front of a refractor, and ultraviolet
radiation is largely absorbed by the glass lens. Third, it is impossible to
produce such a large lens that is completely corrected to eliminate chromatic aberration. Fourth, it is difficult to support such a heavy lens without blocking the path of light into the telescope. All of these problems
can be avoided by using mirrors instead of lenses.
Reflection can be described very simply.

ine
drawing a perpendicular
To understand
reflection,
imag-
where a
The angle between
to a mirror's surface at the point
light ray strikes the mirror, as
shown
in
Figure 4-9.
the arriving (incident) light ray and the perpendicular is always equal to
the angle between the reflected ray and the perpendicular. Knowing
this,
Isaac
Newton
realized that a concave mirror will cause parallel light
rays to converge to a focus as
shown
in
Figure 4-10.
The
distance be-
Lij^hl. uj)tHs. mill tekscupe.
tween the reflecting surface and the focus
is
called
lie
focal length of
the mirror.
An image
of a distant object is formed at the focus of a concave mirorder to view the image, Newton placed a small, flat mirror at a
45 angle in front of the focal point as sketched in Figure 4- Hi. This
secondary mirror deflects the light rays to one side of the reflecting telescope, or reflector, where the astronomer can place an evepiece lens to
magnify the image. .\ telesc()])e ha\ing this optical tlesign is appropriatel) called a Newtonian reflector. Ihe magnifying power of such a reror. In
flecting telescope
is
focal length of the
calculated in the
primary mirror
is
same way
as for a refractor: the

divided by the focal length of the
eyepiece.
Figure 4-9
lllie
The angle
Reflection
which a beam
iij
light iijijiioache!,
angle of iniidenie)
is
Willi
a mirrui
ahra\.\ equal tu the
angle with which Ihe beam
i\
repected fiom the
miriui (the angle of reflectiunj.
Useful modifications of Newton's original design have since been

made. The primary mirrors of many major reflectors are so large that
the astronomer can actually sit at the undeflected focal point, directly in
front of the primary mirror. This arrangement is called a prime focus
(see Figure 4-1 \a). The "observing cage," in which the astronomer rides,
blocks only a small fraction of the incoming starlight.
.Another popular optical design, called a Cassegrain focus, has the
advantage of placing the focal point at a convenient and accessible location. A hole is drilled directlv through the center of the primarv mirror.
A convex secondary mirror placed in front of the original focal point is
used to reflect the light rays back through the hole (see Figure 4-1 \c).
Alternatively, a series of mirrors can be used to channel the light rays
away from the telescope to a remote focal point. Heavy optical equipment that could not be mounted directly on the telescope is located at
the resulting coude focus (see Figure 4-1 \d).
To make a reflector, an optician grinds and polishes a large slab of
glass into the appropriate concave shape. The glass is then coated with
or aluminum or a similai highly reflective substance. Defects inside
the glass such as bubbles or flecks of dirt do not detract from the telescope's effectiveness, as they would in the objective lens of a refracting
silver
telescope.
Figure 4-10
concave mirror
mirror causes parallel tight rays
tu
ccniiave
converge
a focus. The distance between the mirror

Ihe focus
IS
Figure 4-11
the must
Reflecting telescopes
popular optical desigin /or
telescopes: (a)
l-oui of
lelleitiiig
prime focus, (b) Neielonian
/of lo, (c) Cassegrain focus,

focus.
In
and
the jural length of the mirror.
and
(d) coude
Furthermore, reflection is not affected by the wavelength of the light.

Therefore, the image formed by a mirror does not suffer from chromatic aberration. Finalh, the miiror can be fulK supported bv braces on

Light, optics,
and
telescope:
its back, so that a large and heavy mirror can be mounted without much
danger of breakage or shape distortion.
Eleven reflectors exist with primary mirrors measuring at least 3 m in
diameter. The largest, located in the Soviet Union, has a mirror 6 m
(19.7 ft) in diameter. The second largest is at the Palomar Observatory
southern California, with a mirror 5 m (16.4 ft) in diameter. In the

matching pair of telescopes was built in Arizona and
Chile. Both have mirrors 4 m (13.1 ft) in diameter. These two telescopes
(see Figure 4-12) allow astronomers to observe the entire sky with essenin
early 1970s, a
the same instrument.

Astronomers strongly prefer large telescopes. A large mirror intercepts and focuses more starlight than does a small mirror (see Figure
4-13). A large mirror therefore produces brighter images and detects
tially
Figure 4-12
The 4-meter
Cerro Tololo
This telescope
telescope at
(j
located on a
mountaintop near Santiago, Chile.

at the Kitt
telescopes
Peak Obsen>atory
Its
early 1970s.
twin
in Arizona.
have been in operation since
is
Both
the
(NOAO)
The
power of a
primary mirror.
For example, the 200-inch mirror at Palomar Observatory has four times
the area of the 100-inch mirror at Mt. Wilson Observatory. Therefore,
the Palomar telescope has four times the light-gathering power of the
Mt. Wilson telescope.
A large telescope also increases the sharpness of the image and the
extent to which fine details can be distinguished. This property is called
resolving power. With low resolving power, star images are fuzzy and
blurred together. With high resolving power, images are sharp and crisp.
The resolving power of a telescope is measured as the angular distance between two adjacent stars whose images can just barely be distinguished under ideal observing conditions. Large modern telescopes
such as those at Palomar, Kitt Peak, and Cerro Tololo are calculated to
have resolving powers better than 0.1 arc sec. In practice, however, this
exceptionally high resolving power is never achieved. Turbulence and
impurities in the air cause star images to jiggle around, or twinkle. Even
when photographed through the largest telescopes, a star looks like a
fainter stars than does a small mirror.
telescope
is
light-gathering
directly related to the area of the telescope's
tiny circular blob rather than a pinpoint of light.
The angular diameter
of a star's image is called the seeing disk and is

measure of the best possible resolution that can be achieved.
The size of the seeing disk varies from one observatory site to another.
At Palomar and Kitt Peak, the seeing disk is roughly 1 arc sec. The best
conditions in the world (seeing disk = i arc sec) have been reported at
the observatory on top of Mauna Kea, a 14,000-foot volcano on the island of Hawaii. Only in the vacuum of outer space could the theoretical
resolving power of a large telescope be achieved.
Significant engineering problems are associated with building large
refiectors. Very large mirrors (more than about 4 m in diameter) are
slabs of glass so heavy that the mirror's shape changes slightly as the tela realistic
/
////////
Figure 4-13
Light-gathering power
large mirror intercepts more starlight than does
a small mirror. Large mirrors therefore produce brighter images than small mirrors.
'4/
Liglil, optics,
The Multiple-Mirror
Figure 4-14
telescopes
Tele-
This aerial pliotugiaph shows Ike six
scope
1
and
.S-m minors that together constitute the
first
multipte-minor telescope. The total area

of the SIX mirrors
mirror.
is
equal
to
one 4.5-m
(MMT)
is turned toward different parts of tfie skv. Tlie mirror actually

under its own weight, thereby detracting from the sharpness of the
focus and the quality of the resulting image. New techniques for build-
escope
sags
ing thin, light-weight mirrors should help alleyiate this problem.

Another approach is to mount seyeral smaller mirrors together, aimed
at the same focal point. The Multiple-Mirror Telescope atop Mt. Hopkins in Arizona has six mirrors, each measuring 1.8
(6 ft) in diameter,
mounted together as shown in Figure 4-14. The total light-gathering

power of this arrangement is equi\alent to one 45-m mirror. The MMT,
as
it
is
called, thus really ranks as the third largest telescope in the
world.
Figure 4-15
The 10-meter Keck
This model shou": the
dcMgn
telescope
a giuni
scope planned for the summit of
tele-
Mauiui Kea
on the island of Hawaii. Thirly-six hexagonal

mirrors, each measuring 1.8 m (5.9 ft) across,
will together
10
111
have the
effect
of one minor
diameter. (California Institute
1990s.
of Technologyj
MMT
The
has proyen so successful that astronomers around the
world are now planning even larger multiple-mirror telescopes. For example, 36 indixidualK controlled hexagonal mirrors will be mounted together as shown in Figure 4-15. The result is equiyalent to one mirror
10 m (32.8 ft) in diameter. This enormous telescope will be housed at
the summit of Mauna Kea and will probably be in operation in the
radio telescope uses a
large
concave dish to
radio
waves
reflect
to a focus
all information that astronomers could gather about the

universe was based on ordinary visible light. But with the discovery of
nonvisible electromagnetic radiation, scientists began to wonder if objects
Until recently,
in the
universe might also emit radio waves,
rays,
and infrared and
of the universe at these nonvisible

wavelengths would enhance our understanding of the cosmos.
The first evidence of nonvisible radiation irom outer space came from
the work of a young radio engineer, Karl Jansky of Bell Telephone Laboratories. Using long antennas, Jansky was investigating the sources of
radio static that affects short-wavelength radiotelephone communication.
In 1932, he realized that a certain kind of radio noise is strongest when
the constellation of Sagittarius is high in the sky. The center of our galaxy is located in the direction of Sagittarius, and |ansky concluded that
he was detecting radio waves from elsewhere in the galaxy.
Astronomers were not quick to pursue this line of research. Only one
person, Grote Reber (an electronics engineer living in Wheaton, Illinois),
pursued the matter. In 1936, Reber built the first radio telescope in his
backyard to map radio emission from the Milky Way. His design was
modeled after an ordinary reflecting telescope, with a parabolic "dish"
ultraviolet radiation. Surely, views
Light, optics,
Figure 4-16
this
radio telescope
radio telescope
is
45.2
The
m {140 ft)
and
tflrscope.
dish oj
in di-
one of several large instrumeriLs

at the Natiotial Radio Astronomy Obsenmton
ameter. It
is
near Green Bank. West Virginia.
(NR.W)
measuring 9. ni in diameter. The radio receiver at

the focal point of the metal dish was timed to a wavelength of 1.85 m.
By 1944, when Reber completed his map of the Milky Way, astronomers had begun to take notice of these developments. Shortly after
(retlecting antenna)
World War
II, radio telescopes began to spring up around the world.

Radio observatories are today as common as major optical observatories.
Like Reber's prototype, the standard radio telescope has a large parabolic dish (see Figure 4-16). A small antenna tuned to the desired frequency is located at the focus. The incoming signal is relayed to amplifiers and recording instruments, which are typicallv located in a room at
the base of the telescope's pier.
Astronomers were at first not enthusiastic about detecting radio noise
from space, in part because of the low resolving power of early radio telescopes. The resolving power of any telescope decreases with an increase
in wavelength. In other words, the longer the wavelength, the fuzzier
the picture. Because radio radiation has verv long wavelengths, astronomers thought that radio telescopes could produce only blurrv, indistinct
views.
Very large radio telescopes can produce somewhat sharper radio images: the bigger the dish, the better the resolving power. For this reason,
most modern radio telescopes have dishes more than 100 ft in diameter.
There is, however, another clever way to achieve high resolution. Two
radio telescopes separated by many kilometers can be hooked together
with electrical wires. This technique is called interferometry because the
incoming radio signals are made to "interfere" or blend together, so that
the combined signal is sharp and clear. The result is verv impressive: the
effective resolving power is equivalent to that of one gigantic dish with a
diameter equal to the distance between the two telescopes.
Interferometry techniques were exploited for the first time in the late
1940s, and gave astronomers their first detailed \'iews of radio objects in
the sky. Radio telescopes separated by thousands of kilometers were
linked together to give resolving power much higher than that of optical
telescopes. This technique is called very-long-baseline interferometry
(VLBI). The best possible resolution would be obtained b\ two telescopes
Ulil. njytus.
Figure 4-17
lu'e)il\-semi
The Very Large Array

rndw
lrlescope\
are arranged along the
anm
iif
llw
\'1..\
and
Irlfscnfit
Tlir
wslrm
of a Y in lenlial
Xeu' Mexico. The ninth arm of ihe array '

19 km long: Ihe uiiilhwesl and sinilhea\l aims
are each 21
km
hnnj.
(XKAOl
on opposite
sides of the Eartli. In that case, features as small as O.OOOOl

arc sec could be distinguished at radio wavelengths
a resolution
100,000 times better than the sharpest pictures from ordinary optical telescopes.
One of
the finest systems of radio telescopes began operating in 1980

near Socorro, New Mexico. Called the Very Large Array
(VLA), it consists of 27 parabolic dishes, each 26
(85 ft) in diameter.
The 27 telescopes are arranged along the arms of a gigantic Y that covers an area 27 km (17 miles) in diameter. Only a portion of the VLA is
in the desert
shown
in Figure 4-17. This system produces radio views of the sky with
resolutions comparable to that of the very best optical telescopes.
Telescopes in orbit around

the Earth detect radiation
that does not penetrate
the atmosphere
.\s the success of radio astronomy began to mount, astronomers started

exploring the possibilit\ ol making observations at other nonvisible wavelengths. Unfortunately, the Earth's atmosphere is opaque to many wavelengths. Very little radiation other than visible light and radio waves
manages to penetrate the air we bieathe.
Water vapor is the main absorber of infrared radiation from space.
Infrared observatories are therefore located at sites of low humidity. For
example, the 14,000-ft summit of Mauna Kea on Hawaii is exceptionally
dry, and infrared observations are the primary fiuiction of NAS.\'s 3-m
telescope there.
.\nolher
jjossibility is lo lake a telescope up in an .iii|)laiie. Fliat is the

behind the Kuiper .Airborne Obser\ator\ (K.\0) shown in Figin e 4-18. File airplane carries a 1-m reflecting telescope to an altitude of
12 km (40,000 ft), placing the observatory above 90 ]x'rcent of the atmospheric water \apor.
A telescope in Earth oibii of lei s ilie best arrangement. In 198,1, the
Infrared .Astronomical Satellite (IRAS) was laiuuhed into a 90()-km-high
polar orbit. As shown in Figure 4-19. the satellite was designed around a
,'i7-cm (222-in.) reflecting telescope that viewed the entire sk\ <ii infrared
wavelengths that never penetrate the Earth's atmospheie.
basic idea
Light, optics,
Figure 4-18
The Kuiper Airborne
Observatory
This C-141 jet airplane carries
and
telescopes
a 1-m reflecting telescope specifically designed
for infrared obsen'ations. The observing portal

(through which the telescope
is
aimed) can be
seen on the fuselage just in front of the
wing.
(NASA)
made from space. During

and Skylab astronauts carried small ultraviolet
telescopes above the Earth's atmosphere to give us some of our first
iews of the ultraviolet sky. Small rockets have also been used to place
ultraviolet cameras above the Earth's atmosphere briefly. A typical view
is shown in Figure 4-20, along with a corresponding infrared view from
IRAS, a view in visible light, and a star chart.
Some of the finest ultraviolet astronomy has been accomplished by the
The
best ultraviolet observations are also
the early 1970s, Apollo
International Ultraviolet Explorer (lUE), which was launched in 1978.

satellite (see Figure 4-21) is built around a Cassegrain telescope
(22 ft).
with a 45-cm (18-in.) mirror and a total focal length of 6.74
The
The Infrared Astronomical
Figure 4-19
This
Satellite
satellite
contains a reflecting
telescope providing complete coverage oj the
sky at infrared wavelengths that cannot be detected
from
launched
the gtound.
The
1983. (NASA)
satellite
was
Observations cover the range from 1160 to 3200 A.

For decades, astronomers have dreamed of having a major observatory in space. Although satellites like the IRAS and lUE give excellent
views of selected wavelength regions, astronomers are tantalized by the
prospects of one very large telescope that could be operated at any wavelength
from the infrared through the visible range and out into the
far-ultraviolet. This is the mission of the Hubble Space Telescope (see
Figure 4-22) to be carried aloft by the Space Shuttle in 1988. This instrument will dominate astronomy for the rest of the twentieth century.
The primary mirror of the Hubble Space Telescope has a diameter of
2.4
sky
is
('7.8 ft). The resulting light-gathering power is so large and the

so dark that the telescope will be able to detect stars forty times
dimmer than those that can be photographed from Earth-based observais scattered by the atmosphere.
not be degraded by the atmosphere, the actual
resolution of the Hubble Space Telescope will be very close to the theoretical value. For example, when operated along with an auxiliary instrument called the faint-object camera the telescope will have a resolution
of 0.02 arc sec. That is a fiftvfold improvement over the best conditions
tories
where ambient
Because
its
images
light
will
observatories such as Falomar, where the seeing disk is I arc sec.

The extraordinary light-gathering and resolving powers of the Hubble
Space Telescope will permit astronomers to make observations that were
unthinkable only a few years ago. One observing program will have the
at
Hubble Space Telescope search
for
new
planets
around other
stars.
Lighl,
Figure 4-20
infrared,
Orion as seen
and
visible
ofilirs.
and
telescope.
in ultraviolet,
An
wavelengths
til-
Iravwtet view (a) 0/ Ihe conitellatum of Orion
was obtained diinng a hnej rocket flight on

5, 1975. The 100-sec exposure cov-
December
ers the
wavelength range
"false color"
12502000
IRAS uses
view (b) from
to display specific
A, The
color
ranges of infrared wave-
lengths: red indicates
long-wavekngth
i?ifrared
radiation: green indicates intermediate-wave-
length radiation:
and
blue shows short-wave-
length infrared radiation.
an ordinary
For comparison,
optical photograph (c)
and a
star chart (d) are included. (Courtesy
George R. Carrutliers,
of
NRL: NASA:
Palomar Observatory)
Many
and galaxies whose photographs are

be seen with unprecedented clarity and brilliance. Currently unsuspectecl details and subtleties should be reyealed.
Perhaps the greatest surprises will come from the discoyery of totally
new objects in space. Eyer since Galileo tin ned his telescope toward the
familiar nebulae, star clusters,
throughout
Figure 4-21
Explorer
671-kg
The International Ultraviolet
Since
its
mtellile has
launch
arrays
tlial
satellite
are solar-
provide electrical power for
the radio transmitters
equipment.
this
obseri'a-
The dark blue
panels at the midsection of the

cell
197S.
produced superb
tions in the far-ultravinlet.
(NASA)
and
other electronic
this
book
will
Lig:lil. iijtiiis. 1111(1 Iclesailii
The Hubble Space Telescope
Figure 4-22
The Space Shuttle

2.-1-111
in
will
used
hi'
lo
Hubble Space Telescupe
1988. Duriiijf
it.\
place tim
into
Earth
orbit
anticipated fifteen-year
lijetime, the telescope will be
used
to .study the
heavens over wavelengths from the infrared

through the ultraviolet.
(NASA)
and saw four moons orbiiini^ Jupiter, each new generation of astronomical instrument has disclosed pre\ioiisly iinimagined objects and
processes, often more bizarre than the strangest science fiction. It happened twice in the 1960s, when radio telescopes found quasars and pulsars, and again, in the 1970s, when X-ray telescopes detected bursters.
skies
Figure 4-23
[left]
Superb views
of the
glimpses of the
from
HEAOs
The three
X-iu\ sky and tuntuliznig
gamma-ray
sky were obtained
these three Earth-orbiting satellites.
satellite is
roughly the size
and
Each
ma.ss of a large
automobile. In addition to solar panels (for

electrical power), the sides of
each spacecraft
are covered with sensitive X-ray

ray detectors.
Figure 4-24 [right]
The
Observatory
gamma-ray
eiierijy
ellite,
which
and gamma-
(NASA)
IS
best
The
Gamma Ray
views oj the high-
sky will
come from
1980s. The 8000-kg (19.000-pound)

will be placed into
Shuttle.
this sat-
scheduled for launch in the late
(NASA)
Earth
orbit by the
satellite
Space
With serendipity so commonplace, many astroncjmers predict that it will

happen again in the 1990s.
Neither X rays nor gamma rays penetrate the Earth's atmosphere, so
observations at these extremely short wavelengths also must be done
from space. Astronomers got their first quick look at the X-ray sky
with brief rocket (lights during the late 1940s. Several small satellites
lainiched during the early 1970s viewed the entire X-ray and gamma-ray
skv, revealing hundreds of previously unknown soinces, including at
least one black hole.
Although heroic in their day, these preliminary efforts pale in comparison to the detailed views and results from three huge satellites
launched between 1977 and 1979. Called High Energy Astrophysical
Observatories (HEAO). these satellites each carried an array of X-ray
and gamma-ray detectors. Thousands of sources were discovered all
Light, uplics.
and
telescopes
^^î^^^f^f^^*.)^^
Figure 4-25
and X-ray
The radio, infrared,
skies
visible,
Thi\e jour views show the
eiilne sAv at (a) rnilio. (b) infrared, (c) visible,
tion
anil (d) .V-rav wavelengths.

IS
drawn
Each
illustra-
with the .Milky \\'a\ stretching
liorizontalh across the picture. (Griffith Uh-
sen>ati)n: /el Propulsion Lahoratory)
^:--'-^'''^':;i^-i^^''^^
The second satellite in this series was especially successful

producing high-quality X-ray images of a wide range of exotic objects.
Because it was launched near the hiuidredth anniversary of Albert Einstein's birth, this satellite was called the Einstein Obser\atory. X-ray yiews
from this observatory appear throughout this book.
The shortest-wavelength, highest-energy form of light is gamma radiation. Astronomers have had onU their first tantalizing glimpses in this
exotic region of the electromagnetic spectrimi, and mauN hopes and expectations lie with the C.amma Ray Observatory (CiRO) to be launchetl in
across the sky.
in
the late 1980s (see Figure 4-24).
The advantages and
benefits of these Earth-orbiting observatories can-
We are no longer limited to the narrow ranges

of wavelengths that manage to leak through the shimmering, hazy atmosphere we breathe (see Figure 4-25). For the first time we are really
seeing the universe.
not be overemphasized.
Summary
Light
is
wave phenomenon consisting of oscillating electrical and magthrough space at a speed of about 3 x 10'' m/scc.
netic fields that travel
The wa\clcngth of light is associated with its color: wavelengths of visible

range from .ibout 4000 A for violil lit;hi lo 7000 Afor red light.
light
\'isible light forms only a small [jortion of ihe electromagnetic spectrum.

Infrared radiation, nticrowaves. and radio waves have wavelengths larger
than those of visible light. L'ltra\iolet radiation. X ravs. and gamma rays
lia\e wa\elengllis shorter tiian ihose ol \isil)ii- liniil.
Refracting telescopes (refractors) pnxiun- iniat;es

they pass through glass lenses.
Chromatic aberration
lengths
is
is
an optical dilcd uluieln
bent different aniounls
1)\
,i
lens.
l)\
hendini;
iigiil
lii;hi
ol (iill(i<.-iu
ia\s as
wave-
Ligbl, uplirs, inul telescope.
Reflecting telescopes (reflectors) produce images b\ reflecting light ravs to

from ciir\'ed mirrors.
a focus point
Reflectors are not subject to

ness of refractors.
many
of the problems that limit the useful-
Radio telescopes have large reflecting antennas (dishes) that are used to
focus radio waves.
Very sharp radio images are produced with combinations or arrays of
radio telescopes with a technique called interferometrv.
The
Earth's atmosphere is transparent to visible light and radio waves, but

absorbs much of the radiation at other wavelengths arriving from space.
observations
at other wavelengths, astronomers depend upon teleFor
scopes carried above the atmosphere bv high-altitude airplanes, rockets, or
it
satellites.
new information
about the universe, permitting coordinated observation of the sky at all
wavelengths.
Satellite-based observatories are giving us a wealth of
Review questions
Give everyday examples of the phenomena of refraction and reflection.
What
defect,
is
chromatic aberration? What kinds of telescopes suffer from

it be corrected?
this
and how can
What
is
meant bv the resolving power of
a telescope?
4 What are the principle advantages of a reflecting telescope over a refractor?
5 What limits the resolving power of the 5-meter telescope
at
Palomar?
Why
Why
must X-rav telescopes be placed above the Earth's atmosphere?
Wh\
will very
can radio astronomers observe at any time during the day whereas
optical astronomers are mostly limited to nighttime observing?
Advanced questions
large telescopes of the future
make
use of multiple mirrors?
9 Quite often, advertisements appear for telescopes that extol their magnifyabilities. Is this a good criterion for evaluating telescopes? Explain your
answer.
ing
10
Show bv means of
ing telescope
is
diagram why the image formed by a simple
refract-
"upside down."
11 Discuss the advantages and disadvantages of using a small telescope in

Earth orbit versus a large telescope on a mountain top.
12 If you were
tors
in
charge of selecting a
would you consider iinportant?
site
for a
new
observatory, what fac-
Lif;lil. (iptics.
For further reading
and
telescopes
...
Bok. B. "The Promise of the Space Telescope." Mercian, Mav/Iune 1983

p. 60.
Cohen, M.
In Quest of Telescopes. Sk\ Publishing Corp. and Cambridge

University Press, 1980. An astronomer's narration of what it is like
observe with the ivorld's largest telescopes.
Harrington,
S.
"Selecting
Your
First
to
Telescope." Mercun: |ulv/.\ug. 1982,
p. 106.
Henbest, N., and Marten, M. The New Astronomy. Cambridge University

Press, 1983. A beautiful album of the universe as seen in all the hands
of the electromagnetic spectrum.
Learner, R. Astronomy Through
the Telescope.
Van Nostrand Reinhoid,
1981. Illustrated history of telescopes and the discoveries they made possible.
Robinson, L. "The Frigid World of IRAS." Sk\

p.
Schorn. R. "Astronom\
p.
Sullivan,
iff
Telescope
.Apr
1984
339.
in the
Next Decade." Sky
isf
Telescope,
Apr 1982
339.
W. "Radio Astronomv's Golden .Anniversarv." Sky

'
Dec. 1982,
p.
544.
ef Telescope
'
spectra and spectral lines
When a
electnc
spark passes through a gas. atoms of the gas

emit radiation at certain characteristic wavelengths. This
photograph shows (from tup
to
bottom) the spectra of lithium, iron, bnnniii,
and
calcium. At bottom
i.s
and Lomh)
of light
Our knowledge
about the planets, stars, and galaxies comes primarily from

analyzing the electromagnetic radiation they emit. As soon as we have more
information about the properties of light as background for the chapters that
we will be ready to turn to the study of the universe. We find that
follow,
light has both waiwlike
a continuous spec-
trum obtained h\ passing white

prism. (Hau.scli
The laws
light
through a
and
particlelike properties.
object. Finally,
we
an
We
learn that the electro-
affected by the temperature of the

discuss spectral lines and explore their relationship to
magtietic radiation emitted by
object
is
the structures of atoms.
Most of our knowledge about the universe comes from stud\'ing light
from distant objects. When we speak of "light"
from stars and galaxies, we mean the full electromagnetic spectrum (see
Figure 4-3). The human eye detects only the narrow band of wa\elengths that we call visible light. Astronomers also view the universe with
gamma rays, X rays, ultraviolet and infrared radiation, microwaves, and
arriving at the Earth
radio waves.
The laws
In
oj light
Chapter
65
4,
we saw
that light consists of oscillating electric
netic fields (see Figure 4-2) that
nearlv
km/sec
;^()().()()()
and mag-
move through empty space at a speed of

Thus light is often called elec-
(186.()()() miles/sec).
tromagnetic radiation. Ihc distance between adjacent crests in a light

wave (shown in Figure 4-2) is called the wavelength.
All objects emit electromagnetic radiation. Stars like the Sun emit radiation primarily at visible wa\clengths (from 4000 A to 7000 A). During
the late hSOOs, physicists discoveied how the intensity and wavelengths of
light emitted b) an object depend on the object's temperature. Before we
learn about these important discoveries, we first must see how scientists
denote temperature.
Temperatures are usually

given on the Celsius or
Kelvin temperature scales
rhere are three temperature scales that are commonly used, for various
jKuposes (see Figure 5-1). Temperatures are expressed throughout most
of the world in degrees Celsius (C;). The Celsius temperature scale is
ba.sed on the behavior of water, which freezes at 0C and boils at
00C:
at sea level. This scale, once known as the centigrade scale, was renamed
in honor of the Swedish astronomer Anders Celsius, who proposed it in
1
1742.
For many purposes,
scientists prefer to
express temperatures in a unit
named alter the British physicist William Thomson

who made man\ important contributions to our knowl-
called the kelvin (K),
(Lord Kehin).
edge about heat and temperature. On the KeUin temperature scale,

water freezes at 273 K and boils at ;?73 K. Because water must be heated
through a change of either 100 K or 100C to go from the freezing
point to the boiling point, the "size" of a kelvin is the same as the "size"
of a degree Celsius. A temperature expressed in kelvins is always equal
to the temperature in degrees Celsius phis 273. Scientists usualh prefer
the Kelvin scale because of its phvsical interpretation of the meaning of
temperature.
All substances are made of tin\ atoms (t\ pical atomic diameters are
about 10"** cm) that are constantly in motion. The temperature of a subMelting point
;------ of gold
stance
of silver
directly related to the average
speed of its atoms. If something

atoms are moving at high speeds. If a substance is cold, its
atoms are moving much more slowly.
The coldest possible temperature is the temperature at which atoms
is
Meltine point
hot,
is
its
move as slowly as possible (they can never i|uiie stop

minimum possible temperature, called absolute zero,
^
Boiling point
sullur
for the Kelvin scale. Absolute zero is

In the United States, many people
K, or
still
Ireez.ng ponil
water
.1
Boiling point
ol
__
oxygfn
Figure 5-1
Temperature scales
l>emliire sriiles ine

CeLstiis.
and
iti
(cdiiiikiii
l-'ahietiheit.
is
Thirr
its
starting point
the coldest possible temperature
absolute zero.
ten,
Keh'in.
Scientnts usually jm-
fer the Kelvin scale because

(0 K)
ii.\e:
is
The
the starting point
-273C.
use the archaic Fahrenheit
expressing temperatures in degrees Fahrenheit (F). When the

Cierman physicist Gabriel Fahrenheit introduced this scale in the early
17()()s, he intended 0F to lepreseni the coldest temperature then achiexable (with a mixture of ice and saltwater) and 1()()F to represent ihe
temperature of a healthy human bodv. On the Fahrenheit scale, water
lieezes at 32F and boils at 212F.
Fable 5-1 on page 6(5 relates various teiiipeiatiires on the Kelvin, Celsius, and Fahrenheit scales.
Most scientists who study the solai system express temperatiues (such
as the temperatures on the surfaces of the planets) using the C:elsius
scale. However, when we discuss stars, galaxies, and other astronomical
scale,
Boiling point
of water
complctelv).
phenomena, we
will
use the Kelvin scale.
TABLE
5-1
The lawi of
tight
Celsius (C)
Fahrenheit
(F)
-273
-460
Various temperatures on
Kelvin (K)
different scales
Absolute zero
Liquid helium boils
-269
-452
Liquid oxygen boils
90
-183
-297
Water freezes
273
"Room temperature"
295
22
72
Water
373
100
212
boils
Sulfur boils
Iron melts
Any
718
445
833
1808
1535
2795
Iron boils
3273
3000
5432
Carbon
5100
4827
8721
The
object emits
electromagnetic radiation
with intensity and
wavelengths related to the
temperature of the object
32
boils
amount of energy radiated by an object depends upon its temhotter the object, the more energy it emits as electromagnetic radiation. The dominant wavelength of the emitted radiation also
depends upon the temperature. A cool object emits most of the energy
total
perature.
at
The
long wavelengths, whereas a hotter object emits most of the energy
at
shorter wavelengths.
These basic phenomena are familiar to anyone who has watched a

welder or blacksmith heat a bar of iron (see Figure 5-2). As it starts to
heat up, the bar begins to glow with a deep red color. As the temperature rises, the bar begins to give off a brighter, reddish-orange light. At
still higher temperatures, it shines with a brilliant, yellowish-white light.
If the bar could be prevented from melting and vaporizing, at very high
temperatures it would emit a dazzling, blue-white light.
In 1879, the Austrian physicist Josef Stefan summarized the results of
his experiments on this phenomenon by stating that an object emits energy
at a rate proportional to the fourth power of the object's temperature (measured in
kelvim). In other words, if you clouble the temperature of an object (for
example, from 500 K to 1000 K), the energy emitted from the object's
surface each second increases by a factor of 2** = 16. If you triple the
temperature (for instance, from 500 K to 1500 K), the rate of energy
emission increases by a factor of 3^ = 81.
Five years after Stefan announced this law, another Austrian physicist,
Ludwig Boltzmann, showed how it could be derived mathematically from
basic assumptions about atoms and molecules. The law today is commonly known as the Stefan Boltzmann law.
Figure 5-2
Heating a bar of iron
As
the
temperature of the heated bar increases, the
amount of energy radiated by the bar increases. The dominant wavelength of light
emitted by the bar also decreases, causing the
glowing iron
to
change
color.
These
effects
described by the Stefan Boltzmann law
Wien's law.
are
and
îÔ
The laws of
light
The Stefan-Boltzmann law can be stated as a simple equation. If

the energy emitted from each square centimeter of an object's surface
each second, then
E =
is
(tT'
where T
number
is
the temperature of the object in kelvins

Stefan-Boltzmann constant.
and a ("sigma")
is
called the
Boltzmann showed that this law is best obeyed by an object that is a

perfect absorber of energy and thus does not reflect any light. Such an
object would be absolutely black at all wavelengths and is therefore
called a blackbody. The energy that a blackbody does emit depends
we
only on its temperature. Ordinary objects are not perfect absorbers
can see them because they reflect some light and thus the amount of
energy they emit is slightly different from the amount calculated from
the Stefan-Boltzmann law.
A star efficiently absorbs
all radiation falling on it from the outside.

a star behaves like a blackbody and astronomers can use the
Stefan-Boltzmann law to relate the star's brightness to its surface tem-
Thus
perature.
object emits radiation over a wide range of wavelengths, but
always a particular wavelength (A^^^) at which the emission of
energy is strongest. This dominant wavelength gives a glowing hot object
its characteristic color.
In 1893, the German physicist VVilhelm Wien discovered that the domi-
Any
there
is
nant wavelength (K^^^) of radiation emitted by a blackbody is inversely proporshorter

tional to Its temperature. In other words, the hotter an object, the
This
the dominant wavelength of the electromagnetic radiation it emits.
is today called "Wien's law.
Wien's law can also be stated as a simple equation. If
ured in centimeters, then
relationship
A,ax
is
meas-
in kelvins.
is the temperature of the blackbody measured
the Stefan-Boltzmann law, we see that any object with a temperature above absolute zero (0 K) emits some electromagnetic radiation.
From "VX'ien's law. we find that a very cold object with a temperature of
only a few kelvins emits primarily microwaves. An object at "room temperature" (about 300 K) emits primarily infrared radiation. One with a
where T
From
temperature of a few thousand kelvins emits mostly \ isible light. Something with a temperature of a few million kelvins emits most ot its radiation in the X-ray wavelengths.
Wien's law is very useful in computing the surface temperatures of
brightstars because it does not require knowledge of the star's size or
star's elecness; all we need to know is the dominant wavelength of the
tromagnetic radiation.
Studies of blackbody
radiation led to the
discovery that light has
particlelike properties
law and Wien's law describe only two basic propblackbody radiation, the electromagnetic radiation emitted by a
hypothetical blackbody. A more complete picture is given by blackbody
The Stefan-Boltzmann
erties of
curves such as those in Figure 5-3. These curves show the relationship
between the wavelength and the intensity of light emitted by a blackbody
at a given temperature.
The
Figure 5-3
linos
of light
Blackbody curves Three represhown here.
sentative blackbody curves are
Each curve shows
tlw intejisity oj light at every
wavelength emitted by a blackbody at a particular temperature.
measured
i?i
On
microns
this
(fi)
graph, wavelength
where Ifi
10.000 A. The range of wavelengllis of

light
is
is
=
visible
indicated.
Wavelength
The
total
(/i)
area under a blackbody curve
is
proportional to the energy
emitted; the wavelength corresponding to the peak of the curve
dominant wavelength
is
the
Note that the blackbody curves clearly illustrate both of the laws we have discussed: a cool object has a low curve
that peaks at a long wavelength, and a hotter object has a much higher
A.^^^-
curve that peaks at a shorter wavelength.

Figure 5-4 shows how the intensity of sunlight varies with wavelength.
Note that the peak of the curve is at a wavelength of about 0.5 /i. =
5000 A, which is nearly in the middle of the wavelength range for visible
light. The blackbody curve for a temperature of 5800 K is also plotted
in Figure 5-4. Note how closely the observed intensity curve for the Sun
matches the blackbody curve. This close correlation between the observed intensity curves for most stars and the idealized blackbody curves
is the reason that astronomers are interested in the physics of blackbody
radiation.
By the end of the nineteenth century, physicists realized that they had
reached an impasse. All attempts to explain the characteristic shape of
blackbody curves in terms of then-known science had failed.
The breakthrough came in 1900, when the German physicist Max
Planck discovered that he could derive a mathematical formula for the
blackbody curves provided he assumed that electromagnetic energy is
emitted in separate packets of energy. This important assumption was
verified in 1905 by Albert Einstein who explained that light consists of
particlelike packets called photons. The energy carried by a pholon of light is
inversely proportional to the luavelenglh of the light. In other words, longwavelength photons (such as radio waves and microwaves) carry litde
energy, whereas short-wavelength photons (like X rays and gamma rays)
The laws
Figure 5-4
The Sun as a blackbody
of light
This
graph compares
the intensity of sunlight over a

wide range oj wavelengths with the intensity of
radiation
from a blackbody
at a temperature
of
5800 K. Measurements of the Sun's intensity

were made above the Earth's atmosphere. The
Sun mimics a blackbody remarkably well.
Wavelength
((u)
can N much more energy. This relationship between the energy of a photon and its wavelength is called Planck's law.
The energy of a photon is usually expressed in electron volts (eV).
An electron yolt
common unit of
a tiny amount of energ\ especially compared to the

kilowatt-hours in which household electrical energy is
usually measured (a lOOO-watt electric appliance consumes 1 kilow-aitis
hour of energy in 1 hour). Visible photons each carry between 2 and 3

electron yolts. whereas 1 kilowatt-hour equals 2.25 x 10'--^ e\'. Table 5-2
gives the energy carried by photons of various wavelengths.
Planck's law and VVien's law together allow us to relate the temperature of an object to the energy of the photons that it emits, as in Table
5-2. A cool object emits primarily long-wavelength photons that carr\ little energy, and a hot object gi\es off mostly short-wavelength photons
that carry much more energy. In later chapters, we shall see that these
ideas are very important in understanding how stars of various temperatures interact with gas and dust in space.
TABLE
5-2
Some
properties of
Wavelength
Photon energy
(cm)
(eV)
Radio
>10
<io-^
Microwave
10 to 0.01
10"= to
electromagnetic radiation
Infrared
0.01 to 7 X 10"''
Visible
Ultraviolet
Xray
Gamma
Xole: The symbol
>
<0.03
10-'-
0.03 to 30
4100
0.01 to 2
30
2 to 3
4100
to
4 X I0-' to 10-'
3 to 10'
7300
to 3
10"
10' to
X 10"'
to 4
X 10"^
to 10"-'
<io-'
ray
Blackbody
temperature (K)
means "greater than." The symbol
>10''
<
lO'*
to
X 10*
>3 X
7300
X 10*
to 3
10*
X 10"
Tlw laws of
Each chemical element

produces its own unique set
In 1814, the
German master
Figure 4-1). but he subjected the resulting rainbow-colored specTo his surprise. Fraimhofer discovered
that the solar spectrimi contains hundreds of fme dark lines, which became known as spectral lines. Fraunhofer counted over 600 such lines,
and today we know more than 20,000. Hundreds of spectral lines are
visible in the small portion of the Sun's spectrum shown in Figure 5-5.
trum
to intense magnification.
|m;ii{||i|!llt:li(IIH!il
mm
ni
portion of the Sun's
and deep-blue portion of

the solar spectrum (from 3900 A to 4700 A)
is shown here. The G, H, and K Fraunhofer lines are most prominent, but many other
faint spectral lines are also visible. (Mount
Wilson and Las Campanas Obsen'atories)
The
if:
iiffii-ra
V:
ijiniif
I
spectrum
optician Joseph von Fraunhofer repeated

beam of sunlight through a prism
the classic experiment of shining a

(recall
of spectral lines
Figure 5-5
light
violet
Half a century
chemists discovered that they could produce

had long
been identified by the distinctive colors emitted when thev were sprinkled into a flame. About 1857. the German chemist Robert Bunsen invented a special gas burner that produces a clean, colorless flame. This
Bunsen burner became very useful for analyzing substances because its
flame had no color of its own to be confused with the color produced
when a substance is sprinkled into it.
Bunsen's colleague, Prussian-born physicist Gustav Kirchhoff, suggested that light from the colored flames could best be studied bv passing it through a prism (see Figure 5-6). Bunsen and Kirchhoff collaborated in the design and construction of a spectroscope, a device
consisting of a prism and several lenses by which the spectra of flames
could be magnified and examined. Thev promptly discovered that the
spectrum from a flame consists of a pattern of thin bright spectral lines
against a dark background. They next found that each chemical element
later,
spectral lines in laboratory experiments. Certain substances
produces
its ou'ti
characteristic pattern of spectral lines.
Thus was born
in
1859
the technique of spectral analysis, the identification of chemical substances by their unique patterns of spectral lines.
New elements were discovered through spectral analysis. After Bunsen and Kirchhoff had recorded the prominent spectral lines of all the
known elements, they soon began to discover other spectral lines in mineral samples. In 1860, for instance, they found a new line in the blue
portion of the spectrum of mineral water. After chemically isolating the
The Kirchhoff-Bunsen
experiment In the mid-1850s, Kirchhoff and
Bunsen discovered that when a chemical substance is heated and vaporized the resulting
Figure 5-6
spectrum exhibits a
lines.
series
of bright spectral
In addition, they found that each chemi-
cal element produces
tern of spectral lines.
its
oum
characteristic pat-
The lawi of
light
previously unknown element responsible for the line, they named it cesium (from the Latin caesium, '"gray-blue"). The next year, a new spectral
line in the red portion of the spectrum of a mineral sample led them to
the discovery of the element rubidium (from rubidium, "red").
During a solar eclipse in 1868, astronomers found a new spectral line
in light coming from the upper surface of the Sun while the main body
of the Sun was hidden bv the Moon. This line was attributed to a new
element that was named helium (from the tlreek helios. "Sun"). Helium
was not actually discovered on the Earth until 1895, when it was located
in gases obtained from a uranium mineral.
By the earlv 1860s, Kirchhoffs experiments had progressed sufficiently for him to formulate three important statements that todav are
called Kirchhoffs laws of spectral analysis:
Law
plete
A hot, glowing object emits a continuous spectrum

rainbow of colors without any dark spectral lines.
a com-
Law 2 When a continuous spectrum is viewed through a cool gas.

dark spectral lines (called absorption lines) appear in the continuous
spectrum.
Law
3 If the gas is viewed at an angle away from the source of the

continuous spectrum, a pattern of bright spectral lines (called emission
lines) is seen against an otherwise black background.
Figure 5-7 illustrates Kirchhoffs three laws. The bright lines in the emission spectrum of a particular gas occur at exactly the same positions
Emission
Figure 5-7
analysis
Kirchhoffs laws of spectral

glownig nhjert finite a con-
hot,
timwus spectrum. If thu white
light
is
passed
through some cool gas, then dark absorption

lines
appear
same
ga.s
the resulting spectrum. If the
examined from an
oblique angle,
bright emission tines are visible against a
dark background.
line spec
The laws of
light
(wavelengths) as the dark lines in the absorption spectrum of the same

gas.
Somehow, atoms in a gas extract light of specific wavelength from

white light passing through it. Thus, dark absorption lines are created in
the continuous spectrum of the white-light source. The atoms then radiate light of precisely these same wavelengths in all directions, so that an
observer at an oblique angle (without the white-light source in the background) detects bright emission lines.
Why does an atom absorb light only at particular wavelengths? Why
does it then emit light only at these same wavelengths? Traditional theories of electromagnetism could not answer these questions. The answers
came early in the twentieth century with the development of quantum
mechanics and nuclear physics.
An atom
consists of a small,
dense nucleus surrounded

by electrons
The
first important clue about the internal structure of atoms came

from an experiment conducted in 1910 by Ernest Rutherford, a gifted
chemist and physicist from New Zealand. Rutherford and his colleagues
at the University of Manchester in England were investigating the re-
cently discovered
phenomenon of
radioactivity. Certain radioactive ele-
to emit particles. One

type of particle, called an alpha (a) particle, is quite massive (comparable
to four hydrogen atoms) and is emitted from a radioactive substance
with considerable speed.
In one series of experiments, Rutherford and his associates were
ments such
as
uranium and radium were known
using alpha particles as projectiles to probe the structure of solid matter.

directed a beam of alpha particles against a thin sheet of metal
(see Figure 5-8). Almost all the alpha particles passed through the metal
sheet with little or no deflection. To the sinprise of the experimenters,
however, an occasional alpha particle bounced back from the metal sheet
as though it had struck something very dense. Rutherford later remarked, "It was almost as incredible as if you fired a 15-inch shell at a
piece of tissue paper and it came back and hit you."
Rutherford was quick to realize the implications of this experiment.
Most of the mass of an atom is concentrated in a compact, massive lump
of matter that occupies only a small part of the atom's volume. The majority of the alpha particles pass freely through the nearly empty space
that makes up most of the atom, but a very few particles happen to
strike the dense mass at the center and bounce back.
They
Rutherford proposed a new model for the structure of an atom. According to this model, a massive, positively charged nucleus at the center
Figure 5-8
Alpha
Rutherford's experiment
particles
from a
radioactive source are
directed against a thin metal foil.
Most of the
alpha particles pass through the foil with very
little
deflection. Occasionally, however,
alpha particle
thai
it
an
recoils dramatically, indicating
has collided with the massive nucleus of
an atom. This experiment provided

evidence
tliat
atoms have nuclei.
the first
The laws of
Figure 5-9
atom
light
73
Rutherford's model of the
Electrons orbit the atom's nucleus,
which contaiiu most oj the atom's mass. The

nucleus contains two types of particles: protons
and
neutrons.
of the atom is orbited bv tiin negatively charged electrons (see Figure

5-9). Rutherford concluded that at least 99.98 percent of the mass of an
atom is concentrated in a nucleus whose diameter is onlv about one tenthousandth the diameter of the atom.
We know today that the nucleus of an atom contains two tvpes of particles: protons and neutrons. X proton has almost the same mass as a
neutron and each has about 2000 times as much mass as an electron. A
proton has a positive electric charge and a neutron has no electric
chaige. The electrical attraction between the positi\elv charged protons
and the negatively charged electrons holds an atom together.
We also now know that the alpha particle is identical to the nucleus of
a helium atom, consisting of two protons and two neutrons, which happens to be a very stable combination. .Another kind of particle commonly
emitted by radioactive elements, the beta particle, is now known to be a
,
single electron.
Spectral lines are produced

when electrons jump from
one energy
within an
level to
atom
another
The task of reconciling Rutherford's atomic model with the observations

was undertaken by the young Danish phvsicist Niels
Bohr, who joined Rutherford's group at Manchester in 1911.
Bohr began by trying to understand the structure of hvdrogen, the
simplest and lightest of the elements. .\ hydrogen atom consists of a single electron and a single proton. Hydrogen has a simple spectrum consisting of a pattern of lines that begins at 6563.1 .\ and ends at
3645.6 A. The first spectral line is called H, the second H^, the third
H.),, and so forth, ending with H^ at 3645.6 A. The closer vou get to
3645.6 .4, the more spectral lines vou see.
The regularitv in this spcciial pattern had been described mathematically in 1885 bv johann Jacob Balmer, an elderly Swiss schoolteacher.
Balmer used trial and error to discover a fornuila with which the wavelengths of the hvdrogen lines could be calculated. Since his tormula is
successful, the spectral lines of hydrogen at visible wavelengths are today
called Balmer lines and the entire pattern from H,, to H^, is called the
of spectral analysis
Thr laws of
liglil
Balmer lines in the spectrum

193182 This portion of the spectrum
star HD 19B182 shows nearly two
Figure 5-10
ofHD
of the
dozen Balmer
3645.6 A,
lines.
The
just to the
series
left
spectrum also contains the

lines (//
of
converges at
This star's
H.,,,.
first twelve
Balmer
through H12). but they are not
visi-
(Mount
ble in this particular spectrogram.
Wilson and Las Campanas Observatories)
Balmer
The spectrum
HD
193182 shown in Figure

from H,;, through H4(|.
Bohr realized that, if he were to succeed in understanding the structure of the hydrogen atom, he should be able to derive Balmer's formula
directly from basic laws of physics. He began by assuming that the electron in a hydrogen atom orbits the nucleus only in certain specific orbits.
As shown in Figure 5-11, it is customary to label these orbits n = \, n =
2, n = 3, and so on. They are called the Bohr orbits.
For an electron to jump from one Bohr orbit to another, the hydrogen atom must gain or lose a specific amount of energy. For an electron
to go from an inner orbit to an outer one, the atom must absorb energy.
Conversely, if the electron falls from a high orbit to one of the inner
orbits, energy is released by the atom.
The energy gained or released by the atom when the electron jumps
from one orbit to another is the difference in energy between these two
orbits. According to Planck and Einstein, the packet of energy gained or
released is a photon whose energy is inversely proportional to its wavelength. Using these ideas, Bohr mathematicallv derived the formula that
Balmer had discovered by trial and error. Furthermore, Bohr's discovery
elucidated the meaning of the Balmer series: all the Balmer lines are
produced by electron transitions between the second Bohr orbit and
higher orbits ( = 3, 4, 5, and so forth).
In addition to giving the wavelengths of the Balmer series, Bohr's formula correctly predicts the wavelengths of other series of spectral lines
that occur at nonvisible wavelengths. For example, spectral lines occur
because of electron transitions between the lowest Bohr orbit (n = 1) and
all higher orbits (n = 2, 3, 4,
.). These transitions produce the Lyman
series which is entirely in the ultraviolet. This pattern of spectral lines
begins with L ("Lyman alpha") at 1215 A and converges on Lx at
series.
5-10 exhibits
of the star
more than two dozen Balmer
The Bohr model of the
Figure 5-11
An
hydrogen atom
cleus
allowed
(II
photon
tron
is
clirtron circles the nu-
orbits n
jumps from an inner
orbit.
photon
electron falls
sorptions
I, 2. 3,
and
absorbed by the atom as the
and
is
orbit to
so on.
elec-
an outer
emitted by the atom as the
down
to
a low
orbit.
These ab-
emissions of photons occur only
at specific wavelengths, thereby
producing
charactenstic patterns of lines in the
hydrogen spectrum.
lines
The laws of
Figure 5-12
hydrogen
IS
The
energy-level
The \lriulure
light
diagram of
of Ihe hviiugfii
alom
conveniently displayed tn a diagram showing
the allowed energy

slate.
tions,
are shown
lei>els
above Ihe ground
vanety of electron jumps, or transi-
including
some of the most prominent

gen spectrum.
those thai produce
lines in the hydro-
912 A. At infrared wavelengths is the Paschen series, which arises out of

transitions to and from the third Bohr orbit (n = 3). It begins with P^
("Paschen alpha") at 18.751 A and converges on P^ at 8206 A. Addistill longer wavelengths.
Bohr's ideas give us an explanation for Kirchhoffs laws. Each spectral
line corresponds to one specific transition between the orbits of the
atoms of a particular element. .\n absorption line is created when an
electron jumps from an inner orbit to an outer orbit, extracting the required photon from an outside source of energv such as the continuous
spectrum of a hot. glowing object. .Vn emission line is produced when
the electron falls back down to a lower orbit and gi\es up a photon.
Physicists today retain many features of the Bohr model of the atom, although they no longer picture electrons moving in specific orbits about
the nucleus. Instead, electrons are now said to occup\ certain allowed
energy levels in the atom. .An extremeU useful wav of displaving the
structure of an atom is with an energy level diagram, such as that
shown in Figure 5-12 for hydrogen. The lowest energ\ level, called the
tional series exist at
ground state, corresponds to the = 1 Bohr orbit. .\n electron can jump
= 2 level onlv if the atom absorbs a
from the ground state up to the
Lyman-alpha photon of wavelength 1216 A. As noted earlier, the energy
of a photon is usuallv expressed in electron volts (e\'). The Lvman-alpha
photon has an energv of 10.19 eV, so the energ\ level = 2 is shown on
the diagram as having an energ\- 10.19 e\' above the energv of the
ground state, conventionally assigned a value of eV. Similarlv, the n =
3 level is 12.07 eV above the ground state, and so forth up to the n = y^
level at 13.6 eV. If the atom absorbs a photon of an energv greater than
13.6 eV, an electron from the groimd state will be knocked completclv
out of the atom. This process, in which high- energv photons knock electrons out of atoms is called ionization. In general, an atom that has been
stripped of one or more electrons is called an ion.
The atoms of heavier elements have more complex energy-level diagrams. For example. Figure 5-13 shows the energ\ -level diagram of soit
/;
dium. .Astronomers find
it useful to lefcr to such diagrams to identifv

the spectral lines they observe in the spectra ol siars and ncl)ulae.
With the work of people like Planck, Finstein. Rutherford, and Bohr,
the interchange between astronomv and phvsics came full circle. Modern
physics was hoi n when Newton set oin to understand the motions of the
The laws of
76
The energy-level diagram of
Figure 5-13
sodium
light
Complex aloim have complicated
The energy-level
energy-level diagrams.
gram of sodium
IS
dia-
displayed here, along with
the wavelengths nf photons absorbed or emitted

in
some of
the
major electron
visible wavelengtlis, the
transitions.
sodium spectrum
dominated by two very strong
lines at
5889.9 A and 5895.9 A. These two

called the sodium
they correspond
to
lines,
At
is
lines,
are strong because
two transitions that are the
primary avenue through which electrons cascade from high

ground slate.
levels
doum
to the
planets. Two and a half centuries later, physicists in their laboratories

had probed the properties of light and the structures of atoms. The
fruits of their labors would have immediate astronomical applications.
Summary
Electromagnetic radiation (including visible light) has wavelike properties

travels through empty space at the constant speed r = 3 x 10** m/sec.
and
The Stefan Boltzmann

rate at
which
it
law relates the temperature of an object to the
radiates energy.
Wien's law relates the temperature of an object to the dominant wavelength at which it radiates energy.
blackbodv is a hvpothetical object that is a perfect absorber of electromagnetic radiation at all wavelengths. Since a blackbodv does not reflect
any light from outside sources, the radiation that it does emit depends
only on its absolute temperature. Stars closely approximate the behavior
of blackbodies.
The
of radiation emitted at various wavelengths

temperature are shown by a blackbodv curve.
intensities
at a given
The shape
b\ a
blackbodv
of blackbody curves can be explained only by the assumption

These particles of light are called
that light has particlelike properties.
photons.
Planck's law relates the energy of a photon to its wa\elength; the energy
of a photon is inversely proportional to its wavelength.
Kirchhoff s three laws of spectral analysis describe the conditions under

which absorption lines, emission lines, or a continuous spectrum can be
observed.
The laws of
light
consists of a small, dense nucleus (composed of protons and

neutrons) surrounded bv electrons that occupy only certain allowed energy levels.
An atom
When
an electron jumps from one energy level to another, it emits or

absorbs a photon of appropriate energy (and hence of a specific wavelength).
The
correspond
spectral lines of a particular element
tron transitions between allowed energy levels in
its
to the various elecatoms.
of hydrogen at visible wavelengths consists of the Balmer

which arises from electron transitions between the second energy
level of the hydrogen atom and higher levels. The Lyman series in the
far-ultraviolet is produced by jumps between the ground state (lowest
level) and higher levels; the Paschen series in the infrared is produced by
jumps between the third energy level and higher levels.
The spectrum
series,
and
process in which an atom absorbs an energetic photon

electron entirely is called ionization.
The
Review questions
What
is
blackbody?
a blackbody?
If stars
What does
behave
it
mean
like blackbodies,
2 What is the Stefan-Boltzmann law?

mers are interested in it?
What
is
Wien's law?
Of what
to say that a star
why
Why do
use might
it
loses
behaves
an
like a
are they not black?
you suppose that astrono-
be to astronomers?
4 Using Wien's law and the Stefan-Boltzmann law. explain the color
changes that are observed as the temperature of a hot, glowing object
in-
creases.
5 Describe the experimental evidence that supports our current views of the
structure of an atom.
6 Describe the spectrum of hydrogen and explain how the spectrum

hydrogen atom.
is
re-
lated to the structure of the
Advanced questions
I"
7 Approximately
travel in
"
how manv
times around the world could a
beam of
8 Imagine a star the same size as the Sun, but with a surface temperature
twice that of the Sun. At what wavelength does that star emit most of its radiation? How many times brighter than the Sun is that star?
9 Imagine a star the same size as the Sun with a surface temperature of
2900 K. Suppose both the Sun and this star were located at the same distance from you. Which would be brighter? By how much?
10 Describe how the spectrum of a helium atom might appear
two electrons were stripped off.
light
one second?
if
one of
its
11 Compare chemical identification based on spectral line patterns with the

identification of people by line patterns in their fingerprints.
12 Suppose you look

stars with
up
at
your naked eye.
which are cool? Explain.
the night sky and observe some of the brightest

there anv wa\ of telling which stars are lioi an<t
Is
77,f
For further reading
Cline, B. Meti
Wlw Made
coveries that led la
a A'cic Physics. Signet, 1965.
.4
liisloi-y
/i,^
nf llghl
of thr dis-
our modern understanding of radiation and
the
atom.
Gingerich, O. "Unlocking the Chemical Secrets of the Cx)smos.'" Sky is

Telescope. )ulv 1981, p. 13.
van Heel. A., and' Velzel, C. What Is Light? McGraw-Hill. 1968.
R. "Extending the Visible Frontier: New Tools
Astronomer." Mercury, Sept. /Oct. 1975, p. 2.
Weymann,
oi the Optical
6 Our solar system
Nebulae
in
Sagittarius
Plunvh die jnolnihh
j(inniu)^ alunir willi tinr itais

III
Saifillinnis,
The
Iti
these nebulae
type of planet that
a puttindnr distance
[mm
fotnn
at
a star depends on
such conditions as the temperature there and

the substances
and gases
tem, planets
rock fragments, ice crystals,
at that distance. In the solar s\s-
composed primarily of rock formed
near the Sun, buts

gases.
Far from
the
its
heat drove off
ices
and
Sun, where temperatures
are low, planets retained volatile substances,

resulting in worlds composed primarily of gas.
(Royal Obseri'atory, Edinburgh)
Mi^deni iii.\lniineiits uml space probes have given us a rich body of iuformulion about the solar .syslciii. Wr begin this chapter with a sun'cy of the major
physical characteristics of the planets. We find that they fall into two distinct
classes: the inner (terrestrial) planets and the outer (Jovian) planets, with
Pluto an oddity that cannot be easily classified. We then discu-ts the materials
of which the playlets are composed and the technique oj spectro.scopy that provides information about the chemical compositions of distant objects. Finally,
we outline current theories about the origin of the elements, the solar system,
the planets,
and
the Suti.
Looking up at the heavens and wondering about the nature and origin
of the Sun. Moon, stars, and planets is a universal lunnan experience.
Unlike our ancestors, however, we possess a wealth of information. Especially within the past few decades, telescopic observations and interplanetary spacecraft have given us vast quantities of data from which to formulate and test theories. Ihe Sun, jjlanets. moons, asteroids, comets,
and meteoroids that make up oin tiny niche in the luiiverse are finally
accessible. Many of these objects are exceedingly ancient and contain records of the cosmic events that created our solar system.
In addition to gleaning information from the objects that orbit the
Sun, we can observe active star formation occurring elsewhere in our
Our
solar system
and planetary systems are now being formed in many beaunebulae scattered across the heavens. A general understanding of
star creation coupled with knowledge of the Sun and its satellites gives
us a reasonably comprehensive picture of how the solar system was created. Many details still need to be worked out, but the overall scenario
seems remarkably sound. For the first time, we can truly appreciate what
is imique and what is commonplace about our world. We have begun to
fathom our connection with the rest of the cosmos.
galaxy. Stars
tiful
The planets are
terrestrial or
notice the striking dichotomy in the orbits of the planets as shown

Figure 6-1. The orbits of the four inner planets (Mercury, Venus.
Earth, and Mars) are crowded close to the Sun. In contrast, the orbits of
the next four planets (Jupiter, Saturn, Uranus, and Neptune) are widely
spaced at great distances from the Sun.
As you might expect, the range of surface temperatures that each
planet experiences is related to its distance from the Sun. The four
inner planets are quite warm. For example, noontime temperatures on
Mercury climb to (500 K (=327C = 62 TF). At noon in the middle of
classified as
Jovian by their
in
physical attributes
Figure 6-1
The solar system
drawing shows
orbits
around
crowded
ets
orbiting at
the
Sun.
This scale
the distribution of planetary
the Sun.
close to the
much
brief overview of the solar system distinguishes two classes of planets.
First,
Four inner planets are
Sun, with five outer plangreater distances from
summer on
Mars,
it
is
sometimes
as
warm as 300 K = 27C = 81F). Of

much less solar radiation, are
(
course, the outer planets, which receive
much cooler. Typical temperatures range from about 150 K

= -123C = -189F) in Jupiter's cloudtops to 63 K = -210C =
(
Our
solar s\ilem
346F) on Neptune. Temperature plavs a major role in determining

whether substances exist as solids, hquids, or gases, thus profoundly affecting the appearance of the planets.
Most of the planets' orbits are nearly circular. As discussed in Chapter
3. Kepler discovered that these orbits are aciualh ellipses. Astronomers
denote the elongation of an ellipse by its eccentricity (the eccentricity of
a circle is zero). Most planets have orbital eccentricities that are very
close to zero. The notable exception is Pluto, with an orbital eccentricity
of 0.25. Because of its highly noncircular orbit, Pluto is sometimes closer
to the
Sun than Neptune
is.
The
planetary orbits all lie in nearlv the same plane. In other words,
the orbits of the planets are inclined at only small angles to the plane of
the ecliptic. Again, however, Pluto is a notable exception. The plane of
Pluto's orbit is tilted at 17 to the plane of the Earth's orbit (see Table
6-1).
As we compare the physical properties of the planets, we again fmd

the four inner planets and the
fall naturally into two groups
four outer planets
with Pluto once more an exception. We shall now
look at the important properties of size, mass, and densitv.
that they
The
four inner planets are quite small. The Earth, with a diameter of
(7930 miles) is the largest. In sharp contrast, the four outer
planets are much larger. First place goes to Jupiter, whose equatorial
diameter is 143,800 km (89,400 miles). Pluto, despite its position as the
outermost planet, is even smaller than the inner planets. Its diameter is
onlv about 3000 km (1900 miles), roughly the same size as our Moon.
12,760
km
Our
TABLE
6-1
solar system
Orbital characteristics of the planets
Mean
distance from Sun
(AU)
Planet
km)
(10"*
Inclination to
Orbital (sidereal) period

(years)
the ecliptic
Eccentricity
(degrees)
Mercury
0.39
58
0.24
0.206
7.0
Venus
0.72
108
0.62
0.007
3.4
Earth
1.00
150
1.00
0.017
0.0
Mars
1.52
228
1.88
0.093
1.8
Jupiter
5.20
778
11.86
0.048
1.3
Saturn
9.54
1427
29.46
0.056
2.5
Uranus
19.18
2870
84.01
0.047
0.8
Neptune
30.06
4497
164.79
0.009
1.8
Pluto
39.44
5900
247.7
0.250
17.2
Figure 6-2 shows the Sun and the planets drawn to the same scale. The
diameters of the planets are given in Table 6-2.
Determining the mass of a planet is a more difficult piocess. It is
most easily accomplished if the planet has a satellite. The satellite's orbit
obeys Newtonian mechanics, which relate the orbit to the planet's mass.
Astronomers can observe the satellite and measure its orbital period and
semimajor axis. With this information, they can calculate the planet's
mass from the formulas of Newtonian mechanics.
If the planet does not have a satellite, astronomers must rely on a
comet or spacecraft that passes near the planet. The planet's gravity,
which is directly related to its mass, deflects the path of the comet or
spacecraft. By measuring the size of this deflection and using Newtonian
mechanics, astronomers can determine the planet's mass. The four inner
planets have low masses, the next four planets have substantially larger
The Sun and the planets
Figure 6-2
drawing shows
disk of the
the
same
scale.
hih iitarrsi Ihr

nil, I
Mm
ni'\l jiiui
si
nil
This
the nine planets in front oj the
Sun, with
all ten bodies
drawn
to
The four planets that have orSun (MiKun: Venus. Earth.

uiiiill
anil
iiiiiile
of rock. The
Uranus,
plinifis (Jiipiln. .Snliirn,
Neptune) are large and composed primarily

of gas.
Neptune
Our
TABLE
6-2
solar system
Physical characteristics
Mass
Diameter
of the planets
Planet
(km)
(Earth
1)
(gm)
Average density
(Earth
1)
(g/cm^)
4,880
0.38
3.3 X 10-"
0.06
5.4
Venus
12,100
0.95
4.9
X 10"
0.82
5.2
Earth
12,760
1.00
6.0
10^'
1.00
5.5
Mars
6,800
0.53
6.4
lO^**
0.11
3.9
Jupiter
143,800
11.27
1.9
X 10
317.89
1.3
Saturn
120,000
9.44
5.7
lO^**
95.15
0.7
Uranus
52,300
4.10
8.7
10^*
14.54
1.2
Neptune
49,500
3.88
1.0
X 10^^
17.23
1.7
Mercury
Pluto*
3,000
0.2
10^^
0.002
*Atl of the data for Pluto are somewhat uncettani.
masses. Again, Jupiter is first, with a mass 318 times greater than the
mass of the Earth. The masses of the planets are given in Table 6-2.
Average density (mass divided by volume) is a physical property that

can often be used to deduce important information about the composition of an object. Scientists commonly express average density in grams
per cubic centimeter, as in Table 6-2. The four inner planets have large
average densities. The average density of the Earth, 5.5 g/cm\ may not
seem large until you know that the average density of a tvpical rock is
only about 3 g/cm' and the average density of water is 1 g/cm\ The
Earth must therefore contain a large amoimt of material that is more
dense than rock. This information provides our first clue that Earth-like
planets have iron cores.
In sharp contrast, the outer planets have very low densities. Indeed,
Saturn is less dense than water. This strongly suggests that the giant
outer planets are composed primarily of such light elements as hydrogen
and helium. Again, Pluto is an exception. Although it is even smaller
than the dense inner planets, its average density seems to be much like
those of the giant outer planets. Table 6-2 lists the average densities of
the planets.
These differences in size, mass, and density suggest that we shotild

consider the four inner and four outer planets as two distinct groups.
The four inner planets are called terrestrial planets because thev resemble the Earth. They are small and dense, with craters, canyons, and volcanoes being common on their hard, rock\ siulaccs (see Figure 6-3).
The four outer planets are called Jovian planets because they resemble
Jupiter (^Jove was another name for the Roman god Jupiter). Vast swirling cloud formations dominate the appearance of these enormous gaseous spheres (see Figure 6-4). The solid cores of these planets are probably not much bigger than the Earth, buried beneath atmospheres that
are tens of thousands of kilometers thick.
.Although it is usually called a planet, Pluto clearly is an ()<klii\ Its
.
Figure 6-S
terrestrial planet
Mars
typical tern'strial /jlaufl. lis duimeter
6800 km and
il
3.9 glcm^, indicating that
it
is
and
only
photograph taken
spacecraft in 7976. (iV,4A'.4;
composed
craters
lite
Some astronomers have suggested that Pluto should be classified

many thou,sands of small objects (typical diameter
about 40 km) that circle the Sun between Mars and Jupitci The largest
planets.
is
has an ni'cra^c den.sil\ o/
rock. Volcanoes, canyoirs,

.seen in this
physical properties are not tvpical of either the terrestrial oi the Jovian
is
oj
can he
Viking 2
with the asteroids,
750 km.
Another oddity, an object called Clhiron, was discovered in 1977. It
similar to a large asteroid and moves in a highly eccentric orbit about
asteroid, Ceres, has a diameter of about
is
Our
A Jovian
Figure 6-4
largest of the
diameter
is
planet Jupiter
Jovian planets.
143,800 km.
Its
Its
primarily composed of light elements.
this
the
is
equatorial
average density
only 1 .3 g/cm^, indicating that the planet
Jupiter's moons, lo
and Europa, are
view taken by the Voyager
solar system
Two
is
is
of
seen in
1 spacecraft in
1979. Each of these satellites is approximately

same size as Earth's moon. (NASA)
the
Sun between Saturn and Uranus. Pluto and Chiron should perhaps
be grouped with the asteroids as minor members of the solar system,
leaving only eight planets. On the other hand, some astronomers argue
that at least seven additional objects should be included as major members ot the solar system. These seven objects are large moons that orbit
four of the planets.
the
Seven large
resemble
satellites
terrestrial planets
All the planets except
dozen
Mercury and Venus have satellites. More than

^Jupiter and Saturn each have more than a
and certainly dozens of others await discovery. The known
forty satellites are
known
satel-
There are seven giant satellites

(listed in Table 6-3) that are roughly comparable to Mercury in size. The
other satellites are much smaller, with diameters of less than 2000 km.
The recent interplanetary missions of Voyager 1 and 2 have revealed
lites fall
many
into two distinct categories.
fascinating characteristics of the giant satellites (see Figure 6-5).

For example, Jupiter's satellite lo is one of the most geologically active
worlds in the solar system, with numerous volcanoes belching sulfur-rich
compounds. Saturn's largest satellite. Titan, has an atmosphere nearly
twice as dense as Earth's.
Because of their sizes, it is reasonable to include these seven giant satellites in a listing of the planets, even though they happen to orbit other
planets instead of orbiting the Sun independently. Their hard surfaces
(of either rock or ice) would place them in the terrestrial category. This
expanded definition of a terrestrial planet gives us ten Earth-like worlds
to compare with our own planet.
Our
TABLE
The seven giant
6-3
satellites
Satellite
Parent
planet
Diameter
(km)
Average density
name
Moon
Earth
3480
3.3
lo
Jupiter
3630
3.6
Europa
Jupiter
3130
3.0
Ganymede
Jupiter
5280
1.9
Callisto
Jupiter
4820
1.8
Titan
Saturn
5120
1.9
Triton*
Neptune
4000
2.0
*The data for Triton are
Figure 6-5
giant satellite
taken from Vimiger
in
85
solar s\slrm
(g/cm^)
,"
This view,
1979, shows
Callislo.
moon of Jupiter. Its diameter is almost exactly the same as Mercury's.

Numerous craters pockmark Callisto's icethe second largest
bound
surface.
(NASA)
Spectroscopy reveals the

chemical composition of the
Sun, planets, and stars
The most
accurate inforrnalion about the composition of the planets

that have landed on planets and made direct
comes from spacecraft
chemical analyses of the atmosphere or soil. Unfortunately, we have obtained such direct information for only three worlds: Venus, the Moon,
and Mars. In all other cases, astronomers must rely on their ability to
analyze sunlight reflected from the distant planets and their satellites. In
these circumstances, astionomers bring to bear one of their most powerful tools, sfH'ctrosropy.
As we learned in Chapter b, the (ierman physicist Joseph \()n Fraunhofer examined the Sun's spectrum under high magnification and found
numerous faint dark lines cutting across the spectrum. In other words,
certain specific wavelengths are missing from sunlight. Figure 6-6 shows
a portion of the solar spectrum, including a luimber of these spectral
lines. Spectroscopy is the systematic study of such spectra.
Oiii siilnr system
Iron on the Sun
Figure 6-6
n
from 4200
spectrum
The upper
a portion of the Suit's spectrum

to
4300
tral lines are visible.
A.
Numerous dark
spec-
The lower spectrum
is
corresponding portion of the spectrum of vaporized iron. Several bright spectral lines can
be seen against the black background.
The fact
some of
some iron
that the iron lines coincide with
solar lines proves that there

(albeit
a very
tiny
is
amount) in
the
the Sun's
(Mount Wilson and Las

Campanas Observalones)
atmosphere.
Figure 6-7
prism spectrograph
optical device uses a prism to break

light
of an
object into a spectrum
up
llir
the
and foni
that spectrum onto a photographic plate.
The significance of Fraunhofer's discovery was revealed in experiments by Bunsen and Kirchhoff in the mid-nineteenth century: each
chemical produces its own characteristic pattern of spectral lines. For example, a
portion of the spectrum of vaporized iron is shown in Figure 6-6. No
other chemical can mimic this particular pattern of lines at these wavelengths.
It is
iron's
own
distinctive "fingerprint."
iron appear in the Sun's spectrum,
The
spectral lines of
and thus some vaporized iron must
atmosphere.
astronomers have used spectrographs to examine and
record the spectra of stars and galaxies. These machines have become
one of the astronomer's most important tools, second only perhaps to
the telescope itself. In its basic form, a spectrograph consists of a slit,
two lenses, and a prism arranged to focus the spectrum of an astronomical object onto a small photographic plate, as shown in Figure 6-7. This
optical device typically is mounted at the focal point of a telescope, and
the image of the object to be examined is focused on the slit. After the
spectrum of a star or galaxy has been photographed, the exposed portion of the photographic plate is covered and light from a known source
(usually an iron arc) is focused on the slit. This exposes a "comparison
spectrum" above and below the spectrum of the object being examined,
as in Figure 6-8. The wavelengths of the bright spectral lines in the comparison spectrum are already known from laboratory experiments. These
bright lines can therefore be used to measure the wavelengths of the
lines in the spectrum of the star or galaxy under study.
There are drawbacks to this old-fashioned spectrograph. A prism does
not disperse the colors of the rainbow evenly. The red part of the spectrum is compressed, while the blue and violet colors are more spread
out. In addition, the blue and violet wavelengths must pass through a
substantial thickness of the prism, which absorbs much of the starlight.
Indeed, a glass prism is opaque to near-ultraviolet wavelengths.
exist in the Sun's
Over the
years,
Oiii
Figure 6-8
frrapliir
spectrogram
record / a spectrum
s<ila)
system
The phaUiis
collect
spectrogram. This spectrogram shows the ^pn
Irum of a star. ,\umerotLi dark absorplioii

lines can be seen against the brighter back-
ground of
star's
an
the spectrum.
Above and behne
spectrum are emission
lines
the
produced In
iron arc. These bright lines are the compar-
(Palomar Obsen'atonl
ison spectrum.
A better dexice for breaking starlight into the colors of the rainbow is
a diffraction grating, a piece of glass on which thousands of closely
spaced lines are cut. Some of the finest diffraction gratings have as many
as 10.000 lines per centimeter. These lines are usually cut bv drawing a
diamond back and
lines
The
forth across the piece of glass.
spacing of the
must be verv regular.

diffraction grating can be used
in either of two ways. Incoming

starlight can be reflected off the grating (in which case it is called a reflection grating), or the starlight can be passed through the grating
(ulien it is called a transmission grating). In either case, light rays leaving various parts of the grating interfere with each other so as to pro-
duce a spectrum. Figure 6-9 shows the design of
modern
grating spec-
trograph.
In recent years, the television
and
electronics industries have pro-
duced
a variety of light-sensitive devices that are signiricanily better than

photographic film for recording spectra. For example, many observatories now record spectra with a new invention called a charge-coupled
device (CCD). A CCD is a silicon chip appioximatelv the size of a postage stamp dixided into thousands of tin\ squares. When light falls on
one of these tinv squaics (usually called a "pixel"), an electric charge
builds up in proportion to the amount of light. When the exposure is
finished, electronic
lated in each pixel.
equipment ineasures how much charge has accumuThe final result is a graph on which light intensity is
plotted against wavelength. Absorption lines (dark lines in the rainbow-
figure 6-9
optical dei'ice
up
grating spectrograph
m.sc.s
the light of
an
This
re/leclion grating In liimh
object into
a spectrum and
focus that .spectrum onto a photographic plate.
Our
A CCD
Figure 6-10
solar system
spectrum of Pluto
This graph shows the spectrum of Pluto as
re-
corded by a cliarge-coupled deince (CCD). The
prominent absorption near 9000

met)iane
(CH4} and
is
that substance on the planet.

tions
is
due
to
conclusive evidence of
(From ohsenm-
bij. Apt, N. P. Carleton, and
C. D. Mackay)
7000
Wavelength (A)
colored spectrum) appear as depressions or valleys in the background

light. Emission lines (bright lines in the spectrum) appear as peaks rising
above the background. A CCD spectrum is shown in Figure 6-10.
Spectral lines provide extremely reliable evidence about the chemical
composition of distant objects. If the spectral lines of a certain chemical
are found in the spectrum of a star or planet, it is safe to conclude that
this chemical is present in that object. For example, the spectrum of
Pluto in Figure 6-10 shows absorption due to methane, clearly indicating
the presence of that cheinical (probably as methane ice because of Pluto's
low teinperature) on that planet's surface.
All
substances are composed

elements that
of chemical
are
made
of
atoms
As data from spectrographs accumulated, an important fact emerged.

No matter how far we peer into space, we always find the saine naturally
occurring chemical elements of which the Earth is made. The elements
are fundamental substances because it is impossible to break them down
into more-basic chemicals. Each element is composed of only a single
kind of atom.
A list of the chemical elements is most conveniently displayed in the
form of a periodic table (see Figure 6-11). Each element is assigned a
unit|ue atomic number and the elements are arranged in the periodic
table in that sequence. With a few exceptions, this sequence also corresponds to increasing average mass of the atoms of the elements. Thus
hydrogen (the symbol H) with atomic number 1 is the lightest element.
Iron (Fe) has atomic number 26 and is a relatively heavy element. All
the elements in a single vertical column of the periodic table have similar
chemical properties. For exariiple, the elements in the far right column
are all gases at Earth-surface temperature and pressure, and they all
tend to be very reluctant to react chemically with other eleinents.
In addition to nearly 100 naturally occurring elements, Figme 6-11
lists several artificially produced elements. All of them are heavier than
uranium (the symbol U) and are highly radioactive, which means that
they decay into lighter elements within a short time after being created
in the laboratory.
The atom is the smallest possible piece of an element. Typical atomic

diameters are about 1 A. As we saw in Chapter 5 (see Figure 5-9), laboratory experiments in the early twentieth century revealed that most of
the mass of an atom is concentrated in a dense nucleus composed of
Our
The periodic table of the

Figure 6-11
elements The periodic table is a convemenl
of the elements, arranged according
their weights and chemical properties.
listing
to
solar system
2
He
H
6
10
Ne
15
P
16
17
18
CI
33
As
34
Se
35
36
Br
Kr
54
Xe
Li
4
Be
12
14
Na
Mg
13
Al
19
20
Ca
21
22
23
24
25
Sc
Ti
Cr
Mn
37
38
39
40
41
42
Rb
Sr
Zr
Nb
Mo
43
Tc
57
^72
La
Hf
55
Cs
87
Fr
56
Ba
88
Ra
z ^-
26
Fe
27
28
29
32
Ni
Cu
30
Zn
31
Co
Ga
Ge
44
45
46
47
48
49
Pd
Ag
Cd
In
Sb
52
Te
53
Rh
50
Sn
51
Ru
77
78
79
81
84
Pt
Au
TI
82
Pb
83
Ir
80
Hg
Bi
Po
85
At
63
Eu
64
65
66
67
68
69
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
95
96
97
Bk
98
99
100
101
102
103
Es
Fm
Md
No
Lr
76
OS
73
Ta
74
75
Re
105
106
58
Ge
59
60
61
62
Pr
Nd
Pm
Sm
90
Th
91
Pa
92
93
94
Pu
Si
Np
Am
Cm
Cf
70
86
Rn
71
protons and neutrons. Orbiting this massive nucleus are the electrons.
attractive electrical forces between the positivelv charged protons
The
and negativelv charged electrons keep the atom from coming apart.
Although electrons and protons have equal but opposite charges, they
have unequal masses. The electron is one of the lightest subatomic parti10"-** grams. The proton is nearly 2000
cles, having a mass of only 9.1 x
times heavier: its mass is 1.7 x 10~-'^ grams.
A neutron has almost exactlv the same mass as a proton. As the name
it is electrically neutral. Neusuggests, a neutron has no electric charge
trons serve as buffers between the positively charged protons crowded
together inside the nucleus. The number of neutrons in a nucleus typiis larger than the number of protons, especiallv in the heaviest ele-
cally
ments.
Normallv. the number of electrons orbiting an atom is equal to the
number of protons in the nucleus, making the atom electrically neutral.
Furthermore, the number of protons in an atom's nucleus equals the
atomic number for that particular element. Thus, a hydrogen nucleus
has one proton, a helium nucleus has two, and so forth, up to uranium
with 92 protons in its nucleus.
The number of protons in its nucleus determines what element that
atom is. Nevertheless, the same element may have slightly varving numbers of neutrons in its nuclei. Oxygen, the eighth element on the periodic table, has an atomic number of 8, and every oxvgen nucleus has
exactlv eight protons, but it can have eight, nine, or ten neutrons. These
three slightlv different kinds of oxygen are called isotopes. The isotope
with eight neutrons, written as "Ô or oxygen- 16, is b\ far the most
abundant variety. The rarer isotopes with nine and ten neutrons are designated as ''O and '"O, respectively.
The superscript that precedes the chemical symbol for an element is
equal to the total nimiber of protons and neutrons in the nucleus of that
particular isotope. For example, the common isotope of iron is ''''Fe, or
iron-56, which means that its nucleus contains a total of 56 protons and
neutrons. From the periodic table we see, however, that the atomic number of iron is 26. This number shows that every iron atom has 26 protons in its nucleus. Therefore, the number of neutrons in a "''Fe nucleus
56 - 26 = 30.
Although the number of elements is limited, a wide variety of substances exists in the universe because atoms can combine to form various
is
Ou)
Figure 6-12
Carbon dioxide on Venus
This near-infrared spectrum oj
thai
Venm
of spectral lines beginning at
series
is
solar system
shows
<
8689 A
caused by carbon dioxide, the primary
constituent of Venus's atmosphere.
(Lick Observatory)
8689
molecules. For example, two iiyclrogen atoms can combine with an atom
of oxygen to form a molecule of water. Water's chemical formula, HÔ,
describes the atoms in its molecule. In a similar way, two oxygen atoms
can bond with a carbon atom to produce a molecule of carbon dioxide,
whose formula is COq. The chemical formula for common table salt is
NaCl, which means that a salt molecule contains a sodiinn atom combined with a chlorine atom.
Just as each element can be identified by its spectral lines, molecules
also produce unique patterns of spectral lines in the spectra of astronomical objects. For instance. Figure 6-12 shows a portion of the spectrum of Venus, with a series of spectral lines caused by carbon dioxide.
It is thus reasonable to conclude that Venus's atmosphere contains this
gas. Indeed, direct measurements by Russian and American spacecraft
confirm that carbon dioxide gas forms 97 percent of the Venusian at-
mosphere.
The
relative
abundance
of
the elements is the result of

cosmic processes
Some elements
far the
mass of
are very
common,
others are quite rare.
Hydrogen
is
by
most abundant substance, making up nearly three-quarters of the

the stars
all
and
galaxies in the uni\erse.
the second most abundant element. Together hydrogen and

helimn account for 98 percent of the mass of all the material in the uni-
Helium
verse
is
leaving only 2 percent for
the other elements combined.
all
There is a good reason for this overwhelming abundance of hydrogen

and helium. Most astronomers believe that the universe began roughly
20 billion years ago with a violent event called the Big Bang. Only the
two lightest elements, hydrogen and helium, emerged from the enormously high temperatmes following this cosinic event. All the other elements were manufactured deep inside stars at later times. If it were not
for the nuclear reactions inside stars, there would be no heavy elements
in the universe today. In Chapters 13 and 14, we shall explore the processes by
which elements are created
Near the ends of
at a star's center.
their lives, stars cast
much of
their matter out
one in which a
outer layers are gradually expelled. Figure 6-13 shows the star
65750, which is losing material in this fashion. Alternatively, a star
may end its life with a spectacular detonation called a supernova explosion, which blows the star apart. Either way, the interstellar gases in the
galaxy become enriched with heavy elements dredged up from the dying
star's interior, where they were created. New stars that form out of this
enriched material thus have an ample supply of heavy elements to form
into space. This process can be a coinparativeh' gentle

star's
HD
a system of planets, satellites, comets,
Our
billions
solar system
and
asteroids.
made of matter created in stars that disappeared

The Sun is a fairly young star, only 5 billion years
is
of years ago.
elements other than hydrogen and helium in our solar
old. All the
sys-
Oil I solar syslem
Figure 6-IB
HD
65750.
nebulosity
IS
mass-loss star
shatiliiifr
around
Ike star
light reflected frinii dust
This skii.
malrrml Kijndh.
is
caused In
t'hr
star-
grams. These grains
may have condensed from
the material shed h\
the star. (.Anglo-.\iistialian Ohseivatoiy)
tein
were created and
cast off
years of our galaxy's existence.
by ancient stars during
We
are literally
made
tlie first
fO billion
of star dust.
Stars create the different elements in different amounts. For example,

the elements carbon, oxygen, silicon, and iron are readily produced in a
whereas gold is created only under special circumstances.

Thus, gold is rare in the solar system but carbon is not.
A convenient way to express the abundance of the various elements is
to say how many atoms of a particular element are found for every trillion (that is, 10'-) hydrogen atoms in space. For example, for every trillion hydrogen atoms there are about 60 billion (6 x 10'") helium atoms.
star's interior,
From chemical
analysis of Farth rocks. Moon rocks, and meteorites, scihave been able to determine the relative abundance of the elements in our part of the galaxy. The most abundant elements are listed
in Table 6-4.
entists
TABLE
common
6-4
Solar abundances of the most
elements
Atomic
number
Relative
Symbol
Element
abundance
Hydrogen
10'-'
He
Helium
6x
Carbon
4 X 10"
10'"
Nitrogen
9x
10'
Oxygen
7x
10"
10
Ne
Neon
10"
12
Mg
Magnesium
4 X 10'
5 X 10'
14
Si
Silicon
16
Sulfur
2x
10'
26
Fe
Iron
3x
10'
Our
solar system
In addition to tiiese ten very common elements, there are five that
are moderately abundant: sodium, aluminum, argon, calcium, and nickel.
These elements have abundances in the range of 10^ to 10^ relative to
the standard trillion hydrogen atoms. All other elements are
For instance, for every trillion hydrogen atoms

are only six atoms of gold.
The planets were formed by

the accumulation of material
in the solar nebula during
the birth of the Sun
in the solar
much
rarer.
system there
Hydrogen and helium are the most common elements, but they are not
on the inner four terrestrial planets. Because of warmth from
plentiful
the Sun, temperatures on the four inner planets are comparatively high.
The higher the temperature of a gas, the greater the speed of its atoms.
The
light-weight atoms of hydrogen and helium move swiftly enough to

escape from the relatively weak gravity of these small planets.
A different situation prevails on the four Jovian planets. Far from the
Sim, temperatures are low and the atoms of hydrogen and helium move
slowly. The relatively strong gravity of the massive Jovian planets easily
prevents the lightest gases from escaping into space. In fact, the Jovian
planets are composed primarily of hydrogen and heliimi.
Temperature must
also
have been an important factor
in
determining
conditions inside the vast cloud of gas and dust, called the solar nebula,
out of which the solar system formed. It is thus extremely useful to categorize the abundant elements and their common compounds according
to their behavior at various temperatures.
First, hydrogen and helium are gaseous except at extremely low temperatures and extraordinarily high pressures. Second, rock-forming compounds of iron and silicon are solids except at temperatures exceeding
1000 K. Finally, there is an intermediate class of common chemicals such
as water, carbon dioxide, methane, and ammonia. At low temperatures
below 200 to 300 K), these substances solidify into solids called
At somewhat higher temperatures, they can exist as liquids or gases.
In Table 6-5, the common planet-forming substances are listed according
to these three broad classifications.
By observing the process of star formation elsewhere in our galaxy,
astronomers can deduce the conditions that must have led to the formation of our solar system. For example. Figure 6-14 shows a disk of material surrounding a star. Planets may still be forming in this disk of debris
left over from the birth of that star.
Just before the birth of the Sun, atoms in the scalar nebula were so
widely spaced that no substance could exist as a liquid. Matter in this
vast cloud existed either as a gas or as tinv grains of dust and ice. Astronomers find it useful to speak of condensation temperature to specify
whether a substance is a solid or a gas under these conditions of extremely low pressure. Above its condensation temperature, a substance is
(typically
ices.
TABLE
6-5
substances
Common
planet-forming
Gas
Ice
Rock
Hydrogen (H)
Water (H.,0)
Iron (Fe)
Helium (He)
Methane (CHj)
Iron sulfide (FeS)
Neon
Ammonia (NH3)
Olivine ((Mg,Fe)Si04)
Carbon dioxide (CO2)
Pyroxene (CaMgSisOg)
(Ne)
Uui
Figure 6-14
matter
Willie
circumstellar disk of
J his comliuler-rtilmnted
wilh a
CCI)
slinu's
ii
phiiliif!;iiij>h
oj miilirial
ili.sk
iirhiling the star nilleil ji I'lctiirus. This star
located behind a small circular
center of the picture

the star's tight,
overwhelmed
solar s\slem
tliat
mask
was needed
is
at the
to block
which would have otherwise
light
from
seen nearly edge-on,
is
the disk.
The dkk,
believed to be very
young, possihh no more than a few hundred

million years old. (Unn'er\ily of .{rizona
and J PL)
a gas; below
its
condensation teniperalLire, the substance
solidifies into
tiny specks of dust or snowflakes.
Rock-forming substances have very high condensation temperatincs,

range of 1300 to 1600 K. The ices have condensation
temperatures in the range of 100 to 300 K. The condensation temperatures of hydrogen and helium are so near absolute zero that thev always
existed as gases during the creation of the solar system.
Initially, the solar nebula was quite cold. Temperatures thioughout
the cloud were probably less than 50 K, which is below the condensation
temperatures of all common substances except hydrogen and helium.
Snowflakes and ice-coated dust giains must have been scattered abundantly across the solar nebula, which had a diameter of at least 100 AU
and a mass roughly two or ihiee times the mass of the Sini.
The gravitational pull of the particles on one another caused them to
begin a general drift toward the center of the solar nebula. The densitv
and pressure at the center ol the solar neliula l)egaii to increase, productypically in the
Our
suUir syilem
ing a concentration of matter called the protosun. Because of gravitational contraction, temperatures deep inside the solar nebula began to
The
have had an overall slight amount of

it is properly called. Otherwise,
everything would have fallen straight into the protosun, leaving nothing
behind to form the planets. As the solar nebula contracted toward its
gravitational center, it became transformed from a shapeless cloud into a
rotating flattened disk that was warm at the center and cold at its edges.
Astronomers feel confident in describing a flattened, disk-shaped solar
nebula because all the planets today have orbits that are very nearly in
climb.
solar nebula
rotation, or angular
must
also
momentum,
as
same plane.
Temperatures aroutid the newly created protosun soon climbed to
2000 K while temperatures in the outermost regions of the solar nebula
remained at less than 50 K. Figure 6-15 shows the probable temperature
distribution throughout the solar nebula at this preliminary stage in the
the
Temperature (K)
Figure 6-15
Temperature distribution in
the solar nebula This graph shows how ihe
lemperalure probubly varied across the solar
nebula as the planets were forming. For the

inner planets, temperatures ranged from
roughly
1200
at
Mercury
to
500
at the
of Mars. Beyond Jupiter, temperatures

were everywhere less than 200 K.
orbit
The terrestrial planets

formed out of rocky material,
whereas the Jovian planets
also incorporated vast
amounts
of ices
and gases
formation of the solar system. Obviously, all the common icy substances
in the inner regions of the solar nebula were vaporized by these high
temperatures. Only the rocky substances remained solid, which is why
the four inner planets are composed primarily of dense, rocky material.
In contrast, snowflakes and ice-coated dust grains were able to survive in
ttie cooler, outer portions of the solar nebula. That is why the Jovian
planets have low average densities.
Recent observations suggest that many young stars are sinrounded by
circumstellar disks of pre-planetary material. It seems reasonable to suppose that a similar temperature-driven segregation of rocky and icy substances occurs around such stars. Thus other planetary systems may also
exhibit a dichot(jmy between terrestrial-like inner planets and Jovian-like
outer planets.
The formation of the four inner planets was dominated by the fusing
together of solid, rocky particles. Initiallv, dust grains coalesced and accumulated into objects called planetesimals, with diameters of about
100 km. Planetesimal formation took a few million years, as neighboring
dust grains and pebbles in the solar nebula collided and were held to-
gether by electrostatic or gravitational forces.

During the next stage, gravitational attraction between the planetesimals caused them to collide and coalesce into still larger objects called
protoplanets. This accumulation of material through the action of gravitv is called accretion. In recent years, astronomers have used detailed
computer simulation to improve our understanding of accretion. The
computer is programed to simulate a large number of planetesimals circling a hypothetical newborn sun along orbits dictated by Newtonian
mechanics. Such studies show that accretion may continue for roughly
100 million years and will typically form about half a dozen planets.
A particularly successful computer simulation is summarized in Figure
6-16. The calculations begin with 100 planetesimals each having a mass
of 1.2 X 10"'' grams. This proceedure ensures that the total mass (1.2 x
I0~* grams) equals the mass of the four terrestrial planets (Mercury
through Mars) plus their satellites. The initial orbits of these planetesimals are inclined to each other by angles less than 5, to simulate a thin
layer of asteroidlike objects orbiting the protosun.
In this simulation, after an elapsed time of 30 million years the 100
original planetesimals have coalesced into 22 protoplanets. After 79 million years, II larger protoplanets remain. Nearly atiother 100 million
years elapse before the total number of growing protoplanets is reduced
Our
Figure 6-16
planets
oj
Accretion of the terrestrial
These three ilrawiugs shuie the
a computer simulation of the fnrmotion of
the inner planets, (a)

zoith
100
The simulation begins
planetesimals. (b) After
years, these planetesimals
22
results
30 milium
have coalesced
protoplanets. (c) Tins final view
elapsed time of
441
130
into
for an
million years, but the for-
mation of the inner planets

plete after only
is
is
essential^ com-
million \ears. (Adapted
from a computer simulation by George

W. Wetherill)
solar s\slem
shows four planets following nearly circular orbits

This exercise does not
exactly reproduce the inner solar system because the fointh planet from
the Sun ends up being the most massive. In fact. Earth is nine times
more massive than Mars. Nevertheless, agreement with most characteristo six. Figure 6- 16c
after a total elapsed time of 441 million years.
tics
of the inner solar system
The
is
very striking.
material from which the inner protoplanets accreted was rich in
mineral-forming elements having high condensation temperatures. Iron,

magnesium, and sulfur were particularly abundant, followed
closely by aluminum, calciuin, and nickel. Energy released bv the violent
impact of planetesimals with the growing protoplanets as well as the
decay of radioactive elements melted all this rockv material. The terrestrial planets thus began their existence as spheres of molten rock. During this time, denser iron-rich minerals sank to the center of the planets,
forcing the less-dense silicon-rich minerals to the surface. That is why
planets such as the Earth have dense iron cores surroimdcd bv lessdense rock.
The formation of the outer planets probably proceeded slightly differently from that of the inner planets. Recall that the original solar nebula
was a rotating disk of snowflakes and ice-coated dust particles. .Analyses
of the behavior of the outer regions of such a disk have shown that it
would probably break up into rings around the developing protosun.
separated by regions of space fairly free of matter. The material in each
ring would then coalesce to Ibrm a large protoplanet. This protoplanet
would be largely gaseous because the accumulation of the small particles
would release enough energy to vaporize the abundant ices. These massive protoplanets would then sweep up \ast amounts of gaseous hvdiogen and helium as the\ moved along their orbits. The final result is four
huge planets, each with an enormously thick atmosphere siurounding an
Earth-sized core of rocky material. Figure 6-17 summarizes this storv of
the formation of the outer .solar system.
Some of the details of this story about the formation of the planets
will probably be revised in coming decades as we gather more evidence
from spacecraft and orbiting observatories. However, most of the new
discoveries in the past few decades have confirmed the broad outlines of
this scenario, and astionomers are fairly (onfident that the solai system
did form in the fashion depicted here. Of course, this discussion presents conclusions drawn from many dif fereiU kinds of e\ idence, much of
which we will discuss in later chapters.
silicon,
0)
Figure 6-17
ôlar sysli'm
Formation of the solar system
This series of sketches shows major stages in

the birth of the solar system
spanning 100
lion \ears. Terrestrial planets accrete
rocky material in the

the solar nebula.
warm, central regions of
Meanwhile,
the huge, gase-
ous Jovian planets form in the cold outer

gio7is. (a)
stages, (b)
The
The
solar nebtila in
its
100
re-
initial
early solar system after
lion years, (c) Planetary
complete) after
mil-
from
50
mil-
formation (nearly
million years.
During the millions of years while the planets were forming, temperaand pressures at the center of the contracting protosun continued
tures
to climb. Finally,
temperatures
at the
center of the protosun reached 8
enough to ignite thermonuclear reactions, and the

Sun was born. As we shall see in Chapter 13. detailed calculations have
million degrees, hot
demonstrated that Sun-like stars take approximateh 100 million years to

form from the prestellar nebula. As a result, the Sun must have become
Our
solar i\stem
97
a full-fledged star at roughlv the
same lime the accretion of the inner
protoplanets was complete.
A newborn
star adjusts
clear reactions at
its
core,
somewhat violently to the onset of thermonuand often the star's tenuous outermost layers
are vigorously expelled into space. This brief burst of mass loss is observed in many young stars across the sky and is called a T Tauri wind
Taurus (the bull) where it was first identified. The T

Tauri wind that heralded the birth of the Sun swept the solar system
clean of excess gases, preventing further accretion. Many small rocks
were left behind to pelt the planets over the next half billion years. For
all practical purposes, however, the formation of the solar system was
complete by the time the T Tauri wind began to blow.
The Sun today continues to lose matter gradualK in a mild fashion.
This on-going, gentle mass loss, called the solar wind, consists of highspeed protons and electrons leaking awav from the Sun's outer lavers. In
the next two chapters, we shall see that each (jf the planets carves out its
own distinctive cavit\ in this solar wind.
after the star in
Summary
The
four inner planets of the solar system share many characteristics and
are distinctly different from the four giant outer planets.
The
four inner (terrestrial) planets are relatively small, have high average
and are composed primarily of rock.
densities,
The
giant outer (Jovian) planets have large diameters, low densities, and
are primarily composed of hydrogen and heliinn.
Pluto, the
outermost planet, resembles an
icy asteroid.
Spectroscopy, the study of spectra, provides information about the chemical composition of distant objects.
Spectral lines serve as distinctive "fingerprints" for elements
compounds in objects from which light comes.
and chemical
Each chemical element has a particular kind of atom, and atoms of various elements can combine to form molecules of chemical compounds.
Hydrogen and helium, the lightest elements, were formed shortly after
the creation of the universe: the heavier elements were produced much
later in the centers of stars and cast into space when the stars died.
B\ mass. 98 percent of the matter
in
the universe
Planet-forming substances can be classified as gas,

their condensation temperatme.
is
hydrogen and he-
ice,
or rock depending
on
The
solar system
formed from
hydrogen and heand dust particles.
a disk-shaped cloud of
lium, called the solar nebula, that also contained ice
The
inner planets formed through the accretion of dust particles into

planetesimals and then into larger protoplanets.
The
outer planets probably formed through the breakup of the outer

nebula into rings of gas and ice-coated dust that coalesced into huge
protoplanets.
The Sun formed
by accretion at the center of the neinil.i. .After about 100

million years, temperatures at the protosun's center were high enough to
ignite
thermonuclear reactions.
Our
solar system
When
the protosun became a star, excess gas was vigorously blown away
from the Sun, ending the process of planet formation.
Review questions
Give three differences between
On
ets?
terrestrial
and Jovian
what grounds can certain satellites also be

To which planets do these satellites belong?
Why
planets.
classified as terrestrial plan-
do you suppose there are no Jovian planets near the Sun?
hydrogen and helium account for 98 percent of the mass of all the
material in the universe, why aren't the Earth and Moon composed primarily
4
If
of these two gases?

*
Why
are water (HoO).
methane (CH4), and ammonia (NH
comparatively
abundant substances?
6
Why
is it
reasonable to suppose that the solar svsteni was formed dining a
relatively short interval?
7 Would you expect very old stars to possess planetary systems? If so, what
types of planets would they have? Explain.
8 Compare and contrast the optics of a prism spectrograph with those of a
grating spectrograph. What are the advantages and disadvantages of each?
Advanced questions
9 Suppose you are trying to determine the chemical composition of the atmosphere of a planet by observing its spectrum. Chemicals in the Earth's
atmosphere also produce spectral lines (often called telluric lines) in the spectra you observe. Can you think of a way of distinguishing between the telluric lines and the spectral lines of the distant planet?
^10 Suppose there were a planet having roughly the same mass as the Earth
but located at 50
from the Sun. What clo you think this planet would be
made of? On the basis of this speculation, assume a reasonable density for
this planet and calculate its diameter. How many times bigger (or smaller)
than the Earth is it?
AU
11 Propose an explanation for the fact that the Jovian planets are orbited by
terrestrial-like satellites.
12 Suppose a planetary system is now forming around some protostar

How might this system compare with our own solar system?
in the
sky.
For further reading
Beatty, J., et
al.,
eds. The Xeu< Solar System,
Cambridge University
Press, 1982.
2nd
ed. Sky Publishing
and
A fine,
up-to-date collection of
articles by expert planetologists on our modern view of the solar system.
C:hapman, C. Planets of Rock and
Ice.
Scribners, 1982.
^4 zvell-u'rilten
guide
lo the terrestrial planets.
Cohen, B. "Are
We
escope, Oct.
Kaufmann, W.
Beginning
to
Understand
Tauri Stars?"
.S'A'V
is!
Tel-
1981, p. 300.
Planets
and Moons. W. H. Freeman and Company. 1979. A
nontechnical introduction
to the solar system.
Our
solar syilem
Lewis,
j.
"The Chemistry of the Solar System."

1975.
Murray,
9^
.4
Scientific
American, Sept.
special issue devoted to the solar system.
The Planets. W. H. Freeman and Company. 1983. A colof articles about the solar .system.
Reeves. H. "The Origin of the Solar System." Mercuiy, Mar./Apr. 1977,
B., ed.
lection
p. 7.
The three
largest terrestrial planets
Tin-
Earth and Venus have nearly the same man.

size,
and
surface
us
surface gravity. However, Vetius's

perpetually shrouded by a dense
cloud cover containing poisonous, corrosive

gases.
Mars
is
only about half the size
only one-tenth the mass of Earth.
The
and has
thin,
dry Martian atmosphere does not protect the
The major
terrestrial planets
We begin our study of the solar system by examining our oivn planet and the
two terrestrial worlds that most closely resemble the Earth: Ve7ms and Mars.
We explore these worlds as alien visitors might, first descending into the planet's atmosphere, then probing the planet's interior. On Earth we find an atmosphere profoundly affected by the presence of life. Mars's atmosphere is very
sparse, but
on Venus we find oppressive,
planet from the Suti's ultraviolet radiation.
radioactive elements.
Both the Venusian and Martian environments
planet's interior to
are hostile to
(NASA)
life
fonns that thrive on Earth.
We
its
tures. Earth's crust is
hot,
poisonous gases.
We
also dis-
and
partly molten because of the decaying

see that the transfer of this heat outward from a
cover that the Earth's interior
is
hot
profound effect on the planet's surface feadivided into huge plates that jostle each other, produccrust has a
ing mountain ranges and earthquakes. On Mars, such plate movement ceased
long ago as this small planet rapidly cooled, but on Venus we find two continent-like features. Filially, ive see that molten iron in the Earth's interior produces a magnetic field that shields our planet from the solar wind, whereas
Venus and Mars have no such protection.
Of
all
Earth.
we are most familiar with the

we drink its water, and we breathe its
the planets that orbit the Sun,
We
walk on
its
surface,
we know more about the Earth than about any other

the universe. Sometimes, however, this very familiarity makes it
Naturally,
for us to view the Earth in
member
air.
object in
difficult
of the solar system.

Enlightening comparisons have come from explorations of our nearest
planetary neighbors, Venus and Mars. What we have learned about both
its
place as a
Thf major Ifrmlrial f>himis
on how various conditions affect the evoluVenus shows us what the Earth might be like,
if it were nearer the Sun than it is, and Mars shows us what Earth
might
be like, if it were less massive than it is.
planets gives us perspective
tion oi Earth-like worlds.
Earth's
atmosphere
is
Imagine an alien spacecraft approaching the inner solar system. First

would see Mars with its thin atmosphere and barren desert landscapes. Farther in thcv would sec Venus with its shroud of corrosive
clouds hiding a forbiddingly hot surface. Between these two relatively
unpromising planets, they would see Earth with its ever-changing ballet
ol delicate white cloud formations contrasting against the darker browns
and blues of the continents and oceans (see Figure 7-1). Perhaps it is
only our own bias that leads us to guess that the aliens would focus on
the Earth as the most interesting and inviting of the terrestrial planets.
At first, alien visitors might be puzzled to find that Earth's atmosphere is very different from that of its neighbors. The atmospheres of
both Venus and Mars consist almost entirely of carbon dioxide, whereas
only 0.03 percent of the Earth's atmosphere is carbon dioxide. Earth's
atmosphere is predominantly a 4-to-l mixture of nitrogen and oxygen,
but these two gases are found only in very small amounts on Venus and
Mars, as Table 7-1 shows. (The data for Venus and Mars in the table
come from measurements bv Soviet and American spacecraft that landed
on these planets in the 1970s.)
mixture of nitrogen and

oxygen, whereas the
the\'
atmospheres of Venus
and Mars are nearly pure
carbon dioxide
Figure 7-1
rtfoil
Ih/il
ful object
The Earth
l-Milh
visible
IS
Aslrotianh
oflcii
Ihr most invilmgly heauli
from
blue-and-while world
Iheir spacecraft.
is
Our
the larjrest of the
tei
restrial planets. This photoipiiph teas taken in
1969
by Apollo II astronauts on the
of Africa
seen.
wa\
to
manned landinii on the Moon. .\/m/

and pmtiinn of Europe ion he
(NASA)
the first
The major loirshial planets
TABLE
7-1
The chemical composition of
Venus
Earth
Mars
Nitrogen
3%
77%
3%
Oxygen
Almost zero
21%
Almost zero
Carbon dioxide
97%
Almost zero
95%
Other gases
Almost zero
2%
2%
three planetary atmospheres
Why do Earth's planetary neighbors possess atmospheres of nearly

pure carbon dioxide when Earth's own atmosphere is ahiiost entirely nitrogen and oxygen? From study of the Earth's geological records, we
know that living organisms have been active here for at least the past 3
billion years
and
that such biological processes as photosynthesis are
largely responsible for the present chemical composition of Earth's at-
mosphere today. As far as we know, large amounts of oxygen in a planetary atmosphere can occur only as a direct result of biological activity. So
aliens might single out Earth for investigation because its oxygen indicates the possibility of
life.
Nearing Earth's surface, aliens would find that the temperature in our
atmosphere depends on altitude in a complicated fashion, as shown in
Figure 7-2. A graph like this can be used to deduce fundamental information about a planet's atmosphere because temperature variations result
from how sunlight is absorbed differently at various altitudes.
Seventy-five percent of the mass of Earth's atmosphere lies below an
altitude of 11 km (roughly 7 miles, or 36,000 ft) in a layer called the troposphere. All Earth's weather clouds, rain, sleet, and snow occur in
this lowest laver. Commercial jets generally fly at the top of the troposphere to minimize buffeting and jostling.
Atmospheric temperature decreases upward to 55C (= -67F =
218 K) at the top of the troposphere. Above this level, the region called
the stratosphere extends from 11 to 50 km (7 to 30 miles) above the
surface. Ozone (O3) molecules in the stratosphere efficiently absorb solar
ultraviolet rays, thus heating the air in this layer. The temperature increases upward through the stratosphere to about 0C (= 32F = 273 K)
at its top. Some scientists are concerned that chemicals we have made
and released into the air may destroy much of this ozone layer. If the
ultraviolet radiation from the Sun was not absorbed by the ozone in the
stratosphere, it would beat down on the Earth's surface. Because ultravi-
50-
Temperature profile of Earth's

atmosphere The variation of atmospheric
temperature with altitude shows maxima and
minima became of interactions between sun-
Figure 7-2
light
and various atoms and
different elevations.
-1000
ions al
Temperature (C)-
<
The major
lerreslnut
olet radiation
103
lilaiiets
breaks apart most of the delicate molecules that form
ing tissue, loss of the ozone layer could thus lead to

the planet.
liv-
literal sterilization
of
Above the stratosphere, atmospheric temperature declines with increasing altitude in the mesosphere, reaching a minimum of about
-75C (= 200 K =
This
-10()K) at an altitude of about 80 km (50 miles).

the bottom of the thermosphere, above which the
minimum marks
Sun's ultra\iolet light strips atoms of one or more electrons. Ionized

atoms and molecules of oxygen and nitrogen are the most prevalent elements. The stripped electrons easily reflect radio waves in a wide range
is why you can tune in a distant radio station: its
transmissions bounce off this radioreflective layer.
At sea level on Earth, the air weights down with a pressure of 14.7
pounds per square inch. By definition, this pressure is called one atmosphere (1 atm). Because we are familiar with this pressure, atmospheric
pressure on other planets is usefully expressed in atmospheres also. For
example, the atmospheric pressure on the sur face of Venus is 90 atm
of frequencies, which
whereas the sparse Martian atmosphere exerts a pressure of only 0.01

atm.
There are other common ways of expressing atmospheric pressure. In

is measured in dynes per square centimeter
rather than pounds per square inch. One atm happens to be equal to
the metric system, pressure
almost exactly 1 million dynes per square centimeter. This correlation

has prompted definition of another unit of pressure called a bar, where
1 bar = lO'' dyne/cm" = 0.987 atm. In other words, the atmospheric
pressure at sea level on Earth is 1.013 bar.
In expressing low atmospheric pressures, it is often convenient to use
the millibar, which is simply 0.001 bar. For example, the atmospheric
pressure on Mars is about 10 millibars, approximately the same as the
atmospheric pressure at an altitude of 30 km above the Earth's surface.
The atmospheric pressure above any planet decreases with increasing altitude because, quite simply, the higher you go, the less atmosphere
there is to weigh down upon you.
The surface
of
Venus
is
hidden beneath a very thick,

highly reflective cloud cover
At first glance, Venus looks like Earth's twin. The two planets have almost the same mass, diameter, a\'erage density, and surface gravity.
However, Venus is closer to the Sun than the Earth is and is thus exposed to a greater intensity of sunlight, transforming its potentially
Earth-like environment into an extremely hostile world.
Since Venus is closer to the Sun than is the Earth, we always find
Venus near the Sun in the sky. Venus is most easily viewed above the
western horizon after sunset (when it is called an evening star), or before
sunrise above the eastern horizon (when it is called a morning star).
Venus is easy to identify because it is often one of the brightest objects in the night sky. Venus's cloud cover reflects 76 percent of the sunlight that falls on it. From the Earth, Venus can be 16 times brighter
than the brightest
star.
Indeed, only the Sun and
Moon
oulshine Venus
at its greatest brilliancy.
Earth-based telescopic \iews of Venus reveal a thick, nearh featureunbroken layer of clouds (see Figine 7-3). Notice that Venus exhibits phases because of its nearness to the Sun (recall Figiue 3-9). Views
from spacecraft (see Figure 7-4) show details in the clouds, but never a
view through the clouds to the ground. This high reflective cloud cover
less,
makes Venus dazzlingly bright is also responsible for om- long-standing ignorance of the planet's surface.
that
T/w major
Figure 7-3
Earth-based views of Venus
This series of phulugtaphi shows
how
pearance of
moves along
its
orbit.
VeniL',
changes as
it
The number below each view
the ap-
is
the
angular diameter of the planet in seconds of

arc. Vemis reaches greatest brilliancy when its
angular diameter
about
is
40
arc
sec.
(New
Mexico Stale University Observatory)
Fûre
7-4
Venus
Venus's thick cloud cover
efficiently traps heat
from
the
a iuiface temperature (480''C

holler than
Sun
resulting
= 900" F)
even
on Mercury. Unlike Earth's clouds,
which are made of water droplets, Venus's
and contain droplets of

concentrated sulfuric acid. This photograph
clouds are very dry
was taken
in
1979
by
in orbit about Venus.
an American
(NASA)
spacecraft
24'
lerrcslnal planets
Thf major
lenestrial planets
105
Spacecraft have descended into Venus's clouds and provided detailed

information about the Venusian atmosphere. During the 1960s, while the
Americans were busy landing astronauts on the Moon. Soviet scientists
concentrated on building spacecraft that could survive a descent into the
Venusian cloud cover. The task proved to be more frustrating than anyone had expected because the spacecraft would stop transmitting data
before reaching the ground. Finally, in 1970, a Soviet spacecraft managed to transmit data for a few seconds directlv from the Venusian surface. Soviet missions during the early 1970s measured a surface temperature of 750 K (= 480C = 900F) and a pressure of 90 atmospheres
which is equal to a crushing pressure of f ton per square inch.
After some initial puzzlement, astronomers quickly realized that
Venus's high surface temperature has a straightforward explanation. Perhaps you have had the experience of parking your car in the sunshine
on a warm summer day. You roll up the windows, lock the car, and go
on an errand for a few hours. When vou return, you are annoyed to
discover that the interior of vour automobile has become stiflingly hot,
typicallv at least 20C warmer than the outside air temperature.
What happened to make your car so warm? First, sunlight entered the
car through the windows. This radiation was absorbed by the dashboard,
the steering wheel, and the upholstery causing their temperatures to rise
to
about 330
K (= 57C =
135F). At this temperature, the seats in
car reradiate the energy primarily at infrared wavelengths.
your
Your car win-
dows are opaque to these wavelengths. This energy therefore is trapped

inside your car and absorbed by the air and interior surfaces. As more
sunlight comes through the windows and is trapped, the temperature
continues to
Infrared radiation
cannot get back out
rise.
phenomenon
occurs in Venus's atmosphere, and to a lesser

extent in Earth's atmosphere. It is commonly called the greenhouse effect (see Figure 7-5). Sunlight enters the thick Venusian clouds, where it
is absorbed and reradiated at infrared wavelengths that cannot get back
similar
out. This trapped radiation
produces a high surface temperature.

While descending through the Venusian clouds, Soviet spacecraft
measured atmospheric pressure and temperature. The results are shown
in Figures 7-6 and 7-7. The pressure and temperature profiles of the
Venusian atmosphere are simple: both decrease smoothlv with increasing
altitude. As we saw in F'igure 7-2. Earth's atmosphere has a much more
complicated relationship between temperature and altitude. In the chapters on Jupiter and Saturn, we shall see how pressure and temperature
\ariati()n with altitude can have a profound effect on a planet's appear.uuc.
Figure 7-5
ing sunlight
The greenhouse
piniiiiitrs the
ra.\;/\
greenhouse and
ahwthfd
is
These objects reradiate

wavelengths
to
effect
this
hn lim-
windows of a
by objects inside.
energy at infrared
which the glass windows are
opaque. The trapped infrared radiation

sorbed
Iry
perature
the air
and
in.\ide the
objects,
is
ab-
causing the tem-
greenhouse
to rise.
(Recent
studies indicate that the physical barrier of
glass
windows plats a major
perature
rise inside
role in the tem-
a real greenhouse
or car
because the closed windows pre\>ent air from

mixing. Nex'ertheless. the greenhouse effect
does occur
and
is
ature on Venits.)
the cattse of the high temper-
Both Soviet and American spacecraft found the top of the V'enusian
at an altitude of about 68 km (42 miles = 220.000 ft). However,
the Russians also discovered the bottom of the clouds, at an elevation of
about 31 km (19 miles = 100.000 ft) above the ground. Below this altitude, the Venusian atmosphere is reinarkably clear.
In 1979, an American spacecraft plunged into the \'enusian atmosphere and discovered four distinct cloud layers. The upper cloud layer
extends from altitudes of 68 km down to 58 km. A denser and more
opaque middle cloud layer extends from there down to 52 km. The
lower cloud layer (from 52 to 48 km) contains the densest and most
opatjuc ()( the Venusian clouds, even though it is only 4 km thick. Under
tlie cloud layers is a haze layer between 48 and about 31 km. Below this
altitude of 100,000 feet, the atmosphere is clear all the way to the
ground.
clouds
The major
400
Temperature
[left]
in the
Venusian atmosphere
The temperatute
Venui's atmosphere
smoothly from a mini-
rises
in
mum of about 170 K (about -100C, or

150F) at an altitude of 100 km to a maximum of nearly 750 K (about 480''C, or
900F) on
the
ground.
Figure 7-7 [right] Pressure in the

Venusian atmosphere The pressure at
Venusian surface
is
pounds per square
a crushing
inch).
Above
the
90 atm (1300
the surface,
atmospheric pressure decreases smoothly with

increasing altitude.
planeti
600
Pressure (atm
Temperature (K)
Figure 7-6
lerrt'slrial
Because Venus and Earth are so similar in size and mass, astronomers
have assumed that Venus's clouds were made of water vapor, like
Earth's. However, this assumption was proved wrong in the 1960s as scientists had great difficulty detecting any water at all in the Venusian
atmosphere. The latest measurements demonstrate that Venus is extremely dry with water making up far less than 1 percent of the clouds.
If the clouds are dry, what are thev made oP
Clues to the complex chemistry of the Venusian clouds came in bits
and pieces in the 1960s when Earth-based observations and measurements from spacecraft demonstrated that the clouds efficiendy absorb
radiation at specific infrared and microwave wavelengths. Scientists soon
proposed a surprising explanation for this absorption; droplets of sulfuric acid.
When
sulfuric acid
HSO^" and HaO"^

wave radiation
The
(H2SO4)
is
are produced.
at precisely
mixed with water (H2O), the ions

These ions absorb infrared and micro-
the wavelengths observed in Venus's clouds.
would be at least partly responsible for depleting the

Venusian atmosphere of water by con\erting it into HÔ^. The latest
Soviet and American probes leave no doubt that the Venusian clouds are
sulfuric acid
in fact
made of
droplets of concentrated sulfuric acid.
Venus appears yellowish or yellow-orange to the human eye, and data

from spacecraft have indicated why; the upper clouds contain substantial
amounts of sulfur dust. Over the temperature range of these upper
clouds, sulfur
is
distinctly yellow or yellowish-orange.
large concentrations of sulfur
compounds
At lower elevations,
(especially SOo,
OCS, and
H2S) were found along with droplets of sulfuric acid. Because of the tremendous atmospheric pressure, the droplets do not fall as a rain; they
are more-or-less permanently suspended in the clouds like an aerosol.
The sulfuric acid in Venus's atmosphere causes a number of chemical
and chlorides in surface rocks give

hydrofluoric acid (HE) and hydrochloric acid (HCl). Eurther reactions produce fluorosulfuric acid (HSO^F), one of the most corrosive
substances known to chemists, which can dissolve lead, tin, and most
reactions. Reactions with fluorides
rise to
rocks. Indeed, the

hostile to metals
Venusian clouds are a cauldron of chemical reactions
and other
solid materials.
It is
clear
why
the Russians
had so much trouble getting a fimctioning spacecraft down through the

clouds to the surface.
The major
The Martian atmosphere

and dry
is
thin
lerrestriul planets
107
Mars is the only planet whose surface features can be seen through
Earth-based telescopes (see Figure 7-8). Even with the modest telescopes
of the seventeenth century, astronomers made discoveries which hinted
that Mars might be very Earth-like. For example, in 1666 the Italian astronomer Gio\anni Domenico Cassini spent many nights carefully observing the motions of surface features on Mars and discovered that the
planet's rotation period is 24 hours, 37 minutes. Thus a day on Mars is
only slightly longer than a day on Earth. A century later, the Germanborn English astronomer William Herschcl determined the inclination of
Mars's axis of rotation. Just as Earth's equatorial plane is tilted 232 from
the planet of its orbit, Mars's equator makes an angle of nearly 24 with
its orbit. As a result. Mars experiences seasons much as Earth does.
Cassini also discovered that Mars has polar caps, which are large during the Martian winter but shrink with the coming of summer. Early telescopic observers also reported seasonal color variations that seemed to
indicate vegetation on the Martian surface, and some reported geometric
patterns that might represent a planetwide system of artificial canals. Scientists speculated about the possibility of life on Mars, and science-fiction
writers wove popular stories about invasions of Earth bv hostile Martians.
Ironically, the first actual invasion came in 1976 when automated
spacecraft from the Earth landed on the Martian surface. These probes
sent back pictures and data indicating that Mars is a barren, desolate
world. The possibility of some form of Martian life has not been completely ruled out, but it now seems quite likely that Mars is as sterile and
lifeless as
the
Moon.
The two American spacecraft that journeyed to Mars in 1976 Viking

and Viking 2 made numerous measurements of conditions in the
Martian atmosphere during their descents. Figure 7-10 shows the recorded atmospheric temperature plotted against altitude above the Martian surface. Also shown, for comparison, is the atmospheric temperature
above the Earth's surface. Earth exhibits a maximum temperature at an
altitude of 50 km because of the absorption of ultraviolet radiation by
our ozone layer. The temperature profile of the Martian atmosphere
exhibits relatively little variation with altitude, and so we conclude that
Mars has no ozone layer. The absence of an ozone layer means that the
Sun's ultraviolet radiation strikes the Martian surface directly. The sterilizing effect of ultraviolet light may be an important factor inhibiting the
appearance of life on Mars.
The meteorological instruments on both Viking landers promptly confirmed that the surface atmospheric pressure is in the range of 6 to 8
Mars viewed from
Figure 7-8
the Earth
This high-quality. Earlh-basril phntngraph oj
mars was taken
Mars
distance
(35 million
in 1971 when the Earthwas only 56i million kilometers
miles).
At that time. Mars pre-
sented a disk nearly
25
arc sec in angular
diameter. Circumstances as favorable as these

will not be repeated
for the
rest
of the century.
Thii photograph accurately portrays u'hat you
might typically see through a modernte-sited

telescope
under
excellent observing conditions.
Notice the prominent polar cap. (Courtesy of

.
Stephen At. Larson)
Thr major
Figure 7-9
tures of the
Mars Many
lerri-slriiil j)h>iels
of the major fea~
Martian surface are seen
in this
photograph from Viking 2 as the spacecraft
approached the planet in 1979. Clouds flank

the western slopes of the
pus
Mons
the middle
huge volcano Olym-
near the top of the photograph. In

is the vast rift canyon called Valks
Marineris. This canyon stretches nearly
km along
the planet's equator.
the photograph, carbon dioxide
floor of the Argyre Basin

craters.
400U
At the bottom
snow
of
lines the
and snnvundmg
(NASA)
had been expected from previous flyby missions. The atmospheric pressure at sea level on Earth is about 1000 millibars, so the density of the Martian atmosphere is less than too that of the Earth's atmosphere.
Direct chemical analysis of the Martian atmosphere by the Vikings'
instruments reported a carbon dioxide abundance of 95 percent, with
millibars, as
nitrogen at 2.7 percent and argon 1.6 percent. The remaining fraction
of a percent is mostly oxygen and carbon monoxide, with only a small
amount of water vapor. If all the water vapor could somehow be
squeezed out of the Martian atmosphere, it would not fill one of the five
Great Lakes in North America.
In most of the pictures sent back from Mars, the sks has a distinctly
pinkish-orange tint (see Figure 7-11). This coloration is thought to be
caused by extremely fine-grained dust suspended in the Martian atmosphere. Indeed, Earth-based observers have often reported seeing planetwide dust storms that occasionally obscure the Martian surface features
for several weeks at a time. Although the Martian atmosphere is very
thin, its winds are sometimes strong enough to raise large amounts of
fine dust particles high into the atmosphere.
After the Viking landers had been on the Martian surface for a few
weeks, their data showed clearly that the atmospheric pressure at both
landing sites was dropping steadily. Mars seemed to be rapidly losing its
The major
Figure 7-10
109
Temperature of the Martian

Earth. Mars Uuks an
atmosphere
L'litih-
ozone layer in
its
atmosphere. Comequenth.
the temperature profile of the
phere does not show a

tiluite
terrestrial planets
Martian atmos-
maximum
at
50 km
al-
as does the Earth's temperature profile.
Tempcralure (C)
atmosphere, and some scientists joked that all the air would be gone in a
few months. A straightforward explanation was, however, readily available: winter was coming to the southern hemisphere. At the Martian
south pole, it was so cold that large amounts of carbon dioxide were sohdifying out of the atmosphere, covering the ground with dry-ice snow.
In early 1977, when spring came to the southern hemisphere, the
dry-ice snow rapidlv evaporated, and the atmospheric pressure returned
to prcwinter levels. With the arrival of winter in the northern hemisphere, another decrease in atmospheric pressure was observed as drvice snow blanketed the northern latitudes (see Figure 7-12). The Nlartian
surface clearly experiences extreme seasonal variations in temperature
and atmospheric pressure.
Aside from an occasional dust storm, the weather on Mars seems boring to someone accustomed to the temperate zones of the Earth. Atmospheric pressure varies with the sea.sons in a regular, predictable fashion.
varies with the time of dav, also in a monotonously repetitive way. Actually, the dailv temperature variation on
The atmospheric temperature

Figure 7-1 1
A panorama
landing
This mosaic of three views
site
of the Viking 2
shows about 180 of the Viking 2 landing

site. Xorlhwest is at the left, southeast
on the
right.
The
spacecraft.
flat, featureless
km
(NASA)
approximately 3
(2 miles)
horizon
away from
is
the
Mars is quite similar

on Earth (see Figure
temperature changes observable in a desert

The warmest time of day occurs about two
hours after local noon, and the coldest temperatures are recorded just
before sunrise. The thin, dry Martian atuKJsphere is not capable oi retaining much of the da\'s heat. howc\er, so tlie range of the dailv temperature swings on Mars is larger than that on Earth.
to the
7-13).
The major
Winter on Mars
Figure 7-12
luie. taken in
May
site.
This pic-
979. shows a thin frost
layer that lasted for about a

the Viking
lerrtslrml planets
hundred days
at
Freezing carbon dioxide ad-
heres to water-ice crystals
and
dust grains
the atmosphere, causing them to fall to the
ground. The sky

it
was
7-11.
in the
is
therefore not as pink as
summertime view of Figure
(NASA)
Figure 7-13
^^ ---
Daily temperature variation
The lower curve shows

variation at the Viking
the daily temperature

I
site.
The upper
20
20
40
^^^^""^
Earth
curve shows the daily temperature variation at

a desert
site in
ture range on
California.
Mars
is
The
Martian air does not
daily tempera-
about three limes
greater than that on Earth.
The
thin, diy
retain heat as well as
the Earth's atmosphere does.
"\
/^Mars
^^
60 -
80
1
Time
Earth's crust consists of

large plates
affect the
whose motions
appearance
of our planet
Were
10
12
14
16
of day (hours)
on Earth, they would find it radically different from

neighbors in yet another way: the Earth is very wet. Nearly
7 1 percent of the Earth's surface is covered with water. An alien space
probe to Earth might send back thousands of photographs such as Figure 7-14 representing a random close-up view of Earth's surface. In contrast, Venus and Mars are extremely arid. No place on Earth is as dry as
the surfaces of Venus and Mars. By Venusian or Martian standards, our
Sahara desert is a veritable swamp.
One of the most important geological discoveries of the twentieth century is the realization that our planet has an active, constantly changing
crust. Geologists have learned that the Earth's crust is divided into huge
plates that constantly jostle each other, producing earthquakes, volcanoes, and oceanic trenches.
The idea of huge, moving plates might occur to anyone who carefully
examines a map of the Earth. You can see that South America would fit
aliens to land
either of
its
The major
Figure 7-14
typical close-up view of
Earth's surface
Mure than twu-thmh
Earth's surface
covered with water. In con-
trast, there is
lerreMrtal planets
is
oj
no liquid water on Venus
or Alars.
snugly against Africa, were it not for the Atlantic Ocean. Indeed, the fit
between land masses on either side of the Atlantic Ocean is remarkable
(see Figure 7-15). This observation inspired people such as .Alfred Wegener to propose the idea of "continental drift," suggesting that the continents on either side of the .Atlantic Ocean have simply drifted apart.
Wegener got
the idea of drifting continents around 1910 while looking

globe of the Earth. .After much research, Wegener published the
theory that there was originally a single, gigantic supercontinent he
at a
called Pangaea.
which began to break up and drift apart soine 200 milPangaea first split into two smaller supercontinents called
Laurasia and Gondwanaland. Gondwanaland later split into .Africa and
South America, while Laurasia divided to become North .America and
lion years ago.
Eurasia.
.Most geologists initially greeted
Wegener's ideas with ridicule and

was generallv accepted that the continents do "float"
on denser material beneath them, few geologists could accept the idea
that entire continents could move around the Earth at speeds that must
be as great as se\eral centimeters per vear.
Then in the mid-1950s, scientists began discovering long mountain
ranges on the ocean floors, such as the Mid-Atlantic Ridge (see Figure
7-16), which stretches all the wav from Iceland to Antarctica. During the
196()s, careful examination of the ocean floor revealed that molten rock
from the Earth's interior is oozing upward along the Mid-.Atlantic Ridge,
which must therefore be a long chain of underwater volcanoes. This
upwelling of new material causes sea-floor spreading. The eastern floor
of the .Atlantic Ocean is moving eastward, and the western floor is moving westward. This movement is pushing South .America and Africa
apart at a speed of roughly 3 cm per vear.
Sea-floor spreading provided the mechanism that had been missing
from the theories of continental drift. With some embarrassment, geologists began to review the old theories around 1964 and found a great
deal of new evidence to support them. This time, however, the emphasis
was upon he motion of large plates of the crust (not simplv the continents), and the modern theorv of crustal motion came to be known as
plate tectonics. Cleologists i()da\ realize that the boundaries of Earth's
crusial plates can be identified b\ seismic activitv occurring where these
plates are either colliding or separating. The plate boundaries stand out
clearly when the locations of earthquakes are plotted on a map (see Figure 7-17).
scorn. .Although
Figure 7-15
Comparing
Africa, Europe, Creenlund.
South America
the continents
and \orth and
fit lof;ether as
once joined. The
fit
is
though they were
especially convincing if
the edges of the continental shrives (rather
than today's shorelines) are used. (Adapted
from P.
A/. Hurley)
it
The major tf}n\hiat
Figure 7-16
This
The Mid-Atlantic Ridge
artist's remiilioii ihinvs the
floor of the
North Atlantic Ocean. The xmusxial mountain

range in the middle of the ocean floor, called
the Mid-Atlantic Ridge, is caused by lava
seeping up from the Earth's interior along a
rift
that extends
from Iceland to Antarctica.

and Bruce Heezen
(Courtesy of Marie Tharp
1977)
Fûre 7-17
aries
The major plates
The bound-
of major plates are the scenes of violent
seismic
and
geologic activity.
Most earthquakes
occur where plates separate or
collide.
Plate
boundaries therefore are easdy identified by

plotting the locations of earthquakes (shown
here with red dots) on a map.
jAant'ts
The major tmeilniil planeh
Rift
Ocean
The mechanism of plate
Figure 7-18
tectonics
Convection cut-rents
the soft
upper laser of the Earth's interior (called the

asthenosphere) are responsible for pushing
around
rigid, low-density plates
on the cntsi
(called the litliosphere). A'ni' cntst
oceanic
rifts,
is
formed
where lava oozes upward between
separating plates. Mountain ranges
and deep
oceanic trenches are formed where plates collide.
floor
Trench
/
Mountain range
Geologists believe that convective currents beneath the Earth's crust

are responsible for the movement of the plates. Heat is transported from
the Earth's interior toward its surface by convection, a process in which
warm material rises while cool material sinks. Rock beneath the crust is
not liquid, but it is hot enough to permit an oozing, plastic flow
region called the asthenosphere. As sketched in Figure

(magma) seeps upward along oceanic rifts where
plates are separating. Cool material sinks back down along subduction
zones where plates are colliding. Above the plastic asthenosphere is a
low-densitv, rigid laver called the lithosphere, divided into plates that
simply ride on the convection currents of the asthenosphere.
The regions where plates meet are the sites of some of the most im-
throughout a
soft
7-18, hot material
pressive geological activity on our planet (see Figure 7-19). Great mountain ranges, siich as those along the western coasts of North and South
America, are thrust up by ongoing collisions with the plates of the ocean
floor (see also Figure 7-20).
Figure 7-19
The
The separation of two plates
plates thai carry F.gypI
and Saudi
.Arabia
are moving apart, leaving the trench that conthe Red Sea. In this vieii< taken by a.UronauU ill 1966. .Saudi Arabia is on the left
and Egypt (with the \'ile River) on the
laim
right.
IXAS.'M
Subduction zones, where old crust
is
pushed

The major
Figure 7-20
The
terrestrial planets
of two plates
and China are colliding. The Himalaya Mountains have been
thrust upward as a result of this collision. In
collision
The
plates that carry Itidia
this
photograph taken by astronauts in 1968.
India
Ml.
IS
on the
Ex'erest
is
left.
Tibet on the right,
and
one of the snow-covered peaks
near the center. (NASA)
down into the mantle, are typically the locations of deep oceanic
trenches such as those off the coasts of Japan and Chile.
back
Mars has huge volcanoes

and canyons while Venus
has two continents, but
neither planet
shows
significant tectonic activity
The
convection process associated with sea-floor spreading is the primary

which heat is transported outward from the Earth's interior to
the crust. Here again our planet differs from its neighbors. \'enus and
Mars apparently transport heat outward to their surfaces predominantly
by a mechanism called hot-spot volcanism. Hot areas deep inside Venus
and Mars squirt molten la\a up through the crust. In the absence of any
tectonic activity to move the crust around, millions of years of eruptions
eventually build enormous volcanoes.
Hot-spot volcanism also occurs on Earth. The Hawaiian Islands, which
are in the middle of the Pacific plate, are a fine example. There is a hot
spot in the Earth's interior beneath Hawaii that continuously pumps lava
up through the crust. However, the Pacific plate is moving northwest at
the rate of several centimeters per vear. New volcanoes are created over
the hot spot as older volcanoes move away from the magma source, become extinct, and eventually erode and disappear beneath the ocean. In
fact, the Hawaiian Islands are the most recent additions to a long chain
of extinct volcanoes that stretches all the way back to Japan.
Martian volcanoes were first photographed by a spacecraft that was
placed into orbit about Mars in 1971. The largest Martian volcano,
way
in
Olympus Mons,
rises
24
km
(15 miles) above the surrounding plains
Mount Everest (see Figure 7-21). The highvolcano on Earth is Mauna Loa in the Hawaiian Islands, whose summit is only 8 km above the ocean fioor. The huge size of Olympus Mons
strongly suggests a lack of plate tectonics on Mars. A hot spot in Mars's
mantle kept pumping lava upward through the same vent for millions of
years, producing one giant volcano rather than a chain of smaller volcanearly three times as high as
est
noes.
The major
Olympus Mons
Figure 7-21
This photo-
graph looks straight ilown on the largest
cano on Mars. Olympus

tail as
Mount
Everest,
Mons
and
its
600 km (370
(NASA)
measures nearly
diameter.
terrestrial planets
is
2}
vol-
times as
cliff-nnged base
miles)
The base of Olympus Mons is ringed with clilfs. has a diameter of

nearly 600 km, and covers an area as big as the state of Missouri. At the
volcano's summit are several overlapping volcanic craters, forming a caldera large enough to contain the state of Rhode Island (see Figure 7-22).
Olvmpus Mons is one of several verv large volcanoes clustered together on a huge bulge called the Tharsis Rise, which covers an area
2500 km in diameter centered just north of the Martian equator.
Ground levels over this vast dome-shaped region are tvpically 5 to 6 km
higher than the average ground level for the rest of the planet. Another
major, though less dramatic, grouping of volcanoes is located nearK on
Figure 7-22
view
oj the
The Olympus caldera
summit
oj
Ohmpus Mons
This
is
ba.\ed
on a mosaic of six pictures taken by one of the

Viking spacecraft. The caldera consists of
overlapping, collapsed volcanic craters
measures roughly 70
itself is
km
across.
and
The volcano
wreathed in midmoniing clouds that
(H2O) brought upslope
formed from
ice
air currents.
The cloudtops are about S km
below the volcano's peak. (XAS.\I
by cool
*:2r^
The major
Figure 7-23
400 km,
state
is
segment of Valles
2000 km, shows an
-which
is
nearly the
area
same
300
by
size as the
of Pennsylvania. Earth's Grand Canyon
only as big as one of the tributmy canyons
seen in this photograph.
Figure 7-24
(NASA)
Eastern Valles Marineris
^t^^
This mosaic was constructed from fijteen photograplis taken by
one
oj the
and shows an area 1800

is
j)lfnu't\
This phologrcipli. taken [win an
Marineris
altitude of
Inrctriiil
toward the upper
left.
by
The
'
-^^ ''
Viking spacecraft
2000
km. North
eastern
end
of
Valles Marineris merges with chaotic, cratered
terrain toward the upper right.
(NASA)
.-
the opposite side of the planet.
None
of these volcanoes
seem
to
be ac-
tive today.
For reasons we do not yet understand, most of the volcanoes on Mars

are in the northern hemisphere, whereas craters are mostly found in the
southern hemisphere. Between the two hemispheres is a vast canyon
running roughlv parallel to the Martian equator
(see Figures 7-23 and 7-24).
Valles Marineris stretches 4000 km, beginning with heavily fractured
called Valles Marineris
and ending with ancient cratered terrain in the east.

canvon were located on Earth, it could stretch all the way from
New York to Los Angeles. Many geologists suspect that Valles Marineris
is a fracture in the Martian crust caused by internal stresses, not unlike
the rift valleys on Earth that result from plate tectonics. One such valley
is the Red Sea (recall Figure 7-19), which is 3000 km long.
terrain in the west
If this
The major
Figure 7-25
artist's
tar,
The continent of Ishtar
lerrestnat planets
117
This
impression shows the continent oj Ish-
which consists o] a high plateau ringed
The highest peak on the

Maxwell Montes, appears toward the
side of the drawing. The vertical reltej
with mountains.
planet.
right
slightly
exaggerated here.
is
(NASA)
^^
f^>
jrf(#lN^f'
Current theories suggest that plate tectonics did operate on the young
Mars but that this process long ago ceased there. A small planet loses its
internal heat much more rapidly than a large one. Mars's diameter is
only half that of the Earth, and its mass is one-tenth of the Earth's. Plate
tectonics halted early in Mars's history as its rapidly cooling lithosphere
became thick and sluggish.

Although Venus is perpetually shrouded
in clouds, scientists have

been able to use radar to determine surface features on the planet. In
1978, an American spacecraft placed in orbit about Venus mapped 93
percent of the planet's surface using a radar altimeter that bounced microwaves off the ground directly below. By measuring the time delay of
the radar echo, scientists determined the heights and depths of hills and
valleys.
Venus is remarkably Oat. About 60 percent of the planet's surface is

covered with gently rolling hills varying by less than 1 km from the planet's average radius of 6051.4 km. (This value for the planet's radius is
comiTionly used as a reference level, as we use average sea level here on
Earth.)
There are two large "continents" rising well above the generally level
surface of Venus. In the northern hemisphere, there is Ishtar Terra,
named after the Babylonian goddess of love. Ishtar is approximately the
same
size as Australia and consists of a high plateau ringed by high

mountains. The highest mountain is Maxwell Montes with a summit
km above the reference level. For comparison. Mount Everest on Flarth
rises 9 km above sea level. An artist's sketch of Ishtar is shown in Figure
1
7-25.
The second major Venusian

after the
I
odite
is
"continent," Aphrodite Terra (named

Greek equivalent of Venus), lies just south of the equator. Aphslightly bigger than Ishtar and has an area about one-half that
of Africa.
The spacecraft orbiting Venus also discovered some very large volcanoes southwest of Ishtar. These volcanoes, Rhea Mons and Theia Mons,
rise to altitudes of 6 km, with geiiiU sloping sides that extend o\er an
Tlie
Figure 7-26
The sufaces of Earth, Venus,
and Mars Tlii-se maps of Earth, Veniu. and

Mars are reproduced to the same scale. Each
map shows surface elevation (Earth's oceans
empty) on a color-coded scale. Elevations up
4 km above
to
the planet's average radius (sea
level for Earth) appear in shades of tan,
and
green,
white. Elevations
down
to
km
below the planet's average radius appear in

various shades of blue. Venus's two continents
are prominent, as are Earth's seven continents.
The
large protrusion on
Rise, a
Mars
is
the Tharsis
major ierrestnal planets
area 1000 km in diameter. They are among the largest known volcanoes
in the entire solar system and probably result from hot-spot volcanism.
A color-coded map of the Venusian surface is shown in Figure 7-26,
with comparable maps of Earth and Mars. One of the surprising results
from radar mapping is the discovery that Venus has no long mountain
chains resembling the Mid- Atlantic Ridge. Although Venus and Earth
have nearly the same mass and size, the Venusian surface shows very few
of the sure signs of plate tectonics that are so prominent on Earth's surface. Nevertheless, the existence of two continents does suggest that
some tectonic activity might be beginning there. Venus may just be
evolving more slowly than the Earth.
major volcanic reôn that includes
Olympus Mons.
(S. P. Meszaros;
NASA)
Interesting theories try to

explain the absence of water
on both Venus and Mars
Volcanoes give geologists important information about the gases that

probably were spewed into the original atmospheres of Venus, Earth,
and Mars. For example, during the Mount St. Helens eruption of 1980
(see Figure 7-27), geologists monitored substantial amounts of sulfuric
acid and other sulfur compounds emitted by the volcano. As we have
seen, these compounds are common in the Venusian atmosphere.
The major
terrrstrml planets
Carbon dioxide and water vapor are among the most abundant gases
Venus's atmosphere contains much carbon dioxide
but very little water. In contrast, the Earth has abundant water in its
oceans and atmosphere, but there is very little carbon dioxide in its atmosphere. If Venus and the Earth got their atmospheres mainly by the
same outgasing process, what happened to all the water on \'enus, and
where is all the carbon dioxide on Earth?
In the upper V'enusian atmosphere, intense ultraviolet radiation from
the Sun breaks water molecules into separate hydrogen and oxygen
atoms. The light hvdrogen atoms escape into space. Oxvgen. which is
one of the most chemicallv active elements, readily combines with other
substances in Venus's atmosphere. Thus Venus is left with almost no
water and anv water outgassed from active volcanoes is soon destroyed.
The carbon dioxide on the Earth is dissolved in the oceans and chemicallv bound into carbonate rocks such as limestone and marble that
lormed in those oceans. If the Earth became as hot as Venus, so much
.ubon dioxide would be boiled out of the oceans and baked out of the
rust that our planet would soon develop a thick, oppressive carbon dioxide atmosphere much like that of Venus.
Some scientists have recentlv proposed an interesting outline of
\'enus's earlv historv. Venus and Earth are so similar in size and mass
in volcanic vapors.
^HB^^^S
1Î^Hh^
^^bIB
i^^n^
/
Figure 7-27
May 1980
XJrMMBl/'' VS
Mount
Helens eruption of
A prepundemnce of datu suggi:sts
St.
thai the terrestrial planets obtained their at-
mospheres from volcanic oulgassing. Geologists

study
this
oulgassing process in eruptions of
volcanoes on Earth. This spectacular recent

eruption in the western United States was parttcularh well studied. (L'SGS)
that
it
is
reasonable to suppose that V'enusian volcanoes outgassed an
amount of water vapor roughly comparable t(j the total content of

Earth's oceans. Although some of this water on Venus might originally
have collected in oceans, heat from the Sun soon vaporized the liquid to
create a thick cover of water-vapor clouds. Calculations demonstrate that
this water vapor would have added 300 atm of pressure to the existing
90 atm of carbon dioxide. Thus the early Venusian atmosphere would

have weighed down on the planet's surface with a pressure of three tons
per square inch.
This thick, himTid atmosphere efficientlv trapped heat from the Sun,
creating a greenhouse effect far more extreme than the greenhouse effect that operates todav on Venus. Calculations demonstrate that the
ground temperature would have increased to 1800 K (2700F), which is
hot enough to melt rock. The Venusian surface was probably molten
down to a depth of 450 km (280 miles). Soon, however, dissociation of
the water molecules and a subsequent loss of hydrogen to space left behind the carbon dioxide atmosphere we find todav. Hostile as it may
seem, the modern environment on Venus is probably quite mild compared to that of earlier times.
A few photographs of the arid Venusian surface have come from Soviet spacecraft that landed on the planet. A typical panoramic \iew taken
in 1981
is
shown
in
Figure 7-28. Soviet scientists suggest that
this
region
was covered with a thin laver of lava that fractured upon cooling to
create the rounded, interlocking shapes seen in the photograph. This
hypothesis agrees with the analysis by the spacecraft's instruments, which
indicates that the soil composition is similar to lava rocks called basalt
that are common on Earth and the Moon.
Basaltic lava rock is also common on the Martian surface (see Figure
7-29). Chemical analysis of the Martian soil showed a very high iron content. Iron and silicon comprise about two-thirds of a typical Martian
rock's content. Surprisingly high concentrations of sulfur were also
found apparenth it is 100 times more abimdaiit in Martian rocks than
Earth rocks.
Each Viking lander had a scoop at the end of a mechanical arm to
obtain .samples of rock for analysis. Bits of rock were observed to cling
in
The
Figure 7-28
Venusian landscape
color view o] the Venusian surface has
ange tone because
TIil\
an
or-
the thick, cloudy atmosphere
absorbs the blue component of sunlight. The

rocky plates covering the
gtound may
be frac-
tured segments of a thin layer of lava, or they
may be crusty layers of sediment that have been

cemented together by chemical and wind
erosion.
(TASS)
to a
magnet mounted on
in the soil.
The
Digging in the Martian soil

arm with its small scoop
Viking's mechanical
protrudes from the right side of
this
view
the Chryse plains. Several small trenches
by the
.'coop
near the
left
in the
Martian
regolith
side of the picture.
oj
dug
appear
(NASA)
Inrrshml plmirts
thus confirming a high iron content

may simply
familiar reddish color of the Martian surface
be due to an abundance of rust (iron oxide).

As soon as measurements by spacecraft confirmed a low atmospheric
pressure on Mars, scientists realized that the Martian environment must
be extremely dry. Water is liciuid only over a certain range of temperature and pressure. If the atmospheric pressure above a body of water is
very low, molecules escape from the liquid's surface, causing the water to
vaporize. Any liquid water of Mars would boil furiously and rapidly
evaporate into the thin Martian air, so there are no oceans on Mars.
Although Valles Marineris was not formed by water erosion, a Marsorbiting spacecraft did photograph many features that look exactly like
dried-up riverbeds (see Figure 7-30). These features were totally unexpected because liquid water cannot exist today on Mars.
In 1976, the Viking spacecraft supplied evidence strongly supporting
the idea that liquid water once raged across the Martian surface in great
torrents. Numerous Viking photographs show erosional features that
look like the results of flash floods in the deserts of Arizona and New
Mexico. Scientists were faced with a problem that soon became known as
"the mystery of the missing water."
Researchers immediatel) looked to the Martian pcilar caps as a probable location of the missing water. It was not clear, though, how much of
Figure 7-29
tlie sccjop,
miijiii
\pf:
The major
Figure 7-30
(left]
An
This pair oj photographs
ancient riverbed
shoii's
a JOO-km-long
riverbed in A/an's heavily rratered southern
hemisphere. Features such as these were unex-
pected because liquid water cannot
now
exist
on the Martian surjace. This particular riverbed
is
located about
20 south of
the
Martian
equator. I.\.\S.\)
Figure 7-31 Irightj
cap
The northern polar
This I'iew oj the nolheni polar cap was
taken by a spacecraft in
1972 from a
distance
of 13,700 km (8500 miles). By obsen'ing the

rate of shrinkage with the approach of Martian
summer,
scientists
dediued that the
resid-
ual polar cap contains a substantial amount oj
frozen wain.
(NASA)
leireslrial platieb,
the polar caps consists of frozen carbon dioxide (drv

is
water
ice)
and how much
ice.
In 1972, a spacecraft had sent back photographs from orbit as sumto Mars's northern hemisphere. During the spring, the polar
mer came
cap receded rapidly (see Figure 7-31), suggesting strongly that a thin
layer of carbon dioxide frost was evaporating cjuickly in the simlight.
However, with the arrival of summer, the rate of recession abrupth
slowed, suggested that a thicker laver of water ice had been exposed. Scientists concluded that the residual polar caps that survive through the
Martian summers contain a large quantity of frozen water. Calculating
the volume of water ice in the residual caps is difficult, however, because
we do not know how thick the layer of ice is.
Another important clue about water on Mars came from the X'iking
Manv close-up views show flash-fltjod erosion features

where the water appears to have emerged from collapsed, jumbled terrain (see Figure 7-32). These photographs conHrmed earlier suspicions
that frozen water might form a laver of permafrost under the Martian
surface similar to that beneatii the tundra in far northern regions on
Earth. Apparenth, heat from volcanic acli\ii\ occasionalh melts this subsurface ice. The groimd then collapses as millions of tons of rock push
the water to the surface, producing a brief flash-flood as the water
orbiters in 1976.
quickly boils away.
From photographs such as Figure 7-32, scientists estimated the widths

and depths of the Martian flash-flood channels. These dimensions suggest peak flood discharges of 10' to 19'' mVscc. For comparison, the average discharge of the .\niazon is 10^ mVsec. The largest known flash
flood on the earth occurred about 2 million vears ago in eastern Washington, when a natural dam gave way. The peak discharge is estimated
to have been 10' mVsec. Most of the Martian flash floods must have
been greater than anything known on Earth.
The major lenrslnul planets
Evidence of a flash-flood on
ice and the
Figure 7-32
Mars The
melting of subsurface
subsequent collapse or downfaulting of the
Martian surface
easily explain the features
seen in the mosaic taken by Viking 1 in
1976
torrent of water liberated by this process
apparently flowed eastward

the
(to the left) across
surrounding plains. The area seen here
measures
300
by
300
km.
(NASA)
As mentioned
most
common
earlier,
water vapor and carbon dioxide are two of the
gases in volcanic vapors. Mars's surface gravity, although
only one-third as strong as Earth's, is nevertheless capable of keeping

these gases from escaping into space. Ultraviolet light from the Sun can
dissociate water into hydrogen and oxygen, which are light enough to
escape into space. Detailed calculations show, however, that Mars could
have lost only a small fraction of its water in this fashion. It is therefore
reasonable to suppose that much of the carbon dioxide and water vapor
outgassed during Mars's early history still remains on the planet. Some is
in the polar caps, some is in subsurface ice and permafrost. Some water
and carbon dioxide also may be chemically bound in the rocks and sand
that cover the Martian surface.
Earth's interior consists

of crust, mantle,
iron-ricli
core
and
The
densities of rocks you find on the ground are typically 3 to 4 g/cm^,

but the average density of the Earth as a whole is 5.5 g/cm^. Thus the
rocks of the Earth's crust are not representative of our planet's interior,
which must be composed of a substance much denser than the crust.
Iron is a good candidate for this substance because it

abundant of the heavier elements (recall Table 6-4), and
the Earth's interior
is
is
the most
its
presence in
strongly suggested bv the existence of the Earth's
Furthermore, iron is common in meteoroids that strike

it was common in the planetesimals from
which the Earth formed.
Geologists strongly suspect that the Earth was entirely molten soon
after its formation about 4i billion years ago. Energy released by the violent impact of numerous meteoroids and asteroids and by the decay of
radioactive isotopes melted the solid material collected from the earlier
planetesimals. Gravity caused the abundant, dense iron to sink toward
the Earth's center, forcing less-dense material to the surface. This process, called chemical differentiation, produced a layered structure within
magnetic
field.
the Earth, suggesting that
The major tenestrml planets
the Earth: a central core composed of ahnost pure iron, surrounded by

a mantle of dense, iron-rich minerals, which in turn is surrounded by a
thin crust of relatively light silicon-rich minerals.
The Earth's interior is as difficult to examine as the most distant gal-
The deepest wells go down only a few kilometers, barely

penetrating the surface of our planet. Geologists have, however, deduced
basic properties of the Earth's interior by studying earthquakes.
Over the centuries, stresses build up deep within the Earth's crust.
Earthquakes occur when these stresses are relieved with a sudden vibratory motion. The point on the Earth's surface directly above an earthquake's location is called the epicenter.
Earthquakes produce several different kinds of seismic waves that
axies in space.
around or through the Earth in different ways and at different

speeds. Geologists use sensitive seismographs to detect and record these
vibratory motions. Seismic waves are bent as they travel through the
Earth because of the varying density and composition of the Earth's intetravel
rior. By studying the deflection of these waves, geologists have discovered properties of the Earth's interior.
By 1906, analysis of earthquake recordings led to the discovery that
the Earth's iron core is molten and has a diameter of about 7000 km
(4300 miles). For comparison, the overall diameter of our planet is
12,700 km (7900 miles). More careful measurements in the 1930s revealed that inside the molten core is a solid iron core with a diameter of
about 2800
km
(1700 miles).
our planet therefore has a curious structure: a liquid
core sandwiched between a solid inner core and a solid mantle. To understand why this is so, we must examine the temperature and pressure
inside the Earth and their effects on the melting point of rock.
Both temperature and pressure increase with increasing depth below
the Earth's surface. The temperature of the Earth's interior rises steadily
from about 20C on the surface to roughly 6100C at the center (see Fig-
The
interior of
ure 7-33).
The
Earth's crust
is
only about 30
km
thick. It
whose melting points are far greater than

crust. Hence the crust is solid.
The
typical
is
composed of rocks
temperatures
Earth's mantle, which extends to a depth of about

is largely composed of minerals rich in iron
(1800 miles),
in the
2900 km
and magne-
the Earth's surface, specimens of these ferromagnesian minerslightly over 1000C. However, the melting point
of a substance depends on the pressure to which it is subjected: the
higher the pressure, the higher the melting point. As shown in Figure
sium.
als
On
have melting points
7-33, the melting point of the mantle's minerals is everywhere higher

than the actual temperature, so the mantle is solid.
At the boundary between the inantle and the outer core, there is an
abrupt change in chemical composition from ferromagnesian minerals to

almost-pure iron with a small admixture of nickel. This iron-nickel material has a much lower melting point than ferromagnesian minerals, so
the melting-point curve on Figure 7-33 drops abruptly as it crosses from
the mantle to the outer core. The melting-point curve in fact remains
below the temperature curve down to a depth of about 5000 km. Hence,
from depths of about 2900
to
5000
km
the core
is
liquid.
At depths greater than about 5000 km, the pressure is more than 3
million atmospheres. This pressure is so great that the melting point of
the iron-nickel mixture exceeds the actual temperature (sec Figure
7-33). Hence the Earth's inner core is solid.
The major
Figure 7-33
terre^lrial planets
Tempertature and meltingThe tempera-
point curves inside the Earth

lure rises steadily
its
center.
on
this
By
from
the Earth's surface to
plotting the melting point of rock
graph, we can deduce which portions
of the Earth's interior are solid or liquid.
The Earth's magnetosphere

shields us from the solar
wind, but neither Venus nor
Mars has a magnetic field
Although buried deep within the Earth, the liquid outer core has a great
influence on the Earth's outermost environment. Currents in the molten
iron in our planet's interior give rise to a planetwide magnetic field
through the dynamo effect. As the Earth rotates, these currents produce
a magnetic field, just as a loop of wire carrying an electric current generates a magnetic field. The Earth rotates fast enough to produce a magnetic field that dominates space for tens of thousands of kilometers and
dramatically affects Earth's interaction with the solar wind.
end of Chapter 6, the Sun is constantly

and protons escape from the Sun's
This constant streaming of matter away from the Sun is
As inentioned
briefly at the
losing matter as high-speed electrons
outer layers.
called the solar wind. The Earth's magnetic field carves out a cavity in
the solar wind, deflecting the flow around the Earth as shown in Figure
7-34. This cavity, called the Earth's magnetosphere, shields us from the
high-speed particles that would otherwise strike the upper atmosphere.
The speed of the particles escaping from the Sun is faster than the
speed of sound through the solar wind. Thus we say that the solar wind
is supersonic. Just as a supersonic airplane traveling through the air
creates a "sonic boom," a bow-shaped shock wave occurs where particles
of the supersonic solar wind are abruptly slowed to subsonic speeds. The
is the outer boundary of the Earth's magnetic domain.
Most of the particles of the solar wind are deflected around the inagnetopause through the turbulent region called the magnetosheath. Deep
inside Earth's magnetosphere, our planet's magnetic field is strong
enough to trap charged particles that manage to leak through the magnetopause. These particles are trapped in two huge, doughnut-shaped
rings called the Van Allen belts.
magnetopause
The major
Figure T-34
terrestrial ptatiets
Earth's magnetosphere
Earth's magnetic field can<es out a eavtt\ in

the solar wind.
The Earth's magnetosphere
consists of a shock wave, a magnetopame. and

a magnetusheath. Because of the strength of
Earth's magnetic field, our planet
is
iililr
la
trap charged particles in two huge, duughnutsliaped rings called the \'an Allen belts.
100.000
These
belts
were discovered
in
'
1958 during the
States' first successful Earth-orbiting satellite.
who
km
They
flight of the
are
named
United
after the
insisted that the satellite should carry a geiger
counter to
inner Van Allen belt extends over altitudes
of about 2000 to 5000 km and contains mostly protons. The outer Van
Allen belt contains mostly electrons and is about 6000 km thick, centered
at an altitude of about 16,000 km above the Earth's surface. It is remarkable that the Van Allen belts these two vast features that completely encircle the Earth
were totally unknown until a few decades
physicist
detect charged particles.
The
ago.
on the Sun's surface called a solar flare

protons and electrons toward the Earth. Many of these
particles penetrate the magnetopause and overload the Van Allen belts.
The excess particles move along the Earth's magnetic field and rain
down on the upper atmosphere near the Earth's north and south magnetic poles. As these particles collide with gases in the upper atmosphere, atoms of oxygen and nitrogen fiuoresce like the gases in a fiuoOccasionally, a violent event
sentis a burst of
rescent tube.
The
result
is
a beautiful,
shimmering display
called the
northern lights (aurora borealts) or southern lights (aurora auslralis), delicntling on the hemisphere from which the phenomenon is observed
(see Figiuc 7-35).
Numerous American and Soviet space flights have failed to detect an\'
magnetic field around Venus. This absence of a magnetic field might
seem surprising since Earth and Venus are so similar in mass, size, and
average density. It is in fact reasonable to assume that Venus has an interior structure quite similar to Earth's, with a substantial iron core some
of which is probably molten. Then whv doesn't Venus have a magnetic
field as Earth does?
Although Venus is shrouded in |)erpeiual cloud cover, scientists have
been able to examine the Venusian surface with radar. Pulses of radar
waves transmitted from Earth easily penetrate the Venusian clouds and
are reflected off the rocks below. By analyzing the reflected signals, scientists have been able to construct a radar image of the Venusian surface
(see Figure 7-36).
Radar studies of Venus revealed a reinarkable fact: Venus rotates
backward at a very slow rate. In other words, the Sun rises in the west
The major
Figure 7-35
borealis)
The northern
lights
(aurora
Aurorae are always occurring
limited region
around
the Earth's north
south magnetic poles. However,
terrestrial planets
in a
and
when a deluge
of protons and electrom from a solar flare

strikes Earth's
upper atmosphere,
especially
spectacular displays of glowing gases
may
be
seen over a wide range of latitudes. Aurorae

typically
occur at an altitude of about 110
(70 miles) above
Naval
km
the Earth's surface. (U.S.
Observatory)
Figure 7-36
radar
map
of Venus
Thi\
radar view of Venus shows about half of the

planet's northern hemisphere.
appear
bright,
Rough
areas
smooth areas are dark. The
prominent bright patch toward the upper right

is
the continent Ishtar.
patches on the lower

noes
The two smaller
left
bright
are the giant volca-
Rhea Mons and Theia Mons.
(Arecibo Obseniatoiy)
and sets in the east on Venus. A "day" on Venus (that is, the time from
one sunrise to the next) lasts for 16.8 Earth days.
Recall that Earth's magnetic field results from the dynamo effect: currents in the iron core of the rotating Earth produce our planet's magnetism. Although Venus may have a sizable liquid core, the planet's leisurely
1
rotation rate
is
apparently too slow to induce a planetwide magnetic
field.
With virtually no magnetic field, Venus is incapable of producing a

magnetosphere to protect itself from the solar wind. The solar wind
therefore impinges directly on Venus's upper atmosphere, where many
of the atoms become stripped of one or more electrons. The electromagnetic interaction between these ions and the supersonic charged particles
of the solar wind produces a well-defined shock wave (see Figure 7-37).
Along a boundary called the ionopause, inside the shock wave, the pressure of the ions just counterbalances the pressure from the solar wind.
The ionopause is analogous to the magnetopause surrounding a planet
with a magnetic field.
Although Mars rotates at about the same rate as Earth, Soviet spacecraft to Mars have detected only an extremely weak magnetic field there,
Tlif mtijar lmi\tru>l planiis
127
than 0.004 times as strong as Earth's. The near absence of a Martian

magnetic Held may be clue to the lack of a molten iron core. Both Venus
and Earth ha\e average densities greater than 5 g/cm\ indicating the
presence of substantial iron cores. But Mars's average density (3.9 g/cm"*)
is roughly that of ordinary crustal rocks. Thus Mars may not possess an
Earth-like iron-rich core. Indeed, Mars may not be as chemically differentiated as Venus or Earth. Mars's iron may be more uniformly distributed throughout the planet as evidenced by the high percentage of iron
in rocks on the Martian surface.
Mars's magnetic field is so weak that it is largely ineffectual in warding off the solar wind. In fact, the solar wind may actually impinge directly on the outermost layers of the Martian atmosphere, just as it does
on Venus. Unfortunately, our knowledge of this aspect of the Martian
environment is woefulh incomplete. An alternative possibility is that
Mars's weak field just manages to carve out a magnetosphere just barelv
enclosing the planet's atmosphere.
Searching for a Martian magnetosphere or listening to Marsquakes
with seismographs that would reveal an iron core can be done only by
future missions. The next American mission to Mars will use a small satless
lonopause
lo.onn
Figure 7-37
solar
wind
the solar
km
Venus's interaction with the

\'i'iiii.\
wind
lia.s
no magtuiic
strikes the
field.
the planet's atmosphere. Inleraclions luith
ions in the upper atmosphere produce a
shock
wave and lonopause,
this scale
drawing.
Summary
\ii
uppermost layers of
as
shown
in
scheduled to be launched in the 1990s. In orbit about Mars, this

use remote-sensing techniques to analyze the mineral composition of the Martian surface and also determine the role of water in the
Martian climate. The results could lead to a deeper understanding of the
Earth's weather and geology. This mission is the only one to Mars scheduled by the United States during the remainder of the twentieth century.
Although the possibility of Martian organisms seems quite remote,
some scientists point out that the polar regions of Mars may have conditions more suitable for life. These areas could be explored b^ another
mission to the Martian surface that could also send back data about the
Martian interior. Ideally, this ambitious undertaking would utilize a
rover vehicle that would send back data and pictures from a wide range
of sites. Another possibility is to land a spacecraft that would scoop up
some Martian rocks, blast off, and bring the samples back to Earth for
laboratory analysis. The Soviet Union has been particularlv successful
with missions of this tvpe to the Moon. In the 197()s, three of their unmanned prcjbes all returned samples of Moon rocks to Earth. Soviet scientists are vigorously platming a similar assault on Mars, with the ultimate goal of establishing a manned base there early in the twenty-first
centitry. A courageous mission of this magnitude would realize within
oitr lifetimes some of the great dreams of science-fiction writers.
ellite
satellite will
The
Earth's atmosphere is primarily nitrogen and oxygen, whereas the

atmospheres of Venus attd Mars are almost pure carbon dioxide.
Earth's atmosphere carr be divided into layers. .-Ml weather occurs in the
lowest layer, the troposphere, which extends up to an altitude of 1 1 km.
Venus
is
similar to the Earth in size, mass, average density,
gravity, but
and surface
covered by rreariy featureless unbroken cloirds.

nusian clouds consist of droplets of concentrated sulfur ic ai id.
it
is
The
Ve-
Ihe Verrusian cloud cover is confined to a 37-km-thick layer located 31

above the planet's surface. The atmosphere is clear from the bottom
krrt
ol lite
On
/.')()
(loud laver to
lire
grourrd.
ihe surface of Venus, the pressure is 90 aim aird liie terrrperature is

K. Ihe high lenipiraline is causrd h\ the greenhouse eiiect: carbon
The major
leni'stnal planets
dioxide in the atmosphere prevents infrared radiation from escaping into

space.
is smaller then either the Earth or Venus; a day on Mars is slightly
longer than a day on the Earth; Mars has polar caps that expand and
shrink with the seasons.
Mars
The atmospheric pressure on Mars

Earth's
is
about one-hundredth that of the
atmosphere and shows seasonal
variations.
Liquid water would quickly boil away in Mars's thin atmosphere, but the
polar caps do contain a considerable amount of frozen water. A layer of
permafrost may exist beneath the Martian surface.
Study of seismic waves shows that the Earth has a small solid inner core
surrounded by a liquid outer core; the outer core is surrounded by a
dense mantle, which in turn is surrounded by a thin low-density crust.
Earth's inner and outer cores are composed of almost-pure iron; the
mantle is composed of iron-magnesium minerals; the crust is largely composed of silicon-rich minerals.
The
The
Earth's crust and the upper part of its mantle are divided into huge
plates. Movements of these plates (called plate tectonics) are driven by
convective currents in the mantle. Plate tectonics is responsible for most of
the major features of the Earth's surface.
The
ing
surface of Venus is surprisingly flat, mostly covered with gently

there are two major continents and some large volcanoes.
roll-
hills;
The northern hemisphere
of Mars has numerous extinct volcanoes

whereas the southern hemisphere has numerous flat-bottomed craters; a
huge 4000-km-long canyon stretches along the equator.
The
surfaces of
Venus and Mars show
little
evidence of the motion of
large crustal plates that played a major role in shaping the Earth's surface.
The
magnetosphere
from the solar wind; charged particles from the solar wind
two huge doughnut-shaped rings called the Van Allen
Earth's magnetic field surrounds our planet with a
that shields us
are trapped in
belts.
Venus has no detectable magnetic

weak magnetic field.
field
or magnetosphere; Mars has only
a very
Review questions
Describe the various ways in which the Earth
is
different
from either
Venus or Mars.
2
Why
is
the Earth's surface not riddled with craters like the
3 Describe the process of plate tectonics.
Examine
map
Moon?
of the Earth and
identify the kinds of features that are a direct result of plate tectonics.
4 Describe the Earth's interior. What causes the Earth's inner core to be
whereas its outer core is molten. What gives rise to the Earth's mag-
solid,
netic field?
5 What techniques have astronomers used to examine and

of Venus? What kinds of surface features have they found?
6 Why is
reasonable to suppose that Venus's interior
Why doesn't Venus have a planetwide magnetic field as
it
is
map
the surface
similar to Earth's?
Earth does?
The
129
mtijor Irrreilrml plunels
7 Suppose you were in a spacecraft in orbit about Mars. What kinds ol^ surwould you see? What do these surface features tell you about
face features
plate tectonics
on Mars?
8 Compare Olympus Mons with the Hawaiian Islands. In what way are they
different manifestations of the same physical process?
Compare
and Mars.
on Earth?
the role of water on Venus, Earth,
and Mars very dry but water
still
exists
Whv
are Venus
10 With carbon dioxide just about as abundant in the Martian atmosphere as

Venusian atmosphere, why do you suppose there is no greenhouse
in the
effect
11
it reasonable to suppose that the polar regions of Mars might harbor

forms?
Is
life
Advanced questions
on Mars?
*12 What fractions of the Earth's
total
volume are occupied by the
core, the
mantle, and the crust?
*13 As mentioned in the text, Africa and South America are separating at a
rate of about 3 centimeters per year. Assuming that this rate has been constant, calculate when these two continents must have been in contact.
14 Suppose a planet's atmosphere were opaque to visible light but transparent to infrared radiation. How would this affect the planet's surface temperature? Contrast and compare this hypothetical planet's atmosphere with the
greenhouse effect in Venus's atmosphere.
The human
population on Earth
currently doubling about every 30

on the Earth by uncontrolled
human population growth. Can such growth continue indefiniteh ? If not,
what natural and human controls might arise to curb this growth? It has
been suggested that overpopulation problems could be solved by colonizing
the Moon or Mars. Do you think this is a reasonable solution? Explain voin'
answer.
15
is
years. Describe the various pressures placed
16 If you were designing a space vehicle to land on Venus, what special feawould be necessary? In what ways would this mission and landing craft
differ from a spacecraft designed for a similar mission to Mars?
tures
17 Imagine that you are an astronaut living at a base on Mars. Describe what
your day might be like, what you would see, the weather, the spacesuit vou
would wear, and so on. Suppose you and your colleagues have a motorized
vehicle for exploring the planet. Where would you like to go?
For further reading:
from a Torrid Planet." Sky

Telescope, Mav 1982. p.
452. Results of Soviet missions to Vemis.
Beatty. J. "Radar Views of Venus." Sky &" Telescope. Feb. 1984, p. 1 10.
Calder, N. Tlie Restless Earth. Viking, 1972. A good book on our modern view
of the Earth by a noted science writer.
Carr, M. "The Surface of Mars: A Post Viking View." Mercun, Jan. /Feb.
Beatty, J. "Report
1983, p. 2.
Carrigan, C., and Gubbins, D. "The Source of the Earth's Magnetic
Field." Scientific American, Feb. 1979.
Cooper, H. The Search for Life on Mars. Holt, Rinehart & Winston, 1980.
profile of the Viking project, by a well-knowti journalist.
The major
Gore, R. "Sifting for Life
in
Inreslruil planeli
the Sands of Mars." National Geographic, Jan.
1977.
Hartmann, W. "The Early History of the Planet Earth." Astronomy, Aug.

'
'
1978, p. 6.
Haberle, R. M. "The Climate of Mars." Scientific American, May 1986.
Pettengill, G., et al. "The Surface of Venus." Scientific American, Aug.
1980.
Pollack, J.
"The Atmospheres of the
al.,
eds. The
New
Terrestrial Planets," in Beatty,
Solar System,
2nd
ed. Sky Publishing
J., et
and Cam-
bridge University Press, 1982.

Prinn, R. G.
"The Volcanoes and Clouds of Venus."
Mar. 1985.
Scientific
American,
Jupiter, Saturn,
and Earth
The Jovian planets
Thu mimtagr
shows Juptter, Saturn, aixl Earth reproduced

to the
same
posed
pnmarih
the
scale.
Sun. Note that
urn are much

ter.
Jupiter
and Saturn are comand helium, like
of hydrogen
the cloud features
on Sat-
than those on Jupi-
less distinct
Saturn has a weaker surface gravity than
Jupiter's, which causes the
Satumian clouds
to
be spread over a greater range of depttis in
Saturn's atmosphere than in Jupiter's. Sunlight reflected
from these deeper cloud layers

amount of absorption on Sat-
suffers a greater
urn, giving the planet
compared
its
faded appearance
to Jupiter. (S. P.
Meszaros:
XASA)
Aboitl
7U percent
of the
mass
oj the solar system, besides the
centraled in the single giant planet Jupiter. In this chapter,
Sun itself, is conwe learn about
unusual features of this active, vibrant world whose multicolored, turbusurrounded by an enormous magnetosphere. We examine Salutii with its spectacular .system of thin, flat rings. We find that the rings actually consist oj thousands of ringlets composed of ice fragments and
ice-covered rock. We then di.scover that Jupiter and Saturn have similar intebut for different
rior structures and that they have internal sources of heat
reasons. We then turn to UraniLs and Neptune. We learn that these outer two
Jovian planets differ distinctly from the inner tii'o Jovian planets in many
ways. We find that Uranus has a unicjue orientation and an orbiting system
of thin, dark rings. For detailed information about Xeplune we airait the
1989 flyby of Voyager 2.
ihe
lent clouds are
The Jovian planets
More than any other
single factor, temperatures
throughout the young
solar nebula dictated the final characteristics of the planets that orbit the
Sun. In the warm, inner regions of this ancient nebula, the dust grains
that survived consisted primarily of metals, silicates, and oxides. The
temperature was too high to allow substantial condensation of such volatile substances as water, methane, and ammonia. The four planets that
formed close to the Sim were therefore composed almost entirely of
rocky material. Their surface gravities were too low and their surface
temperatures too high to retain any of the abundant but lightweight
hydrogen and helium gases that made up most of the solar nebula.
The four Jovian planets are much farther from the Sun than the terrestrial planets are (recall Figure 6-1). For example, Saturn is roughly
ten times farther from the Sun than Earth is. In the past, as now, it was
cold at these vast distances. In the young solar nebula, the dust grains so
far from the protosun were coated with a thick layer of frozen water,
methane, and ammonia. These volatile substances thus became important
constituents of the planets that eventually accreted in the outer reaches
of the solar system.
The Jovian planets may have started to form in much the same fashion as the terrestrial ones
by accretion of dust grains (coated with frozen gases, in this case) into a great number of planetesimals, which in
turn accreted to form huge protoplanets. Many scientists think, however,
that the Jovian planets were created in a two-step process. First, accretion led fairly quickly to the formation of four large protoplanets, each
several times more massive than the Earth. Then, the strong gravitational pull of these protoplanets attracted and retained substantial quantities of the hydrogen and helium in the cool outer reaches of the solar
nebula. Calculations show that this process of gathering lightweight gases
would have become very efficient and rapid after the protoplanets had
grown beyond
a certain mass.
The
final result
was the largest planets
in
the solar system.
Jupiter
is
a huge, colorful
world composed largely of

hydrogen and helium
Jupiter is huge: its mass is 318 times greater than Earth's. Indeed, the
mass of Jupiter is 2i times the combined masses of not only all the other
planets, but also all the satellites, asteroids, meteoroids, and comets in
the solar system. Jupiter's equatorial diameter is 1 U times as large as the
Earth's and its volume is about 1430 times larger than the Earth's.
Jupiter's average density can be computed from its mass and size: it is
only 1.33 g/cm\ This low average density is entirely consistent with the
picture of a huge sphere of hydrogen and helium, compressed by its
own gravity. Observations from spacecraft indicate that by weight Jupiter
is composed of 82 percent hydrogen, 17 percent helium, and only I percent all other elements. This composition closely matches that of the Sun
(recall
Table
6-4).
Through an Earth-based
telescope, Jupiter
is
a colorful, intricately
from a
and light
bands shaded in subtle tones of red, orange, brown, yellow, and blue
that are parallel to Jupiter's equator. The dark, reddish bands are called
belts, the light-colorecl bands are called zones. These are not the only
banded sphere
spacecraft.
(see Figure 8-1). Figure 8-2
The most prominent
conspicuous markings.
is
a close-up view
features are alternating dark
large, reddish oval called the
Great Red Spot
is
often visible in Jupiter's southern hemisphere. This remarkable feature,

which has been observed since the mid- 1600s, appears to be a long-lived
storm in the planet's dynamic atmosphere. Many careful observers have
rill-
Figure 8-1
Jupiter from Earth
[left]
and zones are

system.
The Great Red Spot was
prominent when
this
photograph
exceptionally
zi'as
taken.
(McDonald Observatory)
Figure 8-2 [right]
spacecraft
spacecraft in
Thi.s
600 km
Jupiter from a
view was sent hack from a
1979
million kilometers
.small as
at a di.<.tance of only
from
30
the planet. Features as
ter's
cloud motions surround the Great
last
for
onK
tew weeks or months
in Jupi-
turbulent clouds.
Although it is the largest and most massive planet in the solar system,
Jupiter has the fastest rate of rotation. At its equatorial latitudes, Jupiter
completes a full rotation in only 9 hours 50 minutes 30 seconds. However, Jupiter does not rotate like a rigid object. Ihc polar regions of the
planet rotate a little more slowly than do the equatorial regions, which
we can seen by following features in the belts and zones. Near the poles,
the rotation period is about 9 hours 55 minutes 41 seconds. The first
person to notice this differential rotation of Jupiter was the Italian as-
across can be seen in the tur-
bulent cloudtops of this giant planet. Complex
Spot.
reported smaller spots that
Belli
easily identified in this F.arth-
based view of the largest planet in the solar
Joviati f)lanels
tronomer Giovanni Domenico Cassini

observer
Red
who gave
us the
first
in 1690. This was the same gifted

accurate determination of Mars's rotation
rate.
IXASA)
Jupiter's colorful cloudtops are the turbulent, uppermost layer of its
thick atmosphere. .Are the observed rotation rates of these clouds anvthing like the rotation rates of deeper levels or of a solid central core?
Some intriguing clues come from the fact that radio waves that Jupiter
emits vary slightly in intensity. In particular, radio emissions with wavelengths in the range of 3 to 75 cm vary with a period of 9 hours 55
minutes 30 seconds. This radio emission is believed to be directly associated with Jupiier's magnetic field, which is anchored deep inside the
l)lanei.
Ihiis radio observations reveal Jupiier's internal rotation period,
which is slightly different from the atmospheric rotation we (.m ()h^crvc

through a telescope.
The Great Red Spot
is
high-pressure anticyclone
During the 197()s, four American spacecraft flew past Jupiter and sent
back spectacular, close-up pictures of its dynamic atmosphere. Short-term
hanges in Jupiier's cloud cover are most apparent in the area around
the Great Red Spot. Over the past three centuries. Earth-based observers
have also reported many long-term variations in the spot's size and color.
I
The Jovian planets
Figure 8-3
Spot
Changes
These two
the Great
Red
in the
vieu'S,
show major changes
Great Red
taken î years apart,
in Jupiter's clouds aroun''
Spot, (a)
7"/iis
view was taken
a distance of 545.000 km from

cloudtops. (b) This view was taken in 1979
in
1974
at
from a distance of 6
million kilometers.
li:,
Notui
the dramatic increase in turbulent cloud activity
that occurred between the two
spaceflights.
(NASA)
At its largest, the Great Red Spot was so huge that three Earths could
have fit side-bv-side across it. At other times (as in 1976-1977), the spot
faded from view. During two flybys in 1979, the Great Red Spot was
only slightlv larger than the Earth.
Figure 8-3 shows two contrasting \iews of the Great Red Spot. In
1974, the Great Red Spot was embedded in a broad white zone that
dominated the
planet's southern hemisphere.
the cloud structure had changed dramatically.
A few years later, however,

A dark belt had broad-
ened and encroached on the Great Red Spot from the north, and the
much greater turbulence.

Careful examination of cloud motions in and around the Great Red
Spot reveals that the spot rotates counterclockwise with a period of about
six days. Furthermore, winds to the north of the spot blow to the west,
whereas winds south of the spot move toward the east. The circulation
around the Great Red Spot is thus like a wheel spinning between two
oppositely moving surfaces (see Figure 8-4). This surprisingly stable wind
pattern has survived for at least three centuries.
The Great Red Spot is a high pressure system that protrudes above
the surrounding cloudtops. .'Xnvone familiar with weather forecasting
knows that the Earth's weather is dramatically affected by high-pressure
and low-pressure systems. A high-pressure system (commonly called a
"high") is simply a place where a more-than-usual amount of air happens to be located. This excess air weighs down on the Earth's surface,
resulting in high atmospheric pressure on the groimd. .\ low-pressure
system (commonly called a "low") is a place where there is a less-thannormal amount of air, resulting in low atmospheric pressure on the
ground. If we could see air, highs would look like bulges in the atmosphere (where extra amounts of air are piled up) and lows would look
like depressions or troughs.
Gravity has a strong influence on the basic dynamics of the Earth's
weather. Gravity causes the air to flow "downhill" from the high-pressure bulges into the low-pressure troughs. But because the Earth is rotating, the wind flow from the highs toward the lows is not along straight
entire region was apparentlv embroiled in
Figure 8-4
Circulation around the Great
Red Spot The
Great
Red
Spot spins counlti-
clockwise. completing a full revolution in
about
and
six days.
.Meanwhile, winds to the north
south of the spot blow in opposite direc-
tions.
Cotuequenlly, circulation associated with
the spot resembles a wheel spinning between
two oppositely moving surfaces. (Adapted from
Andrew
P. IngersoU)

Tht Jovian jAanets
Figure 8-5
windflaws
cyclonic
Cyclonic
and
135
anticyclonic
Because of a plunet's rotation,
wind flou'ing
into
a low-pressure
re-
gion or anticyclonic winds flowing out of a

high-pressure region rotate either clockwue or
counterclockwise, depending on the hemisphere
in which the weather system
is
located.
Instead, the Earth's rotation causes a deflection of the winds, producing either clockwise or counterclockwise flow about the high- and
lines.
low-pressure regions.
Figure 8-5 shows the resulting wind flow about highs and lows. In
Earth's northern hemisphere, winds blowing toward a low-pressure region rotate counterclockwise about the low, forming a cyclone. Winds
blowing aw'ay from a high-pressure region rotate clockwise about the
high, forming an anticyclone. In the southern hemisphere, the directions of rotation are reversed: cyclonic winds rotate clockwise and the
anticyclonic winds rotate counterclockwise.
This basic pattern of wind flow is the same on Earth, Jupiter, or anv
other rotating planet. The six-dav rotation of the Great Red Spot is
counterclockwise. Thus the (ireat Red Spot is an anticyclone, a long-lasting high-pressure bulge in Jupiter's southern hemisphere.
Studies using computers

help explain some
of the phenomena in
Jupiter's clouds
The llyby pictures showed other anticyclones in Jupiter's southern hemisphere. These features appear as white ovals like those in Figure 8-'M>.
The wind flow in these ovals clearly is coiuiterclockwise.
Most of the white ovals are observed

sphere.
in Jupiter's
southern hemi-
Brown ovals are more common in its northern hemisphere (see

8-6). The white ovals are the high-altitude cloudtops of high-
Figure
pressure systems, but the brown ovals result from holes in Jupiter's
cloud cover that permit us to see down into warmer regions of its atmosphere. Like the Great Red Spot, a white oval is apparently long-lived
Earth-based observers have reported them in the same locations since
1938. However, a brown oval lasts for only a veai or two. Computergenerated Figure 8-7 shows how Jupiter would look if vou were located
directly over the planet's north or south pole. The regular spacing of
cloud features such as ripples, plumes, and light-colored wisps is also
obvious.
Tllr Inviini pinliils
A brown
Figure 8-6
ox'tils
oval
earned iv openings
the
Large brown
hemisphere are
in Jupiter's northern
mam
cloud layer
that reveal loarm. dark-colored gases below.
The length of
this
oval
ts
roughly equal
to the
The spacecraft Voyager 1

kilometers from Jupiter when
Earth's diameter.
was 4 million
this
picture ims taken.
(NASA)
Compuier processing was
also used to "unwrap" Jupiter, producing a

views of the planet in Figure 8-8. Note the changes that
occurred during the four months between the tlybys of Voyager 1 and
Voyager 2. The regular spacing of light-colored plumes is apparent in
the equatorial regions of both pictures. The Great Red Spot moved westward and the white ovals moved eastward during the interval between
map
like the
he two
flvbys.
These beautiful Voyager photographs might suggest
that Jupiter's
clouds are the result of incomprehensible turmoil. You may wonder if

there is anything constant in the [ovian atmosphere. Surprisingly, there
is. Telescopic observations over the past eighty years and the Voyager
data demonstrate that the wind speeds in the Jovian atmosphere are
remarkably stable. Although Jupiter's colorful bands change quite rapthe underlying wind patterns do not.
Jupiter's persistent wind patterns consist of broad streams of counterflowing eastward and westward winds. Wind speed near Jupiter's equator
is roughly 100 m/sec (220 miles per hour) and several reversals of wind
direction occur between the equator and the poles.
Computer simulations involving whirlpools and eddies caught between
counterflowing streams help us understand both the long- and shortterm features in Jupiter's clouds. Andrew P. Ingersoll and his colleagues
at the California Institute of Technology have pioneered the development of these calculations. Figure 8-9, for example, shows the beha\ior
of a small, unstable whirlpool, technically called a vortex. This whirlpool
is spinning too slowly to remain intact and so is torn apart by the counterflowing winds. Larger, rapidly rotating vortices do survive, however,
in these simulations. The white ovals and the Great Red Spot endure by
simply rolling with the wind currents (recall Figure 8-4). Figure 8-10
shows a simulation in which two stable vortices merge to form a larger
one. The long-lived white ovals apparently maintain themselves in this
idly,
Soutli pole
The northern and southern
Figure 8-7
hemispheres
struct these
Computers were used
viezi's
Jupiter's north
that look straight
and
to
con-
down
onto
south poles, (a) In the
northern view, notice that light-colored plumes

are evenly spaced around the equatorial
re-
giom. Several brown ovals are vUible. (b) In

the southern view, notice that the three biggest
white ovals are separated from each other by
almost exactly 90 of longitude. In both views,

the
banded belt zone structure
the poles.
The ragged
photographed by
black
is
absent near
.spot is
the spacecraft.
an area not
(NASAI
fashion.
The Jovian planets
vitw
a \<)va^cr
b Vovagii
comparison of
Computer
prucesiinK produced these two "unwrapped"
\news of JupHer from (a) Voyager 1 and
Figure 8-8 fabove]
Voyager
and Voyager 2 views
(b) Vo\aêr 2.
Each view was aligned with

magnetic axis so
respect to Jupiter's
placements
lo the right
or
left
tliat dis-
represent real
cloud motions. Notice that the Great
moved westward while
Red Spot
the while ovals
moved
eastward during the four mouths between the

twoflybys.
(NASA)
Figure 8-9
[left]
unstable vortex
a small vortex
is
The demise of an
hi
this
computer simulation,
rotating loo skrwh lo remain
intact.
After slighlh more than one week, the
vortex
is
pulled apart between counterflowing
zonal jets. (Adapted from .Andrew P. IngersotI)
Figure 8-10 fright]
The merging and
maintenance of stable vortices
This com-
puter sniiulatiou shows the collision
and
merger of two rapidly spinning, stable
The
result
is
a larger vortex
and
vortices.
the ejection
of some material. Mergers of this type occur

around the Great Red Spot. (Adapted from
Andrew
I'.
Ingersolll
15 days
The Juvian planets
The magnificent
Saturn's spectacular rings

are composed of fragments
of ice and ice-coated rock
rings of Saturn
make
planet one of the most specan amateur astronomer using a

away, however, that our Earth-based telthis
tacular objects in the nighttime sky for
small telescope. Saturn
is
so far
escopes can reveal only the coarsest, large-scale features.

In 1675, G. D. Cassini discovered a dark division in the rings that
looks like a gap about 5000 km wide. Astronomers also discovered
stripes in Saturn's clouds similar to the belts and zones on Jupiter (see
Figure 8-11). The contrast between Saturn's own belts and zones is less
dramatic than the colorful patterns in Jupiter's atmosphere. Also, Saturn's stripes are not restricted to the equatorial regions as are Jupiter's;
Saturn's alternating dark and light bands extend into the polar regions.
After Cassini's discovery of a gap in Saturn's ring, astronomers began
to view the ring as a system of rings. The Cassini division separates the
view
one of the
is
best ever
an Earth-based
by
color images taken
1974
Tim
Saturn from the Earth
Figure 8-11
produced of Saturn
observatory. Sixteen original
during
the
same night
in
with the 1.5-m telescope at the Catalina
Observatory were combined
to
make
this
photo-
graph. Note the prominent Cassini division in

the rings
and
Figure 8-12
and
(NASA)
the belts
nian atmosphere.
Saturn's classic rings
three broad rings of

in this
Saturn are clearly
photograph sent back by Voyager
The faint
to
zones in the Satur-
ring
exists in the
The
visibtr
I
region of 1.21
1.53 Saturn radii from the planet's center.
The
bright
1.53
to
ring occupies the region from
1.95 Saturn
Cassini division
the
ring,
lies
radii.
The 5000-km-wide
between the
ring
and
which occupies the region between
about 2.03 and 2.26 Saturn
radii.
(NASA)
The Juvmii
outer
pliinets
ring from the brighter
1800s, astronomers using
ring closer to the planet. By the mid-
improved telescopes were able
to detect a faint
ring (or crepe ring) just inside the B ring (see P'igure 8-12).
Earth-based views of the Saiurnian ring system change dramatically as
Saturn orbits slowly about the Sun (a Saturnian year is equal to 29*
Earth years). This change can be observed because the rings, which lie in
the plane of Saturn's equator, are tilted 29 from the plane of Saturn's
orbit. Thus, over the course of a Saturnian year, the rings are viewed
from various angles by an Earth-based observer (see Figure 8-13). At
one time, the observer looks "down" on the rings; half a Saturnian year
later, the "underside" of the rings is exposed to view from Earth. .\t intermediate times, the rings are seen edge-on, when they disappear enless
tirely from our view. Thus w'e conclude that the rings are very thin
than 2 km in thickness according to recent estimates.
Figure 8-1}
The changing appearance of
Saturn's rings
from
Saliini's
ruip are
tilled
2T
the plane of Saturn's orbit. Earth-hased
observers thus see the rings at various angles

as Saturn moves
around
its
orbit.
Note
the rings .seem to disappear entirely
lliat
when
are viewed edge-on. (Lowell Observatory)
they
Till-
Astronomers
known
more
Jnviaii pliim-ls
century that Saturn's rings

f he Scottish
physicist James Clerk Maxwell proved mathematically in 1857 that such a
broad, thin, rigid sheet would break apart and he concluded that Saturn's rings are composed of "an indefmite number of unconnected particles." In fact, the rings consist of millions of tiny moonlets, each circling
Saturn along its own individual orbit.
Saturn's rings are very bright so the particles that form the rings must
be highly reflective. Astronomers had long suspected the rings to be
made of ice and ice-coated rocks, but confirming evidence was not obtained until the early 1970s, when astronomers identified the spectral
features of frozen water in the near-infrared spectrum of the rings. In
cannot
pf)ssibly
liave
for
be solid, rigid,
tlian a
tliin siieets of
matter,
Voyager
and Voyager 2 traveled
Measurements from these spacecraft tell us that the temper
atine of the rings ranges from only
180C1 290F) in the sunshine lo
less than 200C 330F) in Saturn's shadow. Water ice is in no danger
the early 1980s, two spacecraft
past Saturn.
of melting or evaporating
at these
temperatures.
Voyager observations also showed that the ring particles range in size
from snowflakes less than
mm in diameter up to icy boulders tens of
meters across. It seems reasonable to suppose that all of this material is
1
ancient debris that has failed to accrete into

particles are so close to
Saturn's rings consist
satellites.
In fact, the ring
Saturn that they can never accrete into
satellites.
During the Voyager flybys of 1980 and 1981, their cameras sent back
pictures showing unexpected details in the ring structures of Saturn. Its
broad rings were seen to consist of hundreds upon hundreds of closely
spaced thin bands, or ringlets, of particles (see Figure 8-14). Although
intriguing suggestions have been proposed, scientists still do not understand just why Saturn's A, B, and C rings are divided into these thou-
thousands of
closely spaced ringlets
of
sands of ringlets.
The Voyager cameras also sent back the first high-quality pictures of
the F ring, a thin ring visible just beyond the outer edge of the A ring
in Figure 8-14. C^lose-up views revealed a startling and mysterious fact:
the F ring
is
kinky and braided, actually consisting of several intertwined

One Voyager image shows a total of five
strands (see Figure 8-15).
strands each about 10
Figure 8-14
Vvyager
Saturn's rings from Voyager
look this view of Saturn's rings
from a distance of about 1} million kilometers.
C nng
The
or
scatters light differently
rings do, so that
broad Cassini division

the
The
it
is
clearly visible, as
narrow Encke division within

thin
edge of the
ring
is
ring.
than the
has a bluer color. The
visible just
(NASA)
the
hewnd
is
ring.
the
miln
km
across.
Voyager
scientists
were
at a loss to ex-
Tlir Jdviini pluiifli
Figure 8-15
Voyager
Details of the
plwtuf^mplt of the
ring
from a distance of 750.000 km. The

width of the
this
ring
is
This
ring wa.s Uikni

total
about 100 km. Within
span are several discontinuous strands
each roughly 10
km
across.
(NASA)
complex structure. The braids, kinks, knots, and twists in the F

ring pose one of the most challenging puzzles in modern astronomy.
plain this
Figure 8-16
Details of the
[left]
ring
This view of the underside of Satiirn\ rings
was taken from a distance of 740.000 km b\

Voyager 1. Both the Encke division and the
thin
ring are clearly visible toward the right
side of the picture.
The Cassini
the outer edge of the
The Cassini
side.
in this
rings.
division
ring are on the
and
left
division appears bright
view of the shaded side of the
(NASA)
Through Earth-based telescopes, we see only the sunlit side of SatFrom this perspective, the B ring appears very bright, the A
urn's rings.
ring moderately bright, the C ring dim, and the Cassini division is dark.
sunlight reflected back toward the Sun is directly related to the density of the fragments or particles in the ring. The B ring
is bright because it has a high density of ice and rock fragments,
The proportion of
whereas the darker Cassini division has a lower densitv of fragments.

Eight hours after it had crossed from the northern to the southern
side of the rings. Voyager
took the photograph in Figure 8-16. The
Sun was shining down on the northern side of the rings at that time, so
Figure 8-16 shows the sunlight that passes through the rings. As expected, the B ring looks darkest here because little sunlight gets through
its dense concentration of fragments, whereas the Cassini division looks
bright because sunlight passes relatively freely through its low densitv of
fragments. However, the fact that the Cassini division does ajjpear bright
1
Figure 8-17 [right j

ring details
False-color view of
(Computer processing
.\eiierely
exaggerates subtle color z'ariations in
this
view
from Voyager 2 of the sunlight side of the

rings. Note that the C ring and the Cassini
division
appear
bluish. Also note distinct color
I'ariations across the
A and B
rings.
(NA.SA)
The
/(ii'iaii
j>lanets
clear evidence that it does contain some fragments. If it contained no

fragments at all, we would see the black of space through it. The brightness in Figure 8-16 must be sunlight scattered by some fragments in the
is
Cassini division.
Another
A ring is visible in Figure

km and is named the Encke divi-
division in the outer half of the
8-16. This division has a width of
270
sion after the German astronomer Johann Franz Encke, who reported
seeing it in 1838. Many astronomers have argued, however, that Encke's
report was erroneous because his telescope was not good enough to produce an image of this narrow gap in the rings. The first undoubted ob-
Encke division was made in the late 1880s by the American astronoiner James Keeler, using the newly constructed 36-in.
refractor at the Lick Observatory. For this reason, the Encke division is
sometimes called the "Keeler gap."
Subtle color differences from one ring to the next also give important
clues about the chemical composition of the particles in the rings. These
differences are clearly visible in Figure 8-17, in which the colors have
been exaggerated by computer processing. The main chemical constituent is frozen water, but trace amounts of other chemicals (perhaps coating the surfaces of the ice particles) are probably the cause of the colors
seen in this computer-enhanced view. Although these trace chemicals
have not been identified, the existence of color variations that have
probably persisted for millions of years suggests that the icy particles do
not wander around or migrate substantially from one ringlet to the next.
servation of the
Saturn's innermost satellites

affect the appearance and
structure of the rings
Astronomers had realized before the Voyager fiybys that one of Saturn's
moons, Mimas, has an effect on the ring system. Mimas is a moderatesized satellite that orbits Saturn every 22.6 hours (see Figure 8-18). According to Kepler's third law, particles in the Cassini division should
orbit Saturn every 11.3 hours. Consequently, on every second orbit, particles in the Cassini division line up between Saturn and Mimas. During
these repeated alignments, the combined gravitational forces of Saturn
and Mimas cause small fragments to deviate from their original orbits.
In this way, the 2-to-l resonance with Mimas depletes the Cassini division of dust that would otherwise scatter sunlight back toward Earth,
which is why Earth-based astronomers see the division as a dark band.
The Voyager cameras also discovered three new ring systems: the D,
E, and G rings. The D ring is Saturn's innermost ring system. It consists
of a series of extremely faint ringlets located between the inner edge of
the C ring and the Saturnian cloudtops. The E ring and the G ring both
lie quite far from the planet, well beyond the outer edge of the A ring.
Both of these outer ring systems are extremely faint, fuzzy, and tenuous.
Each lacks the ringlet structure so prominent
The E
Mimas Mimas Ihe unallest

and innermost of Saturn 's six moderate-sized
Figure 8-18
satellites.
This view was taken by Voyager
1 at
500,000 km. The huge imnamed Arthur after the legendary

English king,
130 km in diameter. Mimas
is only -100 km in diameter. (NASA)
a range of nearly
pact crater,
i.s
in the
main ring systems.
along the orbit of Enceladus, one of Saturn's icy satellites

(see Figure 8-19). Some scientists suspect that geysers on Enceladus are
the source of ice particles in the E ring.
The Voyager cameras also discovered two tiny satellites that follow
orbits on either side of the F ring (see Figure 8-20). The gravitational
forces of these two satellites keep the F ring particles in place. The outer
satellite moves around Saturn at a slightly slower speed than that of the
ice particles in the ring. As the ring particles pass by this satellite, they
experience a tiny gravitational tug that tends to slow them down. These
particles thus lose a little energy, which would cause them to fall into
ring
lies
The Juviaii
Figure 8-19 Heft]

resnlutiim imairi'
iif
Enceladus
Eiitfludu.s
l>linuis
Tin-. Iii^h^
ims
nhkiDii'fl In
Vuwger 2 jrom a dislance of 191,000 km. Iff

jhws and crufks strongly suggest that the surface has been subjected
tivily.
The youngest
estimated to be
old.
less
to recent
geological en
crater-free ice flows are
than 100.000 vcniv
(XASA)
uring about 50
The F ring and its

Two lin\ satftliles. eufh meu\km across, orbit Saturn on fi-
Iher side of the
Figure 8-20 [rightl
two sheperds
fefts o/ these turn
ring.
The gravitatiorml
shepherd
confine the particles in the
about 100
km
satellites
ring
wide. This Vo\ager
was taken from a range of lOi

kilometers.
ef-
focus and
to
a band
piclurf
million
(NASA)
orbits a
This
little
closer to Saturn
satellite orbits
As the
satellite
except for the effect of the inner
the planet a
moves
little
satellite.
faster than the F ring particles do.
past the particles,
its
gravitational pull tends to
speed them up, thus nudging them into a slightly higher orbit. The combined effect of these two satellites focuses the icy particles into a welldefined narrow band about lUO km wide. Because of their confining influence, these two moons are called shepherd satellites.
A shepherd satellite that circles Saturn just beyond the outer edge of
the A ring is responsible also for the sharp outer edge of the A ring. As
particles near the edge of the A ring pass by the slovvlv moving shepherd satellite, thev feel a gravitational drag that slows them down
slightlv, piexenting them from wandering into orbits farther from Saturn.
The atmospheres of Jupiter

and Saturn each have three
main cloud layers
As you nnght expect, it is warm deep inside the atmospheres of Jupiter

and Saturn but cooler at their cloudtops. Infrared measurements of Jupiter confirm that the temperature rises as the depth within the )o\ian
cloud cover increases. This situation is analogous to the Earth's atmosphere, which is cool at the highest cloudtops but is warmer near the
ground.
()\cr the range of temperatures in the Jovian atmosphere, gases emit
energv primarilv as infrared radiation. Figure 8-21 shows nearly simultaneous photographs of Jiipiier at infrared and visible wavelengths. In the
infrared pittine. the brighter parts of the image correspond to hotter
temperatures. There is also a striking correlation between brightness in
the infrared image and color in the visible-light image. In other words,
the various colors in Jupiter's clouds correspond to differing temperatures and hence to differing depths in the atmosphere. (It is customary
to discuss these features in terms of depths measured from the cloudtops rather than altitudes above the surface because the exact location of
The jovHin platwts
Figure 8-21
Infrared and visible views
(a) This infrared phnlogrnjih
was taken
through the 200-in. Patomar telescope. The

brightest parts
of the image correspond
in the clouds where deeper
of the
to holes
and warmer regions

visible. Dark
Jovian atmosphere are
parts of the image correspond to

cloudtops. (b) This
image
tlie
cool
in visible light
was
taken b\ Vo'âger I at almost the same time the
infrared picture was taken. Comparisons be-
tween the two photographs show tlmt cloud

color
is
correlated with depth in the Jovian
atmosphere. The bluish
and brown
clouds are
roughh 100C u'armer and 100 km lower

than the red and whitish clouds. (N.ASA)
any
on Jupiter
solid surface
unknown.) Bluish clouds correspond
is
to
the brightest parts of the infrared picture, so these clouds must be the
warmest and hence the deepest layers that we can see in the Jovian atmosphere. Brown clouds form the next highest layer, followed bv whitish clouds and red clouds in the highest layer.
The Jovian atmosphere has a miniminn temperature of about 160C
260F) at an altitude above the cloudtops where the atmospheric pres(
sure
is
about 100 millibars.
similar
minimum
occurs at the 100-millibar
atmosphere where the temperature is 180C 290F).

By analogv with the Earth's atmosphere (recall Figure 7-2), we can call
this level the boimdarv between the stratosphere and the troposphere.
As on the Earth, all the weather on Jupiter and Saturn takes place below
level in Saturn's
the stratosphere.
Saturn
-NHj
Figure 8-22
"spread nut" than Jupiter's, which

result
is
is
upper atmospheres of Jupiter and Saturn is

displayed
these graphs of temperature versus
depth. Note that Saturn's atmosphere
NH4HS
Temperature profiles of
The structure of the
Jupiter and Saturn
H2O
more
a direct
l\
of Saturn's weaker surface gravity.
(Adapted from
.\ndreit' P. Ingersolt)
Temperature
(C.)
-rempeialurc ("C)
The Joi'ian planets
Figure 8-23
Saturn's clouds
[left]
from
Voyager 1 Thts vmv uj Saturn's cloudtops

was taken by Voyager I at a range of
mil-
lion kilometers.
less contrast
Xote that there
between
belts
is
mbstanlialh
and zones
here than
on Jupiter. The shadow of Dione appears at

the bottom of the picture.
(NASA)
Eddy currents
Figure 8-24 [right]

Saturn's atmosphere
Computer
in
proce.V:ing
exaggerates the colors in this \'o\ager
picture
of Saturn's northern mid-latitudes. Tlie wai'y

line in the light blue ribbon
is
a pattern mov-
ing eastward at IHO misec (300 miles per

hour).
The dark
white spots below
ward
at roughly
hour).
(XASA)
oi'ol
it
20
and two
puffy, blue-
are eddies drifting westmJsec (40 miles per
Spectroscopic observations reveal that the atmospheres of Jupiter and

Saturn contain methane (CH4). ammonia (NH,). and water vapor (HoO).
These compounds are the simplest combinations of carbon, nitrogen,
and oxygen with hvdrogen. From calculations of the behavior of these
chemicals under various conditions of temperature and pressure, scientists conclude diat Jupiter and Saturn both ha\e three main cloud layers.
The uppermost laver, with red and whitish clouds, is composed of crystals of frozen ammonia. Deeper in the troposphere, ammonia (NHO and
h)drogen sulfide (H^.S) combine to produce a mid-level cloud layer of
ammonium hvdrosulflde (NH^SH) crystals. The third and deepest cloud
layer has a bluish tint and is composed of water droplets and snowflakes
of frozen water (see Figure 8-22).
Scientists still do not know what gives the clouds their various colors.
Crystals of ammonia, ammoniiun hydrosulf'ide, and frozen water in the
main cloud layers are all white. The browns, blues, reds, and oranges
therefore must be due to other chemicals or perhaps organic compoiuids.
.Although their atmospheres are similar in structure and composition,

Saturn and Jupiter are not identical in appearance. Saturn's clouds lack
the colorful contrast of Jupiter's, although some of the X'oyager photographs do show faint belts and zones (see Figure 8-23). After computer
processing to enhance the colors in the photographs, details such as
storm systems and ovals became visible (see Figure 8-24).
The differing appearances of Jupiter and Saturn are related to their
different masses. Jupiter's strong surface gravity compresses its three
cloud layers into a range of 75 km in the upper atmosphere. Saturn's
somewhat weaker surface gravity subjects its atmosphere to less depression, so the same three cloud layers are spread out over a range of
nearly 300 km (see Figure 8-22). The colors of Saturn's clouds are less
diamatic because the deeper layers are partly obscured by the thick atmosphere above them.
By folUjwing features in the Jovian and Saturnian clouds, scientists
have determined wind speeds in their upper atmospheres. Both planets
exhibit counierllowing eastward and westward currents. However, Salurn's equatorial jet is much broader and faster than Jupiter's (see Figure
8-2.")). In lad, wind sijeeds near Saturn's etjuator approach 500 m/sec
(1000 miles per hour), approximately two-thirds the speed of sound.
The Jovian planets
Wind speeds on Jupiter and

Average wind speeds on juptler and
Saturn are here plotted for latitudes ranging
from 80 north to 80 south. The positive
Figure 8-25
Saturn
numbers are eastward
velocities; negative
are
westward. Although both planets exhibit counterflowing currents, Saturn's equatorial zonal
jet is
miuh broader and faster than
Jupiter's.
100
Jupiter and Saturn emit
more
radiation than they
receive from the
Sun
Eastward wind
200
300
Eastward wind
speed (m/sec)
speed (m/sec)
100
200
100
Both Jupiter and Saturn have internal sources of energy. Each planet
radiates more energy than it receives in the form of sunlight. Many scientists believe that the excess heat escaping from Jupiter is energy left
over from the formation of the planet. As gases from the solar nebula
fell into the protoplanet, vast amounts of gravitational energy were converted into thermal energy and became trapped far below Jupiter's
clouds. For the past 44 billion years, Jupiter has been slowly cooling off
as this trapped energy escapes in the form of infrared radiation.
Saturn is both smaller and less massive than Jupiter. One would thus
expect Saturn to cool more rapidly than Jupiter and hence to emit less
energy today. But, in fact, Saturn radiates about 2\ times as much energy as it receives from the Sun, whereas Jupiter emits onlv about U
times what it absorbs from sunlight. What might explain why Saturn
much more heat than Jupiter?

For some time before the space probes to the Jovian planets, astronomers had suspected that both Jupiter and Saturn have compositions similar to that of the original solar nebula and the Sun's atmosphere today.
Each of these giant planets is massive enough and cool enough to have
retained all of the gases that originally accreted from the solar nebula.
The Voyager flybys did confirm that Jupiter's atmosphere has a solar
abundance of elements; by weight, 82 percent hydrogen, 17 percent helium, and 1 percent all other elements. Surprisingly, however, the Voyager spacecraft reported that Saturn's atmosphere has less helium than
expected. The chemical composition of Saturn's atmosphere, by weight,
is 88 percent hydrogen, 1 1 percent helium, and 1 percent all other ele-
emits so
ments.
A brilliant hypothesis links Saturn's apparent deficiency of helium to

the excess heat radiated by the planet. According to this theory, Saturn
did indeed cool more rapidly than Jupiter. This cooling triggered a
process analogous to the development of a rainstorm here on Earth.
The Jovian planets
147
When the air is cool enough, humidity in the Earth's atmosphere condenses into raindrops that fall to the ground. On Saturn, according to
this explanation, helium droplets rain downward from the planet's atmosphere toward its core. Helium thus appears to be deficient in Satupper atmosphere merely because it has fallen farther down into
urn's
the planet. Furthermore, as the helium droplets descend through the

molecular hydrogen, the two gases rub against each other. The resulting
friction produces heat that eventually escapes from Saturn's surface.
The precipitation of helium from Saturn's clouds is calculated to have
begun 2 billion years ago. The resulting release of energy adequately

accounts for the extra heat radiated by Saturn since that time. Similar
calculations for Jupiter indicate that only now it is reaching the stage
where a significant amount of helium precipitation can begin in its outer
layers. Saturn has therefore given us important clues about the probable
course of Jupiter's future evolution.
The
internal structures of
Jupiter and Saturn can be
Jupiter's rapid rotation profoundly affects the overall shape of the

planet. Even a casual glance through a small telescope shows that Jupiter
deduced from
slightly flattened, or oblate. The diameter across Jupiter's equator

(143,800 km) is 6.37 percent larger than the diameter from pole to pole
(135,200 km). Thus Jupiter is said to have an oblateness of 6.37 percent,
or 0.0637. Saturn is even more oblate than Jupiter (examine Figure
their slightly
is
flattened shapes
diameter is about 10 percent larger than its

polar diameter, so its oblateness is about 0.10.
If Jupiter and Saturn were not rotating, they would be perfect
spheres. A massive, nonrotating object naturally setdes into a spherical
shape in which every atom on its surface experiences the same intensity
of gravity aimed directly at the object's center. However. Jupiter and Saturn are rotating rather rapidly. Saturn's equatorial rotation period (10
hours 14 minutes) is only slightly longer than Jupiter's (9 hours 50 minutes). Thus every part of each planet also experiences an outwarddirected centrifugal force that is proportional to the distance from the
axis of rotation. Equatorial regions are farther from the planet's axis of
8-26). Saturn's equatorial
rotation than are the polar areas, so the equatorial diameter is slightly
larger than the polar diameter. This centrifugal stretching of the equato-
Figure 8-26
Saturn from Voyager 2
Saturn u the most ohtate planet
in the Milar
system. Saturn's equatorial diameter
km
larger than
its
from
the
this
picture
34 million
planet. (XASA)
spacecraft was
diameter from pole
Voyager 2 sent back
when
kilometers
I2.(M/(l
to
pole
the
awat
Tht JaiWH fiaruti
lial
dimensions gives Jupiter and Saturn dior
cfaaiacxerisiic ablate
shapes-
Al even point throughout each planet, the inward foice of giâvily is

exacth balanced bv the out^-ard pressure of the compressed material
plus the out\.5rd coiimUigal effects of the plaiiet''s rotation. This balance
is called hydnastMic eqpuUbrinaB. The shape and densin of a planet adjust iheniseKx"^ to ensure titat dtis balance is maintained.
nê sliupe of a planet is an e^KceUent indicanor of its iniiemal scructuix'. Tx^-o planets with the same mass, average density, and rotatioa will
Itaw sligiîlN ditTeient oblamvess if one planet has a cxxnpaa cme and
tlte loiter docN jkm. .Ml other ihii^ being equal, a planet iitli a dense
core will be K>te oblate than will a planet wiibout a oeotral core.
Detaikd cakîlatkHts stnx^^ su^^jest ihm 4 percent of Jupiter^s mass
is oiMKTcntrated in a dense, rock^- we, Jupiter's oblaieness is in faa consistetM with a Tiivi.> oore nearb 13 tiimes as massh'e as the entire Eanli.
S.i<ne oS" this cwwe was probablv the ordinal "seed" around -niiidi proioJupiter accnewd.
Jupiter s r<xi^> ooie
is probabih not imaciii hagger ihan tbe Eanii. en

15 hiTjes nxwe masshe dun our planeL Tbe tremexKlaias
craslHtK; wvijrhi <vt" the remiainjng 505 Eanii masses of Jupmer's b^L
compnesses ihc- core dcmij to a qîere f*U.OOG km in diaroeEer ('Eanib's
^-uin\j--u- ;*
vn- -^
Tbe pressure ai lupiter's cenier is sitma. SG nnlsquaezcs rhe rock-^ raaieria] of Jupiter's care to
though
it is
'
'
-Tn
'<
T1e
>-;..
the
yfenet * center is
J^Mttr's ckwid-
eaiijensoiiiate aa, itee
Ir. ^-.-r.iTasi-
efnw
A hvdrogen
iên.
Deep
;';,-:>n.
i;
ill
.J.,
from iSkd"
waader
caiier -vwards. ifeefci^^r
Hse a
be!ha^<e5
,;sesi rhai
of
lOors. are iinee so
'
at-om CKmsisB
inside jTqMEer, Ikwp-
irt srripped
i iniiiiiie .at
meistl. It
xnoleajlaT in'drogen
is
pressure esceeds 3
ociTTTs ai a depii) of appraamaieh
..N. Thus ihe iniemal strucmre of Jxqij.i:j,'>r!> (-see Figure S-27): the rode* rare is
JTOfjen v-hen rhe
*rt.H,\ ^'On
"liri
lavfT of liquid meiallic hvdro^fen.
outer lT.OO(*-fcni-ihick Iha'ct of ordicolorful clcoid panems that we see

ê cunermost 1 00 tm of this omer
^'C
iiT^
T.Tvii-^
oblaieness.
!sc'
niier:
1
N.i
Sv?'*/"^
'
,.
-,
;>
kft^ ITi-ss;
Tockn- cxire
nzm
and thus the pressures inside it are

Sarums Tock\ cxire iherefnre fc
s .Mso. Sarums weaker grawipi k
iiouid TTieialiu: -state over as lar^ a
-ii-^sses
of rhe iaver* drffer in
>-i!7).
,
iiiii
-solid.
-I'^T
km
proijîie
3uia TneiaiUt m'drflen. -surrounded in
tK^nnJ^ li.WMi
and
caicuiaied thai die in-
.i;
Saiums
liquid Tnetalhc
Tor-k'^
cart
hvdragen
riie /oi'iiin jAiitirls
Figure 8-28
Jupiter's magnetosphere
Like
other planetary mag>ielospheres, Jupiter\ has
ii
and a
mai^-
tietopause. In Jupiter's case, gas pressure
from
bow shock wave, a
magtietosheath.
149
a hot plasma keeps the magnetosphere inflated,

thereby holding off the solar wind. Particles
trapped inside the magnetosphere are spewed

out into a vast current sheet by the planet's
rapid rotation. Jupiter's

clined to
its
CLxis
magnetic axv,
try
of rotation
is
in-
about 11.
from Figure 8-27
that a large percentage of Jupiter's enorconductive liquid metal. Because of Jupiter's

rapid rotation, electric currents in this thick layer of liquid metallic hydrogen generate a powerful magnetic field, in much the same way that
liquid portions of the Earth's core produce the Earth's magnetic field.
The intrinsic strength of Jupiter's magnetic field is 19,000 times greater
than the Earth's. The Jovian magnetic field is so much stronger than
ours because Jupiter's liquid metallic region is so much larger than the
Earth's and because Jupiter rotates so much faster than the Earth does.
Jupiter's powerful magnetic field surrounds the planet with an enormous magnetosphere, large enough to envelop the orbits of many of its
moons. From the Earth, our evidence for this magnetosphere is limited
to a faint hiss of radio static. However, lour spacecraft that journeyed to
Jupiter in the 1970s revealed the awesome dimensions of the Jovian
magnetosphere. The volume surrounded by the shock wave is nearly 30
million kilometers across. In other words, if you could see Jupiter's magnetosphere from the Earth, it would cover an area in the sky sixteen
times larger than the full Moon.
The inner regions of our magnetosphere are dominated by two huge
Van Allen belts (recall Figure 7-34) filled with charged particles. The
same sort of belts would probably exist around Jupiter if it were rotating
slowly. But because Jupiter rotates rapidly, centrifugal forces spew out
the particles into a huge electrically charged current sheet (see Figure
8-28). This current sheet lies in the plane of Jupiter's magnetic equator.
Jupiter's magnetic axis is inclined slightly from the planet's axis of rotation ancf the orientation of its magnetic field is the reverse of Earth's (a
compass would point toward the south pole on Jupiter). Spacecraft
traveling to Jupiter risk severe damage to sensiti\e electronic components
when passing through the current sheet.
Saturn's mantle of liquid metallic hydrogen, like Jupiter's, also produces a planetwide magnetic field. However, Saturn's magnetic field is
somewhat weaker than Ju|)iter's becau.se of Saturn's slightly slower rotation and much smaller volume of li(|uid metallic hydrogen. Data from
spacecraft indicate that Saturn's magnetosphere resembles that of Jiqiiter
but that it contains radiation belts similar to those of Earth instead of a
huge current sheet like that in Jupiter's magnetosphere.
It is
clear
mous bulk
is
electrically
The jiwuni planets
Uranus and Neptune appear to be twins. They have

size, mass, density, and chemical composition. Until only
a few years ago, however, astronomers regarded the Earth and Venus as
near-twins. But when spacecraft probed the Venusian environment, we
were startled to discover how different Venus and Earth are, despite
their superficial similarities. Thus we had best curb our desire to speculate about similarities between Uranus and Neptune until Voyager 2 files
past Neptune in 1989.
Through Earth-based telescopes, Uranus and Neptune are dim and
uninspiring sights. Each appears as a small, hazy, featureless disk with a
faint greenish-blue tinge (see Figures 8-29 and 8-30). Earth-based obser-
Earth-based observations
provide basic information
At
first
glance,
nearly the same
about Uranus and Neptune
vations have provided only the most basic information about these two
worlds simply because they are so far away.
Figure 8-29
kilometers
jh
Uranus
[left]
from
the
Nearly 3
billion
Sun. Uranus receives only
we experience
dim and
Uranus appears
we can
the intensity of sunlight that
here on Earth.
frigid world. At
Uranus
its
is
therefore a
brightest,
as bright as the faintest stars
our unaided eyes from Earth. Even
see with
though
diameter
its
is
about four times Earth's.
Uranus always shows us a disk less than

4 arc sec across. (New Mexico State
University Observatory)
Figure 8-30 [right]
from
kilometers
the
Neptune
At 4i
billion
Sun, Neptune's surface
receives only gio the interuity of sunlight
receive
on Earth. Neptune
is
therefore
than Uranus. Although about the same
we
dimmer
size as
Uranus, Neptune looks smaller through Earthbased telescopes because
maximum
tune's disk
same
possible
is
size as
of 1 km.
it
is
farther away. The
angular diameter of Nepsec. That is roughly the
2.2 arc
a dime seen from a distance
(New Mexico
State
The outer two Jovian planets are significantly different from the
inner two. The large bulks of Jupiter and Saturn are composed primarily of hydrogen and helium solar abundances. Uranus and Neptune,
however, are distinctly smaller and less massive. If Uranus and Neptune
also had solar abundances of the elements, their smaller masses would
produce less compression and therefore lower average densities than
those of Jupiter and Saturn. In fact, however, Uranus and Neptune have
average densities comparable to or greater than those of Jupiter or Saturn. We must conclude therefore that Uranus and Neptune contain
greater proportions of the heavier elements such as oxygen, nitrogen,
carbon, silicon, and iron in addition to abundant hydrogen and helium.
From the known physical properties of Uranus and Neptune, astronomers have calculated that their interiors have three-layered structures
(see Figure 8-31). Each planet probably has a rocky core composed primarily of iron and silicon surrounded by a liquid mantle of water (HoO),
ammonia (NH3), and methane (CH4). Thus the mantles contain abundant amounts of oxygen, nitrogen, and carbon in combination with hydrogen. The outer layers of the two planets are predominantly hydrogen
and helium in gaseous state at low density.
The Juviaii
Figure 8-31
Neptune
The
limits
interiors of
and
Xffjliiiif
similar interior siruclures.

rock\ cure surrounded
water, methane,
is
Uranus and
have
vci-y
Each jAanet has a

a liquid mantle /
and ammonia,
enveloped in a
fildiieh
loti'-densit\
u'hich in turn
gaseous la\rr
composed largely of hydrogen and helium.
Neptu
Although similar
slightly
more
to
Uranus
radiation than
it
in
its
Neptune emits
from the Sun, whereas Uranus
internal structure,
receives
does not. Although farther from the Sun, Neptune's cloudtops register
the same temperature as Uranus's (57 K = 216C = 357F).
We saw that both Jupiter and Saturn have internal energy sources.
Jupiter is still releasing energy trapped inside it during its formation and
Saturn's excess heat probably comes from the precipitation of helium.
Neither of these mechanisms would work on Uranus or on Neptune because these outer planets are too small and contain too little hydrogen
and helium. The source of Neptiuie's internal heat therefore remains a
mystery.
Voyager 2 revealed many

details about Uranus, its
rings and satellites
After nearly Si years of coasting through interplanetary space. \'o\ager

Uranus in January 1986. As pictures and data poured in,
scientists learned more about this remote world than all the knowledge
that had accimtulated since Uranus was discovered two centuries ago.
It was already known from Earth-based observations that Uranus's
axis of rotation lies very nearly in the plane of its orbit (see Figure 8-32).
Cônsequently. as Uranus moves along its 84-vear orbit, the planet's north
and south poles alternately point toward or away from the Sun, producing exaggerated seasons. During the sununertime, near Uranus's north
pole the Sun remains high above the horizon for many years while
southern latitudes are subjected to a continuous, frigid winter night.
l''orl\-two years later, the situation is reversed.
Uranus's south pole was aimed almost directly at the Sun as N'ovager
2 approached the planet. I'hus Figme 8-33 shows its illuminated southern hemisphere. No clouds or any other atmospheric features are seen,
but computer processing did reveal a smoglike haze over the pole.
2 arrived at
The Jovian planets
Figure 8-32
Uranus
Exaggerated seasons on
Uranui's axu of rotation o
steeply that
it
planet's orbit.
tilted io
plane of the
Seasonal changes on Uranus are
lies
nearly
the
thus severely exaggerated. For example, during
midsummer
at Uranus's south pole, the
Sun
appears nearly overhead for many Earth years,

while the planet's northern regions are subjected to a long, continuous winter night.
Figure 8-33
Uranus from Voyager 2
distinctive cloud patterns are seen in
No
any of
Voyager views of Uranus. The blue-green

appearance of Uranus comes from methane in
the
the planet's atmosphere that absorbs red xuave-
from the incoming sunlight. The spacewas 18 million kilometers from the planet
lengths
craft
when
this
picture
was taken. (NASA)
Voyager's instruments discovered that Uranus has a magnetic field

roughly half as strong as Earth's. Furthermore, Uranus's magnetic axis
(that is, the line joining the north and south magnetic poles) is tilted
away from Uranus's axis of rotation by 55. This inclination is surprising
because most planets have their magnetic and rotational axes nearly
aligned. For example. Earth and Jupiter each have their magnetic axes
The Jovian planets
Figure 8-34
The rings of Uranus
[left]
Voyager 2 was in the shadow of Uratuis when

it took this H-minute exposure of Vranus's
The
and
rings.
urn's,
very thin rings

seen.
The
much darker than
rings are
this
Sat-
long exposure revealed many
and
dust lanes not previousl-i
short streaks are star images, blurred
because of the spacecraft's motion during the

exposure.
(NASA)
Figure 8-33 [right]
Miranda
Curious
banded features cover much of Miranda,
the
smallest of Uranits's five major satellites.
High-resolution photograplis taken by Voyager

valleys and ndges parallel tn these
Miranda is 4H0 km m diameter. A feui
impact craters 10 lo 30 km in diameter can be
show many
bands.
seen.
(NASA)
153
inclined by only 11 from their rotational axes. Uranus is thus unicjue

because of the unusual orientations of both its rotational and magnetic
axes.
For many years, Uranus's rotation period had been a topic of controamong astronomers whose Earth-based observations gave conflicting results. Voyager 2 was able to detect regular changes in radio emission from Uranus's magnetosphere that repeated every 17.2 hours.
These changes are caused bv the motion of Uranus's oblique magnetic
field as it is carried around by the planet's rotation. The magnetic field
is presumablv anchored deep inside the planet, so 17.2 hours must be
the rotation period of Uranus's core.
Revolving around Uranus in the plane of the planet's equator are
numerous satellites and a system of thin, dark rings. Nine of these rings,
ranging in width from 10 to 100 km, were discovered in 1977 when
Uranus passed in front of a star. The star's light was momentarilv
blocked by each of the rings, which proved their existence to astronomers. A H-minute exposure taken while Voyager was in Uranus's
versy
shadow revealed numerous additional verv thin rings (see Figure 8-34).
nine main rings are apparently composed of boulder-sized chunks
of rock and ice, whereas the fainter rings discovered bv \'oyager are primarily composed of dust particles.
Five of Uranus's satellites, ranging in diameter from 480 to nearly
1600 km, were known prior to the X'oyager mission. Voyager's cameras
discovered ten additional satellites, each less than 50 km across. Several
of these tiny moons are shepherd satellites whose gravity confines parti-
The
cles to the thin rings that circle
Of
Uranus.
Miranda is the most fascinating because it is covered with unusual wrinkled and banded surface features (see Figure
8-35). Miranda is the smallest of Uranus's five main satellites and the
impact of an asteroid could have temporarily broken it into several
pieces. Perhaps Miranda was shattered by such a collision and the blocklike features on its surface show how various chunks of Miranda came
all
these moons,
back together.
The jnvian planets
Summary
Jupiter
by far the largest and most massive planet in the solar system.
is
Jupiter and Saturn are primarily composed of hydrogen and helium. Both
planets have an overall chemical composition very similar to that of the
Sun.
Because of their rapid rotation, Jupiter and Saturn are noticeably oblate,
which provides important clues about their internal structure.
Jupiter and Saturn both probably have rocky cores surrounded by a thick
layer of liquid metallic hydrogen and an outer layer of ordinary hydrogen
gas.
The
visible features
of Jupiter
(belts,
zones, the Great
Red
Spot, ovals,
and colored clouds) exist in the outermost 100 km of its atmosphere.

urn has similar features, but they appear much fainter.
Sat-
There are three cloud lavers in the upper atmospheres of both Jupiter
and Saturn. Saturn's are spread out over a greater altitude range than
those of Jupiter, so the colors of the Saturnian atmosphere are somewhat
obscured.
The
colored ovals visible in the Jovian atmosphere represent gigantic

storms, some of which (such as the Great Red Spot) are stable and persist
for years. The ovals are cvclonic or anticvclonic storms at the boundaries
between wind streams moving in opposite directions around the planet.
Jupiter and Saturn both emit more heat than thev receive from the Sun.
Presumably Jupiter is still coohng. On Saturn, the precipitation of helium
downward into the planet is probably the cause of its excess heat.
Jupiter has a strong magnetic field created by currents in the metallichydrogen layer. Its huge magnetosphere contains a vast cmrent sheet of
electrically charged particles. Saturn's magnetosphere is similar to Jupiter's
but has Earth-like radiation belts instead of a current sheet.
Satmn
broad rings lying in the plane of the

ring lies nearest to Saturn. Just outside it is
the much brighter B ring, then the dark gap called the Cassini division,
and beyond it is the moderately bright A ring. Other, fainter, rings were
observed by the Voyager probes.
is
circled by a system of thin,
planet's equator.
The
faint
Each of Saturn's major rings is composed of a great many narrow

numerous fragments of ice and ice-coated rock.
ringlets
consisting of
Some of
the ring boundaries are produced by shepherd satellites, whose
gravitational pull restricts the orbits of the ring fragments.
Uranus and Neptune are quite similar to each other in appearance, mass,
and chemical composition. They both have three-layered internal
structures: a rocky core surrounded by a liquid mantle of water, ammonia, and methane, and an outer gaseous envelope composed predominantly of hydrogen and helium.
size,
Uranus
orbit,
is unique in that its axis of rotation lies nearly in the plane of

producing greatly exaggerated seasonal changes on the planet.
Uranus has a system of thin, dark rings and five

moderate-sized moons of Saturn.
Review questions
satellites similar to
its
the
1 Describe the appearance of Jupiter's atmosphere. What kinds of features

would you expect to see through a telescope?
2 Describe the structure of Saturn's rings.
What
are they
made
of?
155
Thf Jovian planets
3 If Jupiter does not have any observable solid surface and its atmosphere
how are astronomers able to determine the planet's
rotates at different rates,
rotation rate?
4 Why do features in Saturn's atmosphere appear to be much fainter and

"washed out" compared to features in Jupiter's atmosphere?
5 Why do astronomers believe that Jupiter does not have a large iron-rich
core even though the planet possesses a strong magnetic field?
6 Compare and contrast Jupiter's magnetosphere with the magnetosphere

of the Earth.
how shepherd satellites operate.

term for these objects? Explain.
7 Explain
priate
Is
"shepherd
8 Could astronomers of antiquity see Uranus?

was not recognized as a planet?
If so,
9 Compare the ring systems of Saturn and Uranus.
satellite"
an appro-
why do you suppose
Why
it
were Uranus's
rings unnoticed until the 1970s?
10 Compare and contrast the internal structures of Jupiter and Saturn with
the internal structures of Uranus and Neptune. Can you propose an explanation to account for the differences between the inner and outer Jovian
planets?
Advanced questions
why Saturn
11 Explain
tates
more
is
more
oblate than Jupiter even
though Saturn
ro-
slowly than Jupiter.
12 What sort of experiment would you design in order to establish whether

Jupiter has a rocky core?
13 Is it reasonable to suppose that Neptune is surrounded bv rings? How
might you detect such a ring system using an Earth-based telescope?
14 Describe some of the semipermanent features
What
factors infiuence their longevitv?
lived features with
some
in Jupiter's
atmosphere.
Compare and contrast these longphenomena seen in Jupiter's clouds.
of the transient
15 Suppose that you were designing a mission to Jupiter that involved an

would spend many days (months?) living through
What observations, measurements, and analyses should
this aircraft make? What dangers might it encounter and what design pro-
airplane-like vehicle that
the Jovian clouds.
blems would vou have
to
overcome?
16 N.'XSA and the Jet Propulsion Laboratory have tentative plans to place
spacecraft in orbit about Uranus and Neptune early in the twenty-first century. What kinds of data should be collected and what questions would you
like to sec answered bv these missions?
For further reading
and Neptune." Astronomy, Feb. 1977, p. 6.

"Discovering the Rings of Uranus." Sky is' Telescope. June
1977, p. 412.
Gore, R. "V'ovager X'icws Jupiter." Xalioual Geogiaphic Jan. 1980.
Gore, R. "Saturn; Riddle of the Rings." Nalwnnl Geographic, July 1981.
IngersoU, A. "Jupiter and Saturn." Scietilific Atnencati. Dec. 1981.
Belton, M. "Uranus
Elliot,
[.,
et al.
Johnson,
T.,
and Veates, C. "Return

Aug. 1983, p. 99.
Telescope,
to Jupiter: Project Galileo." Sky C?
The Juvmn planeb
Morrison, D. Voyages
to
Saturn.
NASA
SP-451, 1982.
good, lavishly
illus-
trated guide to the Voyager discoveries.
Morrison, D., and Samz,

cellent
J.
Voyage
to
Jupiter.
NASA
SP-439, 1980.
An
ex-
book on the Voyager mission as reported by one of the partici-
pants.
Pollack, J.,
and Cuzzi,
J.
"Rings
in the Solar
System."
Scientific
American,
Nov. 1981.
Washburn, M. Distant Encounters: The Exploration

Harcourt Brace Jovanovich, 1983.
of Jupiter
and Saturn.
The smaller
terrestrial planets
our moon, und
The smaller
worlds
terrestrial
Merciin.
the liiigrsi saleltiles ofjitpilei
and Saturn are shown here to the same scale.

Each of these worlds has its own unique characterutics and all are large enough to be classified as planets.
Only Tilan
mosphere. Not shown

satellite. Triton,
as Mercury.
is
which
possesses
an
at-
Xeptune's largest
is
roughly the same
sizi-
(NASA)
hi
chapter we examine Mercury
thi.s
enough
alotio^
to (jualify as terrestrial planets.
lunarlike suiface but
an
with seven giant
We find
thai
satellites
Mercuty has a
large
cralered,
Earth-like, iron-rich core with a magnetic field.
We
Moon and examine
lunar rocks containing

important clues about the Moon's histon. Orbiting Jupiter are four more
then follow the astronauts to the
its sulfur volcanoes, Europa covered b\ a

and Ganymede and Callisto each surrounded b\ a
thick mantle of water and ice. At Saturn we fnd Titan, the onl\ satellite u'ilh
a .sub.'\tanlial atmosphere, thicker ex'en than Earth's. Finally, we discuss Triton, the largest satellite of Neptune, and look foneard to the anival of the
unicpie terrestrial worlds: lo u'ith
thin la\er of shifting ice,
Voyager spacecraft that will photograph
The
idea thai there are nine planets
its
is
surface in 19S9.
obsolete. In reeent years, space-
have revealed additional worlds each of which is a terrestrial planet

in its own right. The seven giant satellites (recall Table (i-li) of our solar
system have their own unique geology and are roughly the same size as
Mercury. Along with Mercury, these seven worlds give us a new perspective on the variety of which nature is capable.
craft
Mercury has
a Moon-like
surface and an
Earth-like interior
we knew verv little about Merciny, the small planet that

the wariu inner regions of the solar nebula. Iiilormation
about Mermrv was difficult to obtain tor two simple leasons: it is very
small, and it is verv near the Siui. Indeed, Merciuy is so close to the Sun
L'niil
1971.
formed
that
in
most people (including many astronomers) have never seen
it.
riir swalh'i
The
Mercury occur when the planet is as far

can be, at greatest eastern or western elongation (recall Figure 3-4). For a few days near the time of greatest eastern elongation. Mercury appears as an "evening star," hovering low over the western horizon for a short time after sunset. Near the time of greatest
western elongation, it can be glimpsed as a "morning star," heralding the
rising Sun in the brightening eastern sky.
Mercury travels around the Sun faster than any other object in the
solar system, taking only 88 days to complete a full orbit. Thus Mercmy
passes through inferior conjunction at least three times a \ear, and vou
might expect occasionally to see Mercury silhouetted against the Sim in
what is called a solar transit. Transits of Mercury across the Sun are not
very common, though, because Mercury's orbit is tilted 7 to the plane of
the Earth's orbit. As a result. Mercury usually lies well above or below
the Sun at the moment of inferior conjunction. Only fourteen solar transits of Mercury are scheduled for the entire twentieth century (see Figure 9-1).
Naked-eye observations of Mercury are best made at dusk or dawn,
but the best telescopic views are obtained at midday when it is high
above the degrading atmospheric effects near the horizon. The photographs in Figure 9-2, among the finest Earth-based views of Mercmy
ever recorded, were taken at midday. Because of its small size and nearness to the Sun, you cannot see much surface detail on Mercury through
an Earth-based telescope. At best only a few faint, hazy markings can be
best opportunities to see
from the Sun
Figure 9-1
transit of
a dozen Itamils oj
tury. This
Ruiiirhh
occur each cen-
photograph shows the
houetted against the
November
Mercury
Mercmy
lin\ planet sil-
Sun during a
transit
on
14. 1907. (Yerkes Ohsen'aton)
tnicshiut imrlih
as
it
identified.
Our first detailed knowledge about Mercury's surface was acquired

when Mariner 10 coasted to within 756 km (470 miles) of the
in
1974,
planet's surface.
As the spacecraft closed
in
on Mercury,
scientists
were
surprised by the Moon-like pictures appearing on their television monitors. It became obvious that Mercury is a barren, desolate, heavily cratered world. Figure 9-3 shows a typical closeup view of Mercury and a
picture of our
Moon
Although our
for comparison.
impression is of a kmar-like landscape, closer scruof Mercury's surface reveals some significant nonkmar characteristics. Lunar craters are densely packed, one overlapping the next. In
sharp contrast. Mercury's surface has extensi\e intercrater plains (see
Figure 9-4).
tiny
Figure 9-2
Earth-based views of Mercury
These two views are

graplis of
among
the finest photo-
Mercury ever produced with an
Earth-based telescope. Hazy markings are

faintly visible on the tiny planet.
Stale University Observaton)
(New Mexico
first
'illfi
Mercury and our Moon
Figure 9-3
curs, like
our Moon,
hti.s
Miy
a heavily cratered
and no almosphere. Mercury's diameter

the Moon's is 3476 km. (Both
worlds are shown here to the same scale.) For
suiface
is
4878 km and
comparison, the distance from
Angeles
is
3944 km (2451
Sew
netic field; the
Moon
does not.
York
Los
to
Mercury
miles).
has a substantial iron rich core
and a magDmlime tem-
peratures at the equator on .Mercury reach
430X
(8(>()F). hot
enough
to melt
lead or
r.V/i.V.4.- l.ick
Ohseivalon)
Figure 9-4
Mercurian craters and
intercrater plains
I'lin
t'lew of
tin.
.Menim's
nor/hrni hcmisphnc was taken hf .Manner ID

at
a range of
55.fJ()0
the planet's surface.
km (34.000
Numerous
temive intercrater plains appear in

graph, which covers an area
miles) wide.
(NASA)
miles)
craters
from
and
this
ex-
photo-
480 km (300
liireslrial
iciirlil.\
The smaller (enestnal
u'orlih
Astronomers believe that most craters on both Mercury and the Moon
were produced during the first 700 million years after the planets
formed. The strongest evidence comes from the direct analysis and dating of Moon rocks brought back by the Apollo astronauts. Debris remaining after the planets had formed rained down on these young
worlds, gouging out most of the craters we see today.
Astronomers agree that the Moon and the terrestrial planets must
have been completely molten spheres of liquid rock at first. After a few
hundred million years, their surfaces solidified as the rock cooled. Nevertheless, large meteoroids could still easily puncture the thin cooling
crusts, allowing molten lava to well up from their interiors. Older craters
were obliterated as seas of molten rock fiooded portions of the planets'
surfaces. There is clear evidence still visible of extensive lava fiooding on
the
Moon.
The
planets did not cool
planet can
more
down
easily radiate
its
at
the
same
rate,
however.
internal heat into space
small
and cool off
more rapidly than a big planet. Because Mercury is larger than the
Moon, it took longer for a thick protective crust to form on Mercury
than on the Moon. Throughout Mercury's early history, molten rock
seeped up through cracks in its young, frail crust, and volcanism was
probably pervasive. The resulting lava fiows certainly inundated many
older craters, leaving behind the broad, smooth intercrater plains seen
by Mariner
10.
Mariner 10 measured the surface temperature on Mercury and

searched for traces of an atmosphere. Temperatures on Mercury vary
from 700 K (= 427C = 800F) at noon on the equator to 100 K
(= 173C = 280F) at midnight. This 600 K temperature range is
greater than that of any other planet or satellite in the solar system. Furthermore, the only atmosphere Mariner 10 detected was a thin scattering
of particles that may have been captured from the solar wind. Because
of Mercury's low surface gravity and high daytime surface temperature,
it is not able to retain a substantial atmosphere.
The average density of Mercury (5.4 g/cm') is nearly the same as that
of the Earth (5.5 g/cm'). However, typical rocks from the surfaces of the
terrestrial planets have a density of only 3 g/cm and are composed primarily of silicon and other lightweight elements. The higher average
density of the Earth and Mercury comes from abundant quantities of
iron that sank toward the planets' centers while they were still entirei\
molten. As we saw in Chapter 7, this process is called chemical differentiation. .A.s the result of the action of gravity, dense elements sink toward
a planet's center and force less-dense material toward the surface. Chemical differentiation must have occurred during and immediately after the
formation of the solar system, while the terrestrial planets were still entirely molten and internal mass motion could occur on a large scale.
Mercury's average density is slightly less than the Earth's, so you
might suspect that Mercury has proportionally a smaller iron core than
Earth's. This is not the case. The Earth is 18 times more massive than
Mercury. This larger mass pushing down on the Earth's interior compresses the Earth's core much more than Mercury's core is compressed.
In fact. Mercury is the most iron-rich planet in the solar system, with
iron accounting for 65 to 70 percent of its mass. The scale drawing in
Figure 9-5 shows the interior structures of Mercury and Earth.
Independent evidence of Mercury's large iron core came from Mariner lO's magnetometers, which discovered that Mercury has a magnetic
field. Iron is the only common element that could account for the magnetic field of a terrestrial planet.
'
Mantle
Iron corf
The internal structures of

Mercury and Earth Mercury is the most
Figure 9-5
irun-nch jikmel in the solar system.

core occupies
Its iron
an exceptionally targe fraction of
the planet's interior.
//()
Figure 9-6
Mercury's magnetosphere
Mercury's weak mu^netic
enough
leneilnat worlds
lu cun>i'
fie/il is jiisl
strong
out a cavily in the solar wind,
preventing the im/jinging particles from

ing the planet's surface directly.
strik-
As we saw
Chapter 7, the Earth's magnetic field occurs because of

Magnetism arises whenever electrically charged particles are in motion. The Earth's magnetic field, generated by its rotation,
is similar to the magnetism that surrounds a coil of wire in which
electricity is fiowing. Many geologists suspect that the Earth's magnetic field
originates with electric currents fiowing in the liquid poi lions of our
planet's iron core. These currents, carried around by the Earth's rotation, create the planetwide magnetic field.
Scientists were surprised to find that Mercury also has a magnetic
field. Mercury rotates much more slowly than the Earth (59 days versus
24 hours), so most scientists believed that this leisurelv rotation would be
too slow to induce a magnetic field by the dynamo effect. Nevertheless,
it is clear that Mercury does have a weak field (the
Earth's field is 100
the
dynamo
in
effect.
times stronger).
Mariner lO's magnetometers could not prove that Mercury's field is

produced by the dynamo effect. Mercury's magnetic field may be a "fossil field" frozen in a completely solid iron core. Thus Mercury's magnetism may be similar to that of an iron magnet you can buy in a store.
Charged particle detectors on Mariner 10 inapped the structure of
Mercury's magnetosphere with results shown in Eigure 9-6. As we saw in
Chapter 7 (recall Figure 7-34), when supersonic particles in the solar
wind first encounter a magnetic field, they slow abruptiv, producing a
bow-shaped shock wave at the boundary where this sudden decrease in
velocity occurs. Closer to the planet, there is another well-defined
boundarv, the magnetopause, where the outwaid magnetic pressure of
till- i)l,iiKi's field exactiv counterbalances the impinging gas
pressure of
the sol.u wind. Iktween the shock wave and the ni.igneiopause is the
turluileiil
region called the magnelosluaiii
in wiiiiii niosi
of the subsonic
particles
from the
solar
wind are deflected around the planet, just
as
water is deflected by the bow of a ship. The region inside the magnetopause is the true magnetic domain of the planet. Mercury's magnetic
field is not strong enough to capture particles permanently, so it has
nothing comparable to Earth's \'an Allen belts.
The Moon's early history

can be deduced from
the craters and plains visible
on its surface
own satellite consistently provides one of the most dramatic sights

nighttime sky. The Moon is so large and so near that some of its
surface features are readily visible to the naked eve. Even casual observation reveals that the Moon perpetuallv keeps the same side facing the
Earth. This is a stable situation resulting from the gra\itational interaction between Earth and Moon.
With a small telescope you can see several different types of major
lunar terrain (see Figure 9-7). Most prominent are the large, dark, flat
Earth's
in the
areas called maria (pronounced MAR-ee-uh).
The
singular form, mare

was introduced in the seventeenth century when observers using early telescopes thought they had
seen large bodies of water on the Moon. In fact, bodies of liquid water
could not possibly exist on our airless satellite. Because there is no atmospheric pressure, a lake or ocean would boil furiouslv and evaporate
rapidly into the \acuum of space. The maria were actuallv formed by
huge lava flows that inundated low-lying regions of the lunar surface 'ii
billion years ago. They have still, however, retained their fanciful names
such as Mare Tranquillitatis (Sea of Tranquillity), Mare Nubium (Sea of
Clouds), Mare Nectaris (Sea of Nectar), and Mare Serenitatis (Sea of Se-
(pronounced MAR-ee) meaning "sea"
in Latin,
renity).
The
maria is Mare Imbrium (Sea of Showers).

measures 1100 km (700 miles) in diameter. Although
the maria seem quite smooth in telescopic views from the Earth, closeup
largest of the fourteen
Roughly
Figure 9-7
The Moon
seven large
satellites in the
Our Muuii
Moon's diameter (3476 km

slightly less
to
San
one of
The
= 2160 miles)
New York
is
than the distance from
Francisco. This photograph
posite of first-quarter
so elongated
tures. (Lick
is
solar system.
and
Observatori)
is
a com-
last-quarter views.
shadows enhance
all surface fea-
circular,
it
The imtiUn
Figure 9-8
Details of
Tranquillilatis
terrestrial worlds
Mare
Cluseuj) views oj the luniir
surface reveal numerous tmt craters
and
cracks on the maria. This photograph was
taken from lunar orbit m 1969 by astrottauls

during a final photographic reconnaissance /
potential landing
sites.
(NASA)
photographs by the Apollo astronauts reveal small craters and occasional

cracks called rilles (see Figure 9-8).
Perhaps the most familiar and characteristic features on the Moon are
craters. With an Earth-based telescope, some 30.000 of them are visible, from
km to more than 100 km across. Following a tradition established in the se\enteenth century, the most prominent craters are named
.liter famous philosophers and scientists.
Craters smaller than about
km in diameter cannot be seen from
Earth, but photographs from lunar orbit reveal millions of craters that
escape the scrutiny of Earth-based observers. V'irtuallv all craters, both
large and small, are the result of bombardment by meteoritic material.
Many of the youngest craters are surrounded by light-colored streaks
called rays that were formed by material violently ejected during impact.
In addition, many large craters have a pronounced central peak formed
during a high-speed impact by a sizable metcoroid (see Figure 9-9).
Fhe flat, low-lying, dark maria cover only 15 percent of the lunar surface. The remaining 85 percent of the surface is light-colored, heavilv
its
Figure 9-9
Details of a lunar crater
I'hr
photograph, taken from lunar orhit by .\pullo
cratered terrain
11 astronauts in 1969. shows a typical view
This second kind of terrain
of the Moon's heavily cratered far
side.
The
large crater near the middle of the picture
approximately
80 km (50
.\'ote the crater's
surface
miles) in diameter.
One
central peak
and
the
numer-
ous tmy craters thai puckmark the lunar

surface.
(XA.SAj
means "land"
is
is
at
elevations generallv higher than those of the maria.

is called the terrae, or highlands. ( Terra
in Latin, so in this fanciful
covered by either "land" or
terminology, the entire lunar
"sea").
of the surprises from the early days of lunar exploration is that

there are no maria on the Moon's far side, which consists entirelv of
heavily cratered highlands. Detailed observations bv astronauts in lunar
demonstrated that the maria on the Moon's Earth-facing side are 2

below the average lunar elevation. In contrast, the cratered
terrae on the far side are tvpicallv at elevations up to. 5 km above the
orbit
to 5
km
average surface elevation. These elevation differences are explained by

the Moon's crust being thinner on the Earth-facing side than on the
Moon's far side, as shown in Figure 9-10.
Because of the extraordinary thickness of the Moon's rigid lithosphere, no tectonic plate
movement
is
possible. If the
Moon
has an iron-
The smaller
Figure 9-10
Moon
lerreslrial worlds
The internal structure of the
Like
lite
Earlh.
llir
and a
a crust, a mantle,
Moon probuhh
The lunar
core.
lius
crust
has an average thickness of about 60 km on

the Earth-facing side but about
far
side.
form a
The
crust
and
solid
lithosphere about
plastic (nonrigid) asthenosphere
tends all the

the
Moon
less
way
km
main features of
to those
in diameter.
the
Moon's
of the Earth's
portions are quite different.
here
is
graphs
thick.
the
Tlw
probabh
ex-
base of the mantle. If
has an iron-rich core,
than 700
ogous
to the
100 km on
upper mantle
800 km
it is
solid
and
Although the
interior are analinterior, the pro-
The information
based on analyses of data from seismoleft
on the
Moon
by astronauts.
Possible
iron-rich core^
it does not endow our satellite with a magnetic field. Thus the
does not have a magnelosphere, and particles of the solar wind
strike the lunar surface directly.
The Moon's crust is thin on the Earth-facing side, so large meteoroids
could easily puncture the cooling crust shortly after the Moon formed 4i
billion years ago. Lava welled up, flooding the low-lving areas and producing the maria. Large meteoroids must also have struck the Moon's
far side. However, the crust on the Moon's far side was thick enough to
prevent penetration to the molten interior and no maria were created.
Not much has happened since those ancient days; the entire lunar surface has remained almost unchanged for billions of years.
rich core,
Moon
Lunar rocks were formed

3 to 4; billion years ago
There were six successful manned lunar landings. The first two, Apollo
and Apollo 12, set down in maria. The remaining four (Apollos 14
through 17) were made in progressi\ely more challenging terrain culminating in rugged mountains just east of Mare Serenitatis (see Figure
1
9-11).
The major
were concern
wide variety of geo-
factors in choosing the six landing sites
for the astronauts' safety
and the desire
to explore a
logically interesting features.
The Apollo astronauts brought back 382 kg (843 pounds) of lunar

rocks that proved to be a very important source of information about
the early history of the Moon and Earth. All the lunar rock samples appear to have formed through the cooling of molten lava. The samples
are almost completely composed of the same minerals found in terrestrial volcanic rocks. In addition, the entire limar surface is covered with
a laver of fine powder and rock fragments produced bv A\ billion vears
of relentless meteoritic bombardment. This laver, which ranges in thickness from 1
to 20 m, is called the regolith because the term "soil" as
used on Earth normally suggests the presence of decaved biological matter.
The astronauts who visited the maria discovered that these dark regions of the Moon are covered with basaltic rock similar to the darkcolored rocks formed by la\a flows from volcanoes in Hawaii and Ice-
Till'
Figure 9-11
Moon
less
An
smaller lirreshial worlds
Apollo astronaut on the
The Moon
a desolate, barren,
l\
life-
world. This ly/jiail view of the lunar sur-
face shows an Apollo 17 astronaut near a

large rock. Since the Moon has no atmosphere,
lunar rocks have not been subjected to ivealh-
enng and
thus contain unaltered information
about the early history of the solar
.'system. From
the SL\ manned lunar landings between 1969
and 1972 astronauts brought back a total
of 382 kg (843 lb) of Moon rocks. (NASA)
The rock of these low-King lunar plains

Figure 9-12).
land.
is
called
mare
basalt (see
In contrast to the dark niaria, the lunar highlands are covered with a
lighi-colored rock called anorthosite (see Figure 9-13). On Earth, anor-
found onl\ in such very old mountain ranges as the AdiUnited States. .Anorthosite is rich in calcium and
aluminum in comparison to the mare hasalts. which ha\e more of the
heavier elements such as iron, magnesium, and titanium. .Anorthosite
thositic
rock
rondacks
is
in the eastern
therefore has a lower density than basalt. The anorthositic magma apparently floated to the lunar surface when the Moon was molten, solidif\ing as it cooled to form the lunar crust. The denser mare basalts
Figure 9-12
1
[left]
Mare
basalt
53 1-g (3i-pound) sj/rcimoi
was brought back
Irs
This particular sample
is
This
mine
iij
.\piillo I 5
formed
hiisalt
The
called a vesicular
cias,
basalt because of the tiny holes, or vesicles,

that coi'er 31) percent of the rock's surface.
Gas mint have been

in the lai'a
When
dis.solved
from which
under pressure
this rock solidified.
the lava reached the airless lunar sur-
face, bubbles
Some of
fanned as
the pre.',sure dropped.
the bubbles were frozen in place as the
rock cooled. (S'ASA)
Figure 9-13 (rightj
The
Anorthosite
colored lunar terrae are covered with

cient type of rock called anorthosite.
thositic rock
tlie
is
light
an an-
Anor-
believed to be the material of
original lunar crust. Thi', particular .am
pie. called the '(ienesis rock"
astronauts
who picked
it
up
Iry
the Apollo
I i
at the base oj tin
Apennine Mountains, has an age of apfiroximately
-f.l
billion years.
(NASA)
later
from
lava that
oo/ed out
ol tiie interior to
fill
the
mare
basins.
aslrmiauls.
.Apollo astronauts brought back many specimens ot impact brecwhich are various rock fragments that have been cemented ioi;etliei
Figure 9-14
lunar breccia
Mfleurilic
impacts can cement rock fragments together
form
breccias. This glass-coated
to
lunar rock was
mm
picked up near the rim of a crater by the
Apollo 15 astronauts. Natural glass forms
whenever molten rock
by meteoritic impact (see Figure 9-14). In addition to making breccias

and churning up the regohth, meteoritic impacts also melt rocks to produce glass. Manv lunar samples are coated with a thin layer of smooth,
dark glass created when the surface of the rock was suddenly melted,
then rapidly solidified. Small black glass beads are also common in the
lunar regolith. These glass spheres were presumably formed from droplets of molten rock hurled skyward by the impact of a meteoroid.
Many lunar samples bear the scars of meteoritic dust grains traveling
at thousands of kilometers per hour, which produce tiny zap craters on
moon rocks (see Figure 9-15). Because of these tiny, glass-lined craters.
Moon rocks often seem to sparkle when held in the sunshine.
Meteoritic bombardment is the only source of "weathering" for lunar
rocks. The rate of this weathering is actually quite slow. Geologists estimate that it takes tens of millions of years to wear away a layer of rock
only 1
thick. Features formed 3 billion years ago are well preserved
today, and the astronauts' footprints will remain sharply imprinted on
the lunar surface for millions of years to come.
Although lunar rocks bear a strong resemblance to terrestrial rocks,
there are some important differences. Every terrestrial rock contains
some water, but lunar rocks are totally dry. There is absolutely no evidence that water ever existed on the Moon. In the absence of both an
cools rapidly. Notice the
dark color of the shiny glass coating. (NASA)
atmosphere and water,

traces of
By
it
is
not surprising that the astronauts foimd no
life.
carefully measuring the abundances of trace amounts of radioacelements in lunar samples, geologists confirmed that anorthosite is
more ancient than the mare basalts. This result had been expected because the lunar highlands are densely cratered, whereas the basaltic surfaces of the mare show relatively few craters. Typical anorthositic specimens from the highlands are between 4.0 and 4.3 billion years old (one
rock brought back by Apollo 17 is nearly 4.6 billion years old). All these
ancient specimens represent material from the Moon's original crust. In
tive
Figure 9-15
faces of
zap crater
many Moo7i
The upper
sur-
rocks are covered with
nearly microscopic craters produced by the im-
pact of high-speed meteoritic dust grains. These

glass-lined craters are typically less than 1
in diameter.
(NASA)
mm
iilirr
Icimlrial
it'orlth
167
all the mare basalts are between ;V1 and 3.8 billion years old.
.Apparently, the mare basalts solidified from lavas that gushed up from
the Moon's mantle and flooded the mare basins between 3.1 and 3.8 bil-
contrast,
lion years ago, just about the time the oldest rocks in the Earth's present
siufaco la\ers were being forined.
Volatile elements such as potassiinii and sodium melt and boil at relatively low temperatures, whereas refractory elements like titanium, calcium, and aluminum melt and boil at much higher temperatures. Compared to terrestrial rocks, lunar rocks have slightly greater proportions
and slightly lower proportions of volatile elements. The implication is that the Moon formed from material somewhat hotter than the material out of which the Earth was created. Some
of the \olatile elements boiled away, leaving the \oung moon relatively
enriched in refractory elements.
of refractory elements
The Moon probably forined by accretion dining the final stages of

he creation of the solar system. Great numbers of rock fragments must
have swarmed about the protosun nearly 5 billion years ago. Energy
I
Figure 9-16
n
yoiiiii;
Eratosthenes
Eri:Ui\lhiius
is
km in dmmeter on the
Mare Imbrntm. Another
cniler 61
suuthern edge uf
young
crater,
in this
photograph taken b\ the Apollo
Copemicm.
tronauts in 1972.
is
near the horizon

1
7 as-
fXASA)
from the contracting protosun would ha\e baked the water and volatile
elements out of the smaller rocks, many of which would have soon been
captured into orbit about the protoearth. Then, just as planetesimals had
accreted to form the piotoearth in orbit about the Sun, the fragments in
orbit about the Earth accreted to form the Moon
or so says the most
widely accepted theory.
The Moon probably reached roughly its present mass by about 4.6 billion years ago. At first, the heat produced by rock fragments falling into
the protomoon and by the decay of radioactive isotopes kept the Moon
molten, .\fter a few hundred million years, however, the rain of rock
fragments tapered off As the Moon gradually cooled, low-density lava
floating at the
Moon's surface began
crust that exists today.
about 4
The heavy
to solidify into the anorthositic
barrage of large rock fragments ended

impacts producing the ancient
billion years ago, with the final
craters that cover the lunar highlands.
At the end of this crater-making era more than a dozen asteroid-sized

each at least 100 km across, rained down on the voung Moon,
objects,
blasting out the vast mare basins. Then, from 3.8 to 3.1 billion years
ago, great floods of molten rock gushed up out of the lunar interior, filling the impact basins and creating the maria we see today.
Very little has happened on the Moon since those ancient times. A
few fresh craters have been formed (see figure 9-16). but the astronauts
visited a
world that has remained largeK unchanged for over 3
billion
years.
The formation
of the Galilean
probably mimicked
the formation of the
satellites
solar
system
(.alileo (ialilei was the first person to see the four largest satellites of
Jupiter (recall Figure 3-10). He called them the "Medicean Stars" to attract the attention of this wealthy family of Florentine patrons of the arts
and sciences. Since 1610, these four giant moons have plaved an impor-
tant role in our understanding of the solar system. To Galileo, thev were
observational evidence supporting the heretical Copernican cosmology.
To the modern astronomer, the Voyager fhlns of 1979 rexealed four
extraordinary terrestrial worlds, called the Galilean satellites, different

from anything astronomers had ever seen or imagined. They are named
after the mythical loxeis and companions of Zeus: lo, Europa, Gany-
mede, and
When
look like
Gallisto.
viewed through an Earth-based telescope, the Galilean satellites

mere pinpoints of light. With patience, you can follow these
///-'/
Inirshial world'.
four worlds as they orbit Jupiter. Because their orbital periods are fairly
from 1.8 days for lo to 16.7 days for Callisto major changes in
the positions of the satellites are easily noticeable from one night to the
short
next.
The
its
brightness of each of the
orbit.
These
moons
moves along
and light
or hidden from view as the
varies slightly as
it
variations can logically be attributed to dark
surface areas that are alternately exposed to
Careful measurements show that the brightness of each

with a period equal to the satellite's orbital period. In
other words, each Galilean satellite rotates exactly once on its axis dining
satellite rotates.
satellite varies
each trip around its orbit. We can thus conclude that each Galilean satellite keeps the same hemisphere perpetually facing )upitei just as our
Moon keeps the same side facing the Eariii.
Accurate measurements of the diameters of the Galilean satellites
came from the Voyager flyljys through direct photography that revealed
measurable disks (see Figure 9-17). The two inner Galilean satellites, lo
and Europa, arc approximately the same size as our Moon. The two
outer satellites, Ganymede and Callisto, are comparable in size to Mercury.
Deflections in the trajectories of the Voyager spacecraft as they passed

the Galilean satellites provided important data from which their masses
were computed. Europa, the smallest Galilean

massive.
Ganymede, the
satellite
largest Galilean satellite
is
is
also the least
the most masssive.
Data about the Galilean satellites are listed in Table 9-1.

As soon as reliable mass and diameter measurements were available, it
became apparent that the average densities of the satellites are related to
their distances from Jupiter. The innermost satellite, lo, has the highest
average density (3.55 g/cm'), slightly denser than our Moon. The next
satellite, Europa, has an average density of 3.04 g/cm\ which is not quite
Figure 9-17
our
The Galilean
Moon and
salcllili's
atid Saturn's largest
moon. Titan. All
worlds are reproduced
numerous
smooth,
size as
satellites
with
Titan Jupiter's four large
Moon
arc shown here along with our
lo the
active volcanoes
icy surface.
same
six
lo has
.scale,
and Europa has
Both are roughly
the
same
our Moon. Ganymede and Callisto
are each covered with a 1 OOO-km-thick layr)
of
ice.
They are roughh
Mercury. The remainiii:^
the
same
size as
Inri^f \alillilc
of the solar system. Tnlon. m iml slimni.

Voyager 2 will altempi hi ohlmii jilmloiJiapIn
of Triton during a flyhy of Neptune in
(S. P.
Meszaros;
NASA)
I9,S'J.
Tlir snmtlir lini'slrial u'inlih
TABLE 91
The Galilean
satellites
Name
Distance
Orbital
from Jupiter
period
(km)
(days)
Average
Diameter
(km)
Mass
(Moon =
density
1)
(g/cm')
lo
412.600
1.77
3632
1.21
3.55
Europa
670,900
3.55
3126
0.66
3.04
Ganymede
1.070,000
7.16
5276
2.03
1.94
Callisio
1.880,000
16.69
4820
1.44
1.81
4878
4.49
5.42
3476
1.00
3.34
Mercury
Moon
dense as our Moon. Recalling that rocks in the Earth's crust typically
have densities around 3 g/cm\ it is reasonable to suppose that both lo
and Europa are made primarily of rocky material.
The outer two satellites also exhibit decreasing density with increasing
distance from Jupiter. Ganymede and Callisto each have an average density of less than 2 g/cm \ indicating that these two satellites are composed
of roughly equal amounts of rock and ice.
The arrangement of the Clalilcan satellites parallels the characteristics
of the planets grouped according to their distance from the Sun. For
example, moving outward from the Sun, average densitv declines from
more than 5 g/cm' for Mercury to less than 1 g/cm~' for Saturn (recall
Table 6-2). -Scientists therefore began to suspect that the same general
processes that had formed the solar system were at work during the formation of the Galilean satellites, though on a much smaller scale.
as
NASA scientists recently computed the conditions necessary for the

formation of the Galilean satellites, including in their simulation the fact
that Jupiter emits twice as much energy as it receives from the Sun.
They calculated that frozen water could be retained and incorporated
into satellites at the distances of (ianvmede and Gallisto but that only
rocky material would condense at the orbits of lo and Europa because of
Jupiter's warmth. Jupiter's gravity and heat thus produced two distinct
classes of Galilean saleiiitcs, just as warmth from ihe proiosuii caused a
dichotomy between the small, dense, rock) inner planets and the huge,
gaseous, low-density outer planets.
lo is covered with colorful

deposits of sulfur
compounds ejected from
numerous active volcanoes
lew hours after Vovager

passed near Jupitei lo loomed into
|jrobe began sending back a series of strange and unexpected |>ictines such as the one shown in Figine 9- IS. Baffled h\ what
Within
\
ii'w
and the
they were .seeing, scientists jokingly lompared lo to pizzas and rotten

oranges.
A major clue to these puzzling vistas was luicovered several days after
the Jupiter flyby when a navigation engineer noticed a large lunbrella-
shaped cloud protruding from lo in one photograpli an erupting volNo one had expected lo obtain photogra])l)s of erupting volcanoes
on lo. After all, a probe making a single trip past the Earth would be
cano!
very unlikely to catch a
largi'
voUano
in the
ad
of eru|)ling.
photographs revealed eight

hese volcanoes are named after gods aiui godgiant ongoing eiuptions.
desses iiaditionallv associated with fire in Greek, Norse, Hawaiian, and
Gareliil reexamination of the closeup
1
The smaller
lerreslrud worlds
other mythologies. Figure 9-19 shows two views of the symmetric plume
of
Prometheus.
As it orbits Jupiter, lo is repeatedly caught in a gravitational tug-ofwar between the huge planet on one side and the other Galilean satellites on the other. This gravitational battle distorts lo's orbit, varying its
distance from Jupiter. As the distance varies, tidal stresses on lo alternately squeeze and flex the satellite. This constant tidal flexing in tinn
causes frictional heating of lo's interior. Calculations show that the heat
pumped into lo this way is equivalent to 2000 tons of TNT exploding
everv second. Eventually this energy makes its way to lo's surface, producing the niunerous \olcanoes.
The plumes and fountains of material spewing from lo's volcanoes
rise to astonishing heights of 70 to 280 km above the surface. To reach
Figure 9-18
This closeuj) view
lo
oj lo
was
taken by Vutager I at a range of about
860,000 km. Notice
the extraordinary
of colors from while, yellow,
range
and orange
to
black. Scientists believe that these brilliant col-
became of surface deposits of sulfur

ejected from lo's numerous volcanoes. (NASA)
ors occur
these altitudes, the material must emerge from the volcanic vents with
speeds between 300 and 1000 m/sec, a speed much greater than is found
in the most violent terrestrial volcanoes. For example, Vesuvius, Krakatoa, and Mount St. Helens have eruption velocities of only about 100
m/sec. Scientists therefore began to suspect that lo's volcanoes operate in
a fundamentally different way from volcanoes on Earth. The evidence of
these differences came from Voyager's pictures and data.
No impact craters like those on our Moon were seen on lo. Material
from the volcanoes apparentiv obliterates impact craters soon after they
are created. This lack of craters indicates that lo's sinface is extremely
young perhaps less than 100 million years old.
The Voyager cameras revealed numerous black dots on lo, which are
apparently the volcanic vents from which the eruptions occur. These
black spots are typically ten to fifty km in diameter and form 5 percent
of lo's surface. Lava flows radiate from many of these black dots (see
Figure 9-20), some of which have volcanic plumes.
Evidence supporting the volcanic nature of the black spots came from
Voyager instruments that measured the intensity of infrared radiation
across lo's surface. Some of the black spots have temperatures as high as
20C, in sharp contrast to the surrounding surface temperature of only
-146C.
After the discovery of widespread volcanic activity on lo, scientists
soon concluded that sulfur ejected from the volcanoes is responsible for
lo's brilliant colors. Sulfur is nonnally bright yellow. When it is heated
and then suddenly cooled, however, it can assume a range of colors
from orange and red to black.
Figure 9-19
Prometheus on lo
These two
Voyager views, taken two himrs apart, show

of the plume
details
oj the
volcano called Pro-
metheus. The plume's characteristic umbrella
shape
is
space.
silhouetted against the blackness of
When viewed
groxind of
plume a
rises to
surface.
agairut the light back-
lo's surface, jets
of material give the
spiderlike appearance. This
an
altitude of
(NASA)
100 km above
plume
lo's
The smaller
Figure 9-20
volcanic center on lo
terrestrial
worlds
\'o
impact craters are seen in chiseup pictures such

as this one taken by Vosaêr I at a range
of
130.000 km. Lung, meandering lava flows

from many of the black dots thai ap-
radiate
parently are the

ity.
sites
of
inten.<ie
volcanic activ-
This photograph covers an area
800 km,
approximately twice the
California.
size
1000
by
of
(\ASA)
Voyager detected sulfur dioxide (SO..) in the plumes lioiu lo's

volcanoes. Sulfur dioxide is an acrid gas commonlv discharged from
fumaroles and volcanic vents here on Earth. When this gas is released
vacuum of space from eruptions on lo, it crystallizes into
into the cold
white snowflakes. It is likelv that the whitish deposits on lo (see Figure

9-18) are sulfur dioxide frost or snow.
The abundant
sulfur and sulfur dioxide on lo suggest an explanation

mechanism of its volcanoes. Indeed, the term volcano mav be the
wrong word altogether. We have seen that material is ejected from volfor the
much higher velocities than is observed in even the

most explosive volcanic eruptions on Earth. Another major difference is
that lo's vents are not located at the tops of volcanic mountains. Manv
volcanoes on Earth and Mars have an easily recognizable conical shape
with a caldera at the summit. Few of lo's calderas (the black spots) are
associated with anv major topographical relief.
canic vents on lo at
In all these respects, lo's volcanoes are more similar to terrestrial geysers than volcanoes. In a gevser. such as those in Yellowstone Park in
Wvoming, water seeps down to volcanicallv heated rocks, is suddenlv
changed
and erupts explosively through a vent. If the Old

were to erupt under the low gravity and vacuum that
would send a plume of water and ice to an altitude of 10
to steam,
Faithful geyser
surround lo, it
km.
Both sulfur and suilur dioxide are molten at depths of oulv a lew kilometers below lo's surface because of the heat generated bv tidal flexing. CJeologists have pointed out that sulfur dioxide could be the principal propulsive agent driving lo's eruptions. Just as the explosive
conversion of water into steam produces a gevser on Earth, the sudden
conversion r)f lic]uid sulfur dioxide into a high-pressure gas could produce an eruption on lo, Calculations indicate th.ii this explosive expansion of sulfur dioxide could result in eruption velocities up to 1000
m/sec.
The material erupting from lo's geyserlike volcanoes is composed primarily of sulfur and sulfur dioxide. Voyager's instruments failed to detect any other abundant gases such as the water vapor and carbon diox-
Tin- smiilln
lnn-.hia/ imilih
ide emitted from terrestrial volcanoes. lo has apparently been completely

outgassed by volcanic activity over hundreds of millions of years. lo has
not been able to retain its volatile gases and has almost no atmosphere
because its surface gravity is comparable to that of our Moon.
It is estimated that each volcano on lo ejects roughly 10,000 tons of
material per second. Although this material is a hot mixture of molten
sulfur and sulfur dioxide gas under high pressure as it gushes from a
volcanic vent, the gas liquid mixture rapidly cools and solidifies in the
cold, nearly perfect vacuum around lo. it then takes about half an hour
for the fine particles of sulfur dust and sulfur dioxide snow to fall back
down onto the surface.
Altogether, lo's volcanoes and vents eject an estimated 100 billion tons
of matter each year. This amount represents a sulfur-rich layer 10 m
thick over lo's entire surface each year. Thus the surface is constantly
changing, and it is probably safe to say that there are no long-lived or
even semipermanent features on
Europa
smooth
is
did not pass near Emopa, but Voyager 2 captured the excelshown in Figure 9-21. Europa is a very smooth world with no
mountains and very few craters, crisscrossed with a spectacular series of
streaks and cracks. Most of the cracks appear to be filled with darkcolored material, but some cracks have light-colored substance in them.
covered with a
Voyager
lent view
layer of ice,
crisscrossed with
many
lo.
cracks
We have seen that Europa's average density is about 10 percent less

than the average density of our Moon. Spectroscopic observations troin
Earth indicate that Europa has frozen water on its surface. These two
facts together suggest that Europa's surface may be covered with an ice
layer 100 km thick. That would be consistent with its remarkable
smoothness. An "ocean" of ice 100 km deep would certainly hide mountain ranges and other topographic features. But what causes the network
of cracks, and whv are impact craters so rare?
Tidal squeezing is responsible for the volcanism on lo, which gives lo
many of its extraordinary characteristics. Europa is caught in a similar
tidal tug-of-war. with Jupiter to one side and the two largest Galilean
Figure 9-21
is
Europa
Einnpa's
covered by numerous streaks
ice
and
surjau
cracklih
features that give the satellite a fractured ap
pearance. The streaks are typically
km
20
to
wide. This picture, taken by Voyager
a distance of
240.000 km. allows surface
features as small as 5
seen.
(NASA)
40
2 <il
km
across to be
Thi' imallrr
Irimlniil winlih
173
moons
periodically passing on the opposite side. Hovvc\ci.

farther from Jupiter than lo is, and the tidal effects
are considerably less than those on lo.
much
The
hAuopa is
on Europa
tidal flexing
of K.uropa is responsible for the network of cracks

Some of the darkest streaks in fact follow paths
along which the tidal stresses are calculated to be stiongest.
Ihe tidal
flexmg of Europa is far too weak to produce volcanoes',
but it is thought
to supplv enough energy to jostle and churn
the satellite's icv coating.'
This activity would explain why only a few small impact
craters have survued to the present time, even though Europa's surface is much
older
than lo's.
that cover
its
surface.
The streaks on Europa are presumabh caused bv cracks in the icy

coating through which water gushed up and then froze.
These cracks
are a few tens of kilometers wide. To accommodate
all of them, Europa's
surface area would have to ha\e increased bv 10
to 15 percent since the
cracks first began to appear. It seems unreasonable
to suppose that Euexpanding like an inflating balloon, though. Apparentlv, more is
happening on Europa than meets the eye. Perhaps old surface
material
ropa
is
IS somehow being pulled back

down into the mushv laver below the ice
coating and then recycled. Just as the Earth's crust
is pulled back down
into the Earth's mantle in subduction zones
(recall Figure 7-18). Europa's
surface mav represent a water- and-ice versioti of
plate
tectonics.
Ganymede and
Callisto
have
heavily cratered icy surfaces
Europa's densitv can be explained bv 100 km of ice and

water on top ol
an otherwise rockv world. A much thicker layer of water
and ice must
surround Ganymede and Callisto, however, to be consistent
with average densities slightlv less than 2 g/cm^ the rockv cores
of these two
outer satellites must be enveloped in mantles of water
and ice nearly
1000 km thick. The cross-section diagrams in Figure 9-22 show
the
probable interior structure of all four (ialilcan satellites.
The two outer Cialilean satellites have the kind of ancient, cratered
surface normally associated with Moon-like landscapes.
CauNniede even
looks somewhat like our Moon (see Figure 9-23). Of
course, the craters
on both Ganymede and Callisto are of ice rather than rock.
Although Ganvmcde looks vaguelv like our Moon, there are

significant differences. Ganymede has two very different
kinds of terrain that
are distinguished by both appearance and age (see
Figure 9-24). Dark,
polygon-shaped regions are presumed to be the oldest surface
features
because they exhibit a high densitv of craters. Light-colored,
heavily
grooved terrain found between thi' dark angular' islands is
much less cratered and is therefore Noiingei.
Clanymede is the largest satellite in the solar svsiem. Its diameter

is
5270 km which is slightlv greater than the diameter of Mercurv.
he
largest single feature on (ianymcde is a vast, dark,
circular island of an1
cient crust called Galileo Regio (see Figure 9-23).

in
It measures 4000 km
diameter and covers nearly one-third of the hemisphere of
(;anvmede
that faces
away from
lupiier.
Fhis surface feature
is
the onlv one on the
(ialilean satellites that
can be detected with Earth-based telescopes.

It IS casv to distinguish young craters
from old on CJanymcde. The
youngest craters are surrounded bv bright rays of freshlv
exposed ice
(see Figure 9-24). Older craters clearly have
been covered with deijosits
ol dark meteoritic dust. Ihe most ancient
craters are barelv visible on
(;alileo Regio and on the other dark, angular
island remnants of old
crust. The degradation and near obliteration of
the oldest craters probably also involved the slow plastic flow of
Ganymede's
icv surface.
allet
Figure 9-22
satellites
Interiors of the Galilean
These eross-seclion diagrams show
the probable internal structures of the
Galilean
sities
satellites,
four
based on their average den-
and information from
the
Voyager
flybys.
terrestrial worlds
us about the history of the younger, light-colored terIn some places the cratering is
about as dense as on the ancient crust, but in other places it is only onetenth that amount. We thus suspect that this grooved terrain was formed
over a long period of time. The process probably began quite early in
Craters also
tell
rain covered with
numerous grooves.
Ganymede's history and continued through the period of intense meteoritic bombardment. The age of the grooved terrain therefore ranges
from about 4i to 3| billion years.
Figure 9-23
Ganymede
This view from
Voyager 2, taken at a distance of 1.2 million

kilometers, shows the hemisphere that always
faces
away from Jupiter. The surface
is
domi-
nated by a huge, dark circular region called

Galileo Regio, which
Ganymede's ancient
is
the largest
crust.
(NASA)
remnant of
Thf
Young and old terrain on
Figure 9-24
Ganymede
smiiller tr>it'\liiat wnilih
Tim
closeuj) vieii' oj
was taken by Voyagei 2
Ganymedi-
at a ratige
312.000 km. Features as smalt
of only
as 5
km
across
can be seen. Dark, angular islands oj
Ganymede's ancient crust are separated b\

younger, light-colored, grooved terrain. The
southwest edge oj Galileo Regio appears at the
right side of the picture.
(NASA)
High-resolution photographs such as Figure 9-25 show that this

grooved terrain actually consists of parallel mountain ridges up to 1 km
high and spaced 10 to 15 km apart. These features suggest that plate
tectonics may have dominated Ganymede's early history. Water seeping
upward through cracks in the original crust would freeze and force
apart fragments of the original crust, producing jagged, dark islands of
old crust separated by bands of younger, light-colored, heavilv groo\ed
ice. The cracks play a role in Ganymcdean plate tectonics analogous to
the role of the oceanic rifts on the Earth. But unlike thinner-crusted
Europa, where tectonic-like activity perhaps occurs even today, tectonics
on Ganymede bogged down 3 billion years ago as its cooling crust froze
to unprecedented depths.
Figure 9-25
Ganymede
Grooved terrain on
This picture, taken by Voyager
at a riiiigc oj 145.11011
km. shoies an area
roughly as large as the state of Pennsylvania.
The
smallest vuible jeatures are about 3
across.
Numerous
to 15 km
(NASA)
spaced 10
to 1
km.
km
parallel mountain ridges are
apart
and
have heights
up
When Ganymede formed
41 billion years ago, it may have been comcovered with an ocean roughly 1000 km deep. Dining the next
200 million years, the water cooled and a thick coating ol ice developed.
Today this layer of solid ice is probably about 100 km thick. Beneath it
lies a 900-km-thick skjshy mantle of water and ice.
Callisto, Jupiter's outermost Galilean satellite, looks very much like
Ganymede: numerous impact craters scattered over an ancient, dark, icy
crust (see Figure 9-26). There is one obvious difference, however
Gallisto has no younger, grooved terrain. We thus infer that tectonic activity never began on CaUisto. Perhaps because of its greater distance
from Jupiter, the ocean that enveloped young Callisto 41 billion years
ago froze more rapidly and to a greater depth than on Ganvmede, forever preventing tectonic processes. Callisto's icy crust may in fact be several times thicker than Ganymede's, extending to depths of several hundred kilometers. It is bitterly cold on Callisto. Voyager's instruments
measured a noontime temperature of -118C (-180F), and the nighttime
temperature plunges to -193C (-315F).
Voyager
photographed a huge impact feature on Callisto's Jupiterfacing hemispheie. This feature, called the Valhalla Basin, consists of a
large nuinber of concentric rings, separated by 50 to 200 km and ha\ing
diameters ranging up to 3000 km. Valhalla was produced by an asteroidpletely
Figure 9-26
muil Galilean
Callisto
Callishj. Ihc uutit-
salellite. is
almost exacth the
Numerous
same
size
mark
Callisto's icy surface, as seen in this
as Mercury.
craters pock-
mo-
from Voyager I taken at a distance of about 400,000 km. A huge impact
sized object verv earlv in the satellite's history.
saic of views
basin called Valhalla dominates the Jupiter-
facing hemisphere of
inactive world.
to
this
series
frozen, geologically
of concentric rings up
3000 km in diumcin surrounds

(NASA)
the impact
site.
The
great age of Valhalla
lying impact craters
concentric rings.
when
is
inferred both from the presence of oversubstantial vertical relief of the
and the absence of
The
Valhalla impact probably occurred
around 4
billion
young, relatively thin crust was still plastic

enough to flow and reduce the height of the upraised rings in the ice.
The Voyager pictures also showed traces of a Valhalla-like impact on
Ganymede's Galileo Regio. Segments of a system of concentric rings
cover a large portion of this dark island of ancient crust. However, no
obvious impact feature is found at the center of this ring svstem. Apparently the development of grooved terrain completely obliterated the imyears ago,
the
satellite's
pact basin.
opaque
atmosphere rich in methane,
nitrogen, and hydrocarbons
Titan has a thick,
Long before
the Voyager flybys, astronomers knew Saturn's largest satelbe an extraordinary world. It was discovered in 1655. By the early
1900s, several scientists suspected that Titan might have an atmosphere
because it is cool enough and massive enough to retain heavy gases.
Confirming evidence came in 1944 when astronomers discovered spectral
lines of methane in the sunlight reflected from Titan. Titan is the only
satellite in the solar system known to have an appreciable atmosphere.
Because of this atmosphere. Titan was a primary target for the Voyager missions. To everyone's. chagrin, however, the Voyagers spent hour
after precious hour sending back featureless images such as Figure 9-27.
Titan's cloud cover is so thick that it blocks any view of the surface and
allows very little sunlight to penetrate dow'n to the ground; the surface
of Titan must be a dark, gloomy place.
In size, mass, and average density. Titan is quite similar to the largest
Jovian satellites. We would thus expect its internal structure to resemble
those of Ganymede and Callisto
a rocky core surrounded by a mantle
of frozen water nearly 1000 km thick.
Titan's thick atmosphere distinguishes it from all other satellites. The
atmospheric pressinc at Titan's siuface is 1.6 bars, or 60 percent greater
lite
to
Tlif smaller
lenestual worlds
iliaii ihc atmospheric pressure at sea level on Earth, even though litairs
surface gravity is lower than the Earth's. Considerably more gas must be
weighing down on Titan than on Earth. About ten times more gas lies
each square centimeter ol Titan's surface than above Earth's sur-
.il)o\e
face.
What
factors leading to the formation of Titan's
atmosphere did noi

or Clallisto? Eor one, Titan formed in a much cooler
part of the solar nebula. In contrast to the materials available in the
warmer conditions near Jupiter, the ices from which Titan accreted
probably contained substantial amounts of fro/en methane and ammoexist for
Ganymede
f itan's interior became warm

through the decav of naturally ocurring radioactive isotopes, these ices \a|Jorized, producing an atmosphere around the young satellite. Methane (CH,) is stable in sunlight
and lemained in the atmosphere, but ammonia (NH,) is easily broken
(i(nvn into nitrogen and hydrogen by the Sun's ultraviolet radiation. Titan's gravity is too weak to retain hydrogen, so it escaped into space.
nia. .As
Even today, hydrogen

Figure 9-27
Titan
taken by Voyager
Tlus
vii'u'
iij
I ilaii
2 from a dUtance of 4i
irii\
mil-
lion kilomelen. Veiy feu' features are visible in
the thick,
large
unbroken haze that surrounds
salellile.
nearly
The main haze
300 km above
layer
i.\
this
located
Titan's surface. (.\ASAl
The
dissociation of
is
escaping from Titan
at a substantial rate.
ammonia and
loss of hydrogen leaves Titan with

an abundant supply of nitrogen. Voyager data suggest that roughlv 90
percent of Titan's atmosphere is nitrogen. The two next most-abundant
gases are argon and methane.
The interaction of sunlight with methane induces chemical readions
that produce a variety of carbon-hydrogen compounds called hydrocarbons. Voyager's instrutnents detected small amounts of man\ Indrocarbons such as ethane (CaH,;), acetylene (C._>H._,), ethylene (C2H4), and propane (C;iH) in Titan's atmosphere. Nitrogen combines with these
hydrocarbons to produce other compounds such as hvdrogen cyanide
(HC:N). some of which are the building blocks of the organic molecules
on which life is based. There is little reason to suspect life on Titan its
surface temperature is 95 K (= -178C = -288F)
but a more detailed
stuch of its chemistry may shed light on the origins of life on Earth.
On Earth, the atmospheric pressure and temperature are near the triple point of water, so ii is foimd in all three phases: liquid, solid, and
gas. The atmospheric pressure and temperature at Titan's surface are
near the triple point of methane, however. Methane mav thus plav a role
on Titan similar to that which water plays on Earth. Methane snowflakes
may faff onto frozen methane potar caps and methane raindrops descend into methane rivers, lakes, and seas in warmer areas.
Some molecules are capable of joining together in long, repeating
molecular chains, to form substances called polymers. Many of the hydiocarbons and carbon-nitrogen compounds in
iian's atmospheie form
such pohmers. Droplets of some polvmers rem,iin suspended in the atmosphere to foini the kind of mixtine called an aerosol (see Eigiue
9-28). but the heavier poivmer particles settle down onto Titan's surface,
probablv covering it with a thick laver of siick\. larlike goo. .Some scientists estimate that the deposiis oi Intlrocai Ixin sludge on Titan ma\ be
km
(ieep.
hi- one ol llie m.ijoi objectives foi futiue planespacecraft ec)uip|)ed with radar could map Titan's
cloud-covered surface in much the same wav that the sinlace of Venus
Til.m will pi()l5ai)l\
tary ex|)loralion.
was examined
in the 197()s.
into Titan's aimosiJJiere to
do
glidei or balloon coidd be
could descend to the siuface 10

"goo." Tentatixe plans
Titan in the 199()s.
at
N;\S.\
sam|)lc'
(.ill
loi
some
of the
missions lo
dropped
piobe
hydrocaibon
both Saturn and
diiect chemical anahsis.
.A
Tlie smaller leni'slricil
Figure 9-28
Titan's atmosphere
ager data suggest that
Titati's
middle layer
lowest layer
cles.
is
opaque
Methane rain
the
layer ab-
Sun. The
is
The
Ultraviolet absorption layer
/ .injii'mtrd parti-
cloud', miiy exisl
which
tan's surface,
from
tn visitilr lii^hl.
an aerosal
is
The Voy-
atmosphere has
The uppermost
three distinct layers.
sorbs ultraviolet radiation
uwrUh
near Ti-
at a temperature of
Optical haze
about 95 K. (Adapted from Tobias Owen)
Aerosol layer
/.*/.:'.: \'.,v;'-'
200
.5
<
Methane
Triton has an unusual orbit
rainfall
'
_
Methane clouds
Neptune's largest moon, Triton (see Figure 9-29), was discovered in

1846. Along with Titan, the Galilean satellites, and our Moon, Triton is
one of the seven giant moons of oiu" solar system. Its size is somewhat
uncertain (its diameter is estimated to be 4000 km), but it clearly is of
planetary dimensions.
Neptune's second largest moon, Nereid, was discovered in 1949.
Nereid is quite small, with a diameter estimated to be somewhat less than
1000 km.
Triton and Nereid have unusual orbits. Triton's orbit is retrograde
and Nereid's is the most eccentric of any satellite in the solar system.
Triton completes its backward orbit in only 5 days, but Nereid takes
nearly 360 days to go once around Neptune. The distance between
Nereid and Neptune varies from 1.4 million kilometers to 9.7 million kilometers as this tiny moon moves along its highly elliptical path.
An extraordinary fact has emerged from recent studies of Triton: its
orbit is decaying. Triton is slowing and spiraling in toward Neptune. In
10 million to 100 million years, Triton will be so close to Neptune that
the planet's gravity will tear Triton apart. Thus Neptune will develop a
spectacular ring system as rock fragments gradually spread out along the
satellite's
former
orbit.
Spectroscopic studies suggest that Triton's surface is covered with rock

rather than ice. Infrared observations indicate the presence of a sea of
liquid nitrogen on Triton, and traces of a methane atmosphere have also
been detected. Triton

Figure 9-29
Neptune and Triton
one of the seven planet-sized
Triton,
lites
solar system,
circles
is
Neptune
every six days.
creasing
and
seen here near Neptune. Triton

in
The
a retrograde
size
orbit
of Triton's
is
de-
the satellite will be torn apart by
Neptune's gravity in about 100 million years.

(Yerkes
and McDonald
Observatories)
be a fascinating target for Voyager 2 in 1989.
intriguing theory about the unusual orbits of the Neptunian satelinvolves Pluto. Pluto's orbit about the Sun is more elliptical than the
any other planet in the solar system. In fact, Pluto's orbit is so

it is sometimes closer to the Sun than Neptune is.
Pluto has a moon called Charon that was discovered in 1978. Charon
and Pluto are remarkably comparable in size (see Figure 9-30). Their
orbit of
once
orbit
will
An
satellites in the
eccentric that
measurements are rather uncertain, but reasonable estimates place
Plu-
ii/ln
Figure 9-30
Pluto and Charon
moon Charon appears

lump on one
Inmlrinl
u'(>ilil\
179
Plulos
as a slight elongation a
side o] this greatlt enlarged
image of the planet. Pluto and Charon are

separated by only 19.700 km. The Pluto-
Charon system deserves
to be called
a double
planet because these two objects resemble each

other in mass
any other
and
size
more
closely
than do
planet-satellite pair in the solar sys-
tem. (Courtesy of James
Chnst\ and
Robert Harrington)
diameter at 3000 km and Charon's at 1400 km. Pluto and Charon

to resemble typical moderate-sized satellites of the Jovian planets.
Especially in view of Pluto's unusual orbit about the Sun, it is reasonable to wonder whether Pluto might be an escaped satellite that once
orbited Neptune. Perhaps some sort of cataclysmic event in the ancient
past reversed the direction of Triton's orbit, flung Nereid into its highly
elliptical orbit, and catapulted Pluto awav from Neptune.
The idea that Pluto is an escaped satellite of .Neptune was first proposed in 1936. Problems with this proposal ha\e, however, long been
recognized. For one, the present-day orbits of Neptune and Pluto do not
intersect
the two planets are never closer than 384 million km. If Pluto
was catapulted awav from Neptune (for example, by a near collision with
Triton and Nereid), why does Pluto's orbit not pass through Neptune's
to's
seem
orbit at the point of this event?
With the discovery of Charon, astronomers made the intriguing prounknown, massive, planetlike object may have been involved in Pluto's escape from Neptune. Perhaps Triton, Nereid, and
posal that an
Pluto
all
orbited
cient past.
Then,
Neptune along well-behaved regular

in a near-collision, the
turbed the orbits of
all
unknown
orbits in the anplanet se\erely per-
and Nereid were swung

from the unknown planet
and catapulted them into an orbit about the
three
satellites.
Iriton
into their present unusual orbits. Tidal forces
tore Pluto into two pieces
Sim.
Howevei
might be a massive transby no means commonh accepted by astronomers. One of

the main objections concerns the resemblance of the Pluto-C^haron system to a pair of asteroids. As we shall see in the next chapter, there is
increasing evidence that some asteroids may be double objects
two comparabh sized chunks of roc k in dose pro\imii\ to each other.
Pluto planet
intriguing, the idea that there

is
Summary
Mercury's surface is pocked with craters like those of the Moon, but there
are extensive, smooth iiiteniater plains. These features appeal to have
formed as the crust of the planet .solidified.
The smaller
Inn-slruit wurtdi
Surface temperatures on Mercury range from 100 to 700 K, the greatest

range known on any of the planets.
Mercury has an iron core much

Mercury's magnetic
like that
of the Earth.
produces a magnetosphere that blocks the solar
field
wind from the surface of the
planet.
The
Earth-facing side of the Moon displays light-colored, heavily cratered

highlands (terrae) and dark-colored, smooth-surfaced maria. The Moon's
far side has no maria.
The Moon's
crust
of the Moon).
is
thicker than the Earth's (and thickest
The Moon
has no magnetic
on the
far side
field.
Lunar rocks contain no water and also differ from terrestrial rocks by
being relatively enriched in the refractory elements and depleted in the
volatile elements.
The most widely accepted theory about lunar origin holds that the Moon
accreted from rock fragments orbiting the Earth soon after the Earth itself
had formed.
The
and
and
anorthositic rocks of the lunar highlands

4.3 billion vears ago, whereas the
3.8 billion years ago.
mare
were formed between 4.0

between 3.1
basalts solidified
The Moon was molten in its early stages, and the anorthositic crust solidified from low-density magma that floated to the lunar surface. The mare
basins were created later, by the impact of planetesimals and filled with
lava
from the lunar
interior.
Meteoroid impacts have been the only significant "weathering" agent on

the Moon. The Moon's regolith ("soil" layer) was formed by meteoritic
action, and glasses and breccias of meteoritic origin are very common.
The Moon's
surface has
undergone very
little
change
in the past 3 billion
vears.
The
inner two Galilean moons, lo and Europa, are roughlv the size of
have densities similar to that of the Moon. The outer two
Galilean moons, Ganymede and Callisto, are roughly the size of Mercurv
and are lower in density than the Moon or Mercury.
our
Moon and
lo is covered with a colorful layer of sulfur compounds deposited bv

quent explosive eruptions from volcanic vents.
lo's volcanic
eruptions resemble terrestrial geysers.
The energy
comes from tidal forces that flex the moon

tween the planet and the other large moons.
lo's interior
as
it
fre-
heating
passes be-
Europa is covered with a smooth layer of frozen water that is crisscrossed

by an intricate pattern of long cracks, probably produced bv tidal flexing
of the moon.
The
heavily cratered surface of
Ganymede
is
composed of frozen
water.
Large polygons of dark, ancient surface are separated by regions of heavgrooved, lighter-colored, younger terrain. Plate tectonics apparently
operated during the early history of Ganymede.
ily
Callisto also has a heavily cratered crust of frozen water, but plate tectonics
apparently never operated on this moon, presumably because
developed a
it
quickly
thick, solid crust.
The
Galilean satellites probably formed through a process of accretion

formed the solar system about the Sun. Inil on
a smaller scale.
similar to the process that
tilirt
l/nnliial
imrt/l.s
The largest satellite ol Satuiii. ii.in. Ii.in a dense nitrogen atmosphere in

which methane mav pla\ a role similar to that of water on Earth. A variety of hydrocarbons are lormcd l)v the interaction of sunlight with methane, creating an aerosol laver in Titan's atinospheie and probablv a thick
i
sludge on
The
is
surface.
a thin
Iriton. moves in a retrograde orbit that

a rocky surface with a liquid nitrogen sea
Neptune,
may have
decaying. Triton
and
Review questions
its
largest satellite of
methane atmosphere.
1 Why are naked eye observations of Mercury best made at dusk or dawn,
whereas telescopic ob.servations are best made around noon?
2 Why do astronoiners believe Mercurv to be "the most iron-rich planet in

the solar svstem"?
3 Why is more lunar detail \isible through a telescope when the
near quarter phase than when it is at full phase?
Why
are nearly
facing the Earth?
all
Moon
is
of the lunar maria located on the side of the Moon

are there more craters on the tar side of the Moon
Why
than the near side?

5
How would vou
Why do
prove
to
someone
that the
Moon
has no atmosphere?
you suppose that no Apollo mission landed on the
far side of the
Moon?
it
7 How does the Galilean satellite system resemble the solar system?
different?
8 With
all its
volcanic activity,
why
How
is
doesn't lo possess an atmosphere?
9 Compare and contrast the surface features of the four Galilean satellites,
discussing the relative geological activity and evolution of these satellites.
Advanced questions
10
Why
.solar
is it reasonable to presume that none of the large satellites in the

system, including our Moon, possess a substantial magnetic field?
11 Long before the V'ovager flvbvs. Earth-based astronomers reported that h

appeared brighter than usual for a few hours after emerging from Jupiter's
shadow. Based on what we know about the material ejected from lo's volcanoes, explain this brief anomalous brightening of lo.
12 L'sing the diameter of h) (3630 km) as a scale, estimate the height to

which the plume of Prometheus rises above the surface of lo in Figure 9-19.
13 What evidence do we have that the surface features on Mercurv were not
formed during recent geological historv?
14 Imagine that you are planning a hni.u landing mission. What tvpe of
landing site would you select in order to obtain bedrock? Where might vou
land to search for evidence of recent volcanic activity?
15 The idea has been advanced that withoiu the presence of the Moon in
om- sky astronomy would have developed far more slowlv. Please comment.
16 (Compare the a<lvaniages and disadvantages of exploring
astronauts as opposed to mobile roliots.
ifie
Moon
with
Thr
17 Speculate on the possibility that

bor some sort of marine life.
For further reading
F.iiropa,
Ganymede, or
unalU-i leiri'slnal irnilih
Callisto
might har-
Cadogan, P. "The Moon's Origin." Mercmj, Mar./Apr. 1983, p. 34.

Cooper, H. Apollo on the Moon and Moon Rocks. Dial, 1970. Two accounts
by a respected journalisl of the Apollo 11 mission and the analysis of the
material the astronauts brought back.
French, B. The
Moon
Moon
Book. Penguin, 1977.
layperson's introduction to the
by a lunar scientist.
Hartmann, W. "The Moon's Early History." Astronomy, Sept. 1976, p. 6.

Hartmann, W. "The Significance of the Planet Mercury." Sky & Telescope,
May
1976, p. 307.
Johnson, T. "The Galilean

Solar System,
2nd
Satellites."
In Beatty,
ed. Skv Publishing
J., et al.,
eds.
The
New
and Cambridge University
Press, 1982.
Johnson, T., and Soderbloni, L. "lo." Scientific American, Dec. 1983.

Morrison, D. "Four New Worlds: The Voyager Exploration of Jupiter's
Satellites." Mercury,
Murray,
B.,
and Burgess,
May/June 1980,
p. 53.
E. Flight to Mercury.
Columbia University
Press,
1977.
Scientific American, May 1976.
T. "Titan." Scientific American, Feb. 1982.
Murray, B. "Mercury."
Owen,
Soderblom,
L.
1980.
"The Galilean Moons of
Jupiter." Scientific American, Jan.
10
The head of Comet Halley

five spacecmjt passed near
Interplanetary
vagabonds
In March 1986.
Comet Halley. This
picture was taken by the Halley Multicolor
Camera on board
launched h\
the
the spacecraft Giotto,
European
.Space .Agency. DiJ
fereni colors represent different levels of tighl

inletisity.
The dark area
in the
upper
left
is
illuminates the comet
ner of the scene.

tends
from
from
the lower right cor-
broad, bright
the nucleus
the
Sun
dust-covered icy nucleus of the comet. The
plume
ex-
toward the Sun. Details
of the nucleus are seen in Figure 10-15.
(Max Planck
Inslilul fiir .Aeronomie)
The planets are not

chapter
nificant
the only objecLs that
move in orbits about the Sun. In

and comets that are small but
ive discuss the ttsteroids, meteoroids,
members
oj the solar system.
We
clues about the histoty of the solar system.
meteoroids
and comets may have had a
this
sig-
learn that these objects provide major
We
also disciLss .speculations that
significant effect on our planet, in-
cluding the extinction of more than one-half the species living on Earth some
6?
million years ago.
Many
rocks and chunks of ice that condensed out of the primordial solar
nebula
still
continue lo orbit the Siui. just as heat from the protosun
classes of planets terrestrial and Jovian

two main Ivpes
of inierp!aneiar\ material were created. Near ihe Sun, interplanetary
debris consists ol roi k tragments called asteioids or meteoroids. Far
Irom the Sun. there aie ninnerous dust-covered cluniks of ice called
piodiucd two
comets.
hilnl/laniiiin vaiâhiiiids
Bode's law led to the

discovery of numerous
asteroids between the orbits
of Mars and Jupiter
In the late- 1700s, a young German astronomer, Johann Elert Bode, popularized a simple rule that describes the distances of the planets from
known today as Bode's law an unfortunot a physical law and was not invented by Bode.
It had first been published in 1766 by Johann Titius, a German physicist
and mathematician. Most astronomers now regard this "law" as merely a
coincidence, but it did lead directly to the discovery of a large number
of previously unknown objects that orbit the Sun.
Bode's rule for remembering the distances of the planets from the
the Sun. This rule
nate
name because
Sun goes
1
is
it
usually
is
like this:
Write down the sequence of numbers 0, 3, 6, 12, 24, 48, 96, ....
(Note that each number after the second one is simply twice the preceding number.)
Add
Divide each of the resulting numbers by 10.
4 to each
number
in the secjuence.
in Table 10-1, the final result is a series of numbers that corresponds remarkably well to the distances (in AU) of the planets from
the Sim.
Astronomers regarded Bode's rule as merely a useful trick for remembering the planetary distances until 1 78 1 when William Herschel
discovered Uranus whose average distance from the Sun is very near
that predicted by Bode's scheme. Suddenly it seemed far more likely that
Bode's rule inight actually represent some physical property of the solar
As shown
system.
Astronomers now looked with new interest at the "missing planet" in

the gap between the orbits of Mars and
Jupiter. On January 1, 1 80 1, the Sicilian astronomer Giuseppe Piazzi
noticed a dim. previously uncharted star that shifted its position slightly
over the next several nights. Later that year, the orbit of this object was
determined to lie between the orbits of Mars and Jupiter. At Piazzi's request, the object was named Ceres (pronounced SEE-reez) after the patron goddess of Sicily.
the sequence of Bode's law
TABLE
lO-l
Bode's law
Actual distance
Bode-Titius
Planet
progression
(0-l-4)/10
(AU)
0.4
Mercury
0.39
(3
4)/10
0.7
Venus
0.72
(6
4)/10
1.0
Earth
1.00
(12
4)/10
1.6
Mars
1.52
(24
4)/10
2.8
(48
4)/10
5.2
Jupiter
(96
4)/10
10.0
Saturn
9.54
(192-l-4)/10
19.6
Uranus
19.18
(384 -l-4)/10
38.8
Neptune
30.06
77.2
Pluto
39.44
(768
4)/10
5.20
liilf'iplaiii'fdn'
vn^dhintds
Ceres orbits the Sun once every 4.6 years at an average distance of
AU. This orbit is in remarkable agreement with the distance "predicted" by Bode's law for the missing planet, (leres is \ery small, however
its diameter is estimated to be a scant 1000 km.
I'hus Ceres does
not cjualify as a full-fledged planet, and astronomers continued the
2.77
search.
In 1802, the
German astronomer Heinrich Olbers discovered another

moved against the background stars. He called
faint, starlike object that
Like Ceres, Pallas orbits the Sun every 4.6 years at an average
AU. Pallas is even dimmer and smaller than Ceres, with
an estimated diameter of only 600 km. Obviously, Pallas is not the missing planet either.
The discovery of these two small objects with siinilar orbits at the distance expected for the missing planet led astronomers to suspect that
Bode's missing planet might have somehow broken apart or exploded.
The search for other small objects therefore continued. Only two more
were found ^Juno and Vesta until the mid- 800s, when telescopic
equipment and techniques had improved. Astronomers then began to
stumble across many more such objects circling the Sun between the orbits of Mars and Jupiter. These objects are today called asteroids or
it
Pallas.
distance of 2.77
minor
planets.
fhe next major breakthrough came in 1891 when the German astronomer Max Wolf began using photographic techniques to search for aster-
A total of 300 asteroids had been found up to that time, each

painstakingly discoveied by scrutinizing the skies for faint, uncharted
stars that would shift their positions slowly from one night to the next.
oids.
With the advent of astrc^photography, however, the floodgates were

opened. An astronomer simply aims a cainera-equipped telescope at the
stars and takes a long exposure. If an asteroid happens to be in the field
of view, it leaves a distinctive, blurred trail on the photographic plate
because of its movement along its orbit during the time exposure (see
Figure 1 0-1). Using this technique. Wolf alone discovered 228 asteroids.
Although thousands of asteroids have been sighted, only 3000 have
well-determined orbits. An additional 6000 asteroids have "passable" orbits, but the orbits of 20,000 more have never been determined. The
orbits of all officiaily discovered asteroids are published annually in the
famous Soviet catalogue Ephemmdes of Minor Planets.
To become the official discoverer of an asteroid, you must do a lot
more than just produce one photograph of a blurred trail whose path
does not match any known orbit listed in the Ephemerides. You must track
the asteroid long enough to compute for it an accurate and reliable orbit
(important data may be available from colleagues who may have inadvertently photographed it on earlier occasions). Then you must prove
the accuracy of the orbit by locating the asteroid again on at least one
succeeding opposition. At that time, an official number will be assigned
to vour asteroid (Ceres is I, Pallas is 2, and so forth). \\)\.\ will also be
given the privilege of selecting a name lor vour asteroid.
Ceres is uiuiuestionably the largest asteioid. With its diameter of
nearly 1000 km, Cicres accounts for about 30 percent of the mass of all
the asteroids combined. Only two others (Pallas and Vesta) have diameters greater than 300 km. Thirty other asteroids have diameters between
200 arul 300 km, and there are 200 more asteroids trigger than 100 km
across.
Astronomers estini.iie thai lougliK 100,000 .isicioids exist

enough to a])|)e.n on |)hoI()graphs t.iken from K.ulh.
bright
that are
I
he vast
liilnpluHi'liiiy
Figure 10- 1
delected
liy
Two
asteroids
their blurred Irads
photographs of the
stars.
vagabonds
Asteroids are
on time-exposure
The images of two
asteroids are seen in this picture. Astronomers
sometimes find asteroids accidentally while pho-
tographing various portions of the sky
fin
other purposes. (Yerkes Observatory)
km
and Juno, most

between 2 and 3i AU. This region
of the solar system between the orbits of Mars and Jupiter is called the
asteroid belt (see Figure 10-2). Asteroids whose orbits lie entirely within
majority are
less
than
asteroids circle the
this
The asteroid belt Most asterSun m a U-AU-wide belt between the orbits of Mars and Jupiter. The orbits of Ceres, Pallas, and Juno are indicated.
The orbits of two Apollo asteroids (Apollo and
Figure 10-2
oids orbit the
Icarus) are also shown.
Sun
across. Like Ceres, Pallas,
at distances
region are called belt asteroids.
I tilnl)tn>iii(ir\
vagabonds
187
The combined matter of all the asteroids (including an estimate for

those not yet officially known) would produce an object barely 1500 km
in diameter, considerably smaller than our Moon. If the asteroids are the
fragments of Bode's missing planet, it cannot have been large enough to
rank with the terrestrial planets. Furthermore, geologists and physicists
have never been able to produce a good theory explaining how a planet
could fragment or explode. There is little support today for the hypothesis of a shattered planet. It seems more reasonable that the asteroids
are debris left over from the formation of the solar system out of the
solar nebula.
Constant gravitational perturbations caused bv the enormous mass of

Jupiter probably kept planetesimals from accreting into larger objects in
the region between Mars and Jupiter. The missing planet simply never
had a chance to form. What remains today in the gap between the orbits
and Mars appears to be merely a remnant of scattered debris
from the original solar nebula that elsewhere accreted into planets.
of Jupiter
Jupiter's gravity affects the
structure of the asteroid belt
and captures asteroids along

its
orbit
Consider a belt asteroid moving along its orbit between the orbits of Jupiter and Mars. Each time the faster-moving asteroid catches up with
and passes massive Jupiter, it experiences a slight gravitational tug toward Jupiter. This tug tends to alter the asteroid's orbit slightly. However, over the ages, these close passes occur at different points along the
asteroid's orbit, so their effects cancel each other.
Now
imagine an asteroid circling the Sun once every 5.93 years,

exactly half of Jupiter's orbital period. On every second trip
around the Sun, the asteroid finds itself lined up between Jupiter and
which
is
Sun again and again, always at the same location and with the same
These gravitational effects add up to deflect the asteroid
from its original 5.93-year orbit, leaving a gap in the asteroid belt. -Acthe
orientation.
cording to Kepler's third law, a period of 5.93 years corresponds to a

semimajor axis of 3.28 WJ. Because of Jupiter, there are no asteroids
that orbit the Sun at this average distance.
Similarly, we would expect to find a gap corresponding to an orbital
period of one-third Jupiter's period, or 3.95 years. Comparable gaps
should exist for other simple relationships between the periods of asteroids and Jupiter. The data graphed in Figure 10-3 show that such gaps
do exist. They are called Kirkwood gaps in honor of the American astronomer Daniel Kirkwood, who first drew attention to them.
Although Jupiter's gravitational
pull depletes certain orbits in the ascaptures asteroids at certain other locations much farther
from the Sun. There are two specific points along Jupiter's orbit where
the gravitational forces of the Sun ancf Jupiter work together to hold as-
teroid belt,
it
These two locations are called the Lagrange points /.

honor of the French mathematician whose cak ul.uions revealed this gravitational effect. Point L, is located one-sixth of the way
around Jupiter's orbii ahead of the planet, and point L-, occupies a simiteroids in orbit.
and
lai
/,;,
in
position behind the planet (see Figure 10-4).

I
he asteroids traj^pcd
asteroids,
named
at Jupiter's Lagrange points are called Trojan

indisidually after heroes of the Trojan War. Nearis
two dozen Trojan asteroids have been catalogued, and some astnjnomers
may be as many as 700 rock fragments orbiting near
each Lagrange point.
believe that there
hitnplfuirttn^ vdiâhonds
Distance from Sun
(AU)
3.0
2.5
3.5
4.0
4.5
5.0
Lagrange
-^ /
\^
point L4
\\
"/M
/ \60
Sun
\/ 60'
\
/
//
\
\
/
Lagrange
I
point
L-,
/ /
/
/y'
^\,v //^
\
'^,
Trojan asteroids
(trailing
group)
Orbital pcricid (Jovian synodic years)
Figure 10-3
Incidentally, the other three
The Kirkwood gaps
[left]
Lagrange points (L,, L^, and L,) lie along

and the Sun, but these points are not
This histogram displays the numbers of aster-
a straight line connecting Jupiter
oids at various distances
capable of capturing asteroids.

Asteroids at the L4 and L5 points along Jupiter's orbit have been
known since the first Trojan asteroid was discovered in 1906. In addition, Lagrange points exist at many places in the solar system. While
passing Saturn, the Voyager spacecraft discovered tiny statellites at the
L4 and Lj points of Saturn and two of its satellites (Tethys and Dione).
Small clouds of dustgrain-sized particles have been observed at the L4
and L-, points of the Earth-Moon system. A group called the L-, Society
argues that the Lr, point of the Earth-Moon system would be an ideal
location for a huge space station with a permanent human ptjpulation.
Despite careful searches, no asteroids have been found at the stable Lagrange point of the Earth-Sun and the Saturn-Sun systems.
that very
few
bital periods
(such as
i,
asteroids
I, 3)
from
have
correspond
to
the
orbits
Sun. Notice
whose or-
simple fractions
of Jupiter's orbital period.
Gravitational perturbatiotis caused by repeated
alignments with Jupiter have deflected asteroids
away from
these orbits.
Figure 10-4 [right]
The Trojan asteroids
Aslerinds are trapped at the two
Lagtange
points along Jupiter's orbit by the combined
and the Sun.

are named after
gravitational forces of Jupiter

Asteroids at these locations
Homeric heroes of
the Trojan
War.
Asteroids occasionally collide

with each other and with the
inner planets
In addition to the belt asteroids and the Trojan asteroids, there are
other asteroids distinguished by highly elliptical orbits that bring them
into the inner regions of the solar system. Occasionally one of these asteroids passes quite close to Earth. Figure 10-5 shows Eros as it passed
within 23 million kilometers of our planet in 1931. In 1968, Icarus
passed Earth at a distance of only 6 million kilometers. One of the clos-
near-misses in recent history occurred in 1937, when Hermes passed

km only a little more than twice the distance to the Moon.
During these close encounters, astronomers can examine the details of
the asteroids. For example, an asteroid's brightness is often observed to
vary in a periodic fashion, presumably because of its different surfaces
est
us at a distance of 900,000
being turned toward us as the asteroid rotates. Periodic brightness variations thus reveal the asteroid's rate of rotation. Typical asteroid rotation
periods are in the range of 5 to 20 hours.
Iiilerfilimelun
Figure 10-S
Eros
Eros
uuls that occasionally passes
Tim
when
ptwtograph
iras taken
vagabonds
one of lite aslrinear the Earth.
is
February 1931.
and the Earth

was only 23 million kilometers. The dimen.wins
of Eros are roughly 10 by 20 by 30 km. This
asteroid rotates with a period of 5.27 hours.
the distance between Eros
(Yerkfs Obseri<atury)
Careful scrutiny of an asicroid's brightness variations can also reveal

the asteroid's shape and dimensions. Only the largest asteroids, such as
Ceres, Pallas, and Vesta, are spherical, because onlv thev have enough
gravity to pull themseKes into a spherical shape. Smaller asteroids permanentlv retain the odd shapes produced by interasteroid collisions. A
small asteroid looks dim when seen end-on but appears brighter when
seen br(jadside. Measurements of an asteroid's brightness variations thus

tell astronomers a lot about its shape.
There is ample evidence that interasteroid collisions successfully frag-
ment asteroids into small pieces. In 1918. for instance, the Japanese astronomer Kiyotsugu Hirayama drew attention to groups of asteroids that
share nearly identical orbits. These groupings presumablv resulted from
the fragmentation of parent asteroids.
The collision of kilometer-sized asteroids must be an awesome event.

Tvpical collision velocities are estimated at
to 5 km/sec (2000 to
1,000
miles per hour), which is more than sufficient to shatter rock. Onlv in a
high-velocity collision is there enough energv to shatter an asteroid perm.mently. In collisions at low velocities, the resulting fragments may not
achie\e escape velocity fiom each other and will reassemble because of
1
mutual gravitational attraction. Alternatively, several large fragments may end up orbiting each other, which is probabK what happened to both Pallas and Victoria. They are binary asteroids, eai h lonsisting of a main asteroid and a large satellite.
Interasteroid collisions produce numerous chunks of rock, main of
which eventuallv rain down on Venus, Karth, and Mars. Fortimatelv for
us, the vast majority of these asteroid fragments (usuallv called meteoroids) arc quite small. On rare occasions, liowcxcr. a large fragment does
collide with our planet. The result is an impact crater whose di.nneier
depends on the mass and speed of the impinging object.
C3nc of the most impressive and besi-|jiesei \ed terresti ial impact craters is the famous Barringer Crater near Winslow. .Arizona. The ciater
measures 1.2 km across and is 200 m dee]) (see Figme 10-6). The crater
was formed 25.000 years ago when an iron-rich object measining
roughly 50 m across struck the ground with a speed estimated al
km/
their
Intcrj)lan(l(n-\
The Barringer Crater An

measuring 50 m across struck
Anzona 23,000 years ago. The
Figure 10-6
iron meteoroid
the
ground
in
result
was
crater
measuring 1.2 km
200 m
vagnbunds
this beautifully
deep at
its
center.
symmetrical impact
in diameter
and
(Meteor
Crater Enterprises)
The resulting blast was equal to the detona20-megaton hydrogen bomb.

Iron is one of the more abundant elements in the universe (recall
Table 6-4) as well as one of the most common rock- forming elements
(recall Table 6-5), so it is not surprising that iron is an important constituent of asteroids and their fragments called meteoroids. Another element, iridium one of several elements that geologists call siderophiles,
sec (25.000 miles per hour).
tion of a
or "iron lovers"
common
minerals but is rare in ordinary

can therefore tell us
about the rate at which meteoritic material has been deposited on the
Earth over the ages. A team of geologists, including Walter Alvarez and
his physicist father Luis Alvarez from the University of California at
BerkelcN, made such measurements in the late 1970s.
Working at a site of exposed marine limestone in the Apennine
Mountains in Italy, the Alvarez team discovered an exceptionally high
abundance of iridium in a dark-colored layer of clay between the limestone strata (see Figure 10-7). Since this 1979 discovery, a comparable
layer of iridium-rich material has been uncovered at a variety of sites
around the world. In all cases, geological dating reveals that this apparently worldwide layer of iridium-rich clay is about 63 million years old.
Paleontologists were quick to realize the profound significance of this
particular date. Around 63 million years ago, all the dinosaurs became
extinct. In fact, at that time a staggering 65 percent of all the species on
Earth disappeared within a relatively brief span of time.
The AKarez discovery suggests a startling explanation for the drainatic extinction of more than half the life-forms that inhabited our
planet at the end of the Mesozoic era: perhaps an asteroid hit the Earth.
An asteroid 10 km in diameter slamming into the Earth would have
thrown enough dust into the atmosphere to block out sunlight for several years. As plants died for lack of sunshine, the dinosaurs would have
starved to death along with many other creatures in the vegetation food
chain. Eventually the dust settled, depositing an iridium-rich layer
around the world. Tiny, rodentlike creatures that could ferret out seeds
and nuts were prominent among the animals that managed to survive
rocks.
-is
in iron-rich
Measurements of iridium
in the Earth's crust
Iiilft/>liiiitl(in viifâboiuLs
Figure 10-7
191
The iridium-rich layer of
clay
This phulugtaph of strata in the Apen-
nine
Mountanu of Italy shows a dark-colored

mdium-nch clay sandwiched between
layer of
white limestone (below) from the late Mesozoic

era
and
grayush limestone (above)
from
the
early Cenozoic era. This indium-rich layer
be the result of
an asteroid impact
the extinction of the dinosaurs.
may
that caiuiil
The coin
the size of a U.S. quarter. (Courtesy of

Walter Alvarez)
is
the holocaust, setting the stage for the rise of
mammals
in the
Cenozoic
era.
Some geologists and paleontologists are not yet convinced that such a
meteoroid impact did produce "the great dying out" at the end of the
Mesozoic era, but many scientists agree that this hypothesis fits the available evidence better than other explanations that have been offered.
Meteorites are classified

as stones, stony irons,
or irons, depending on
their
composition
meteoroid. like an asteroid, is a chunk of rock in space. There is no

dividing line between meteoroids and asteroids, but the term as-
official
teroid
is
generally applied only to objects larger than a few
hundred
meters across.
A meteor
that
is
visible at
is
the brief flash of light (sometimes called a shooting star)

night when a meteoroid strikes the Earth's atmosphere
(see Figure 10-8).

If a piece of rock survives its fiery descent through the atmosphere,
the object that reaches the ground is called a meteorite. People have
been finding specimens for thousands of years, and descriptions of mete-
orites
appear
in
ancient Chinese. Greek, and
Roman
literature.
Our
an-
on these "rocks from heaven."

Meteorites are classified into three broad categories: stones, stony-
cestors placed special significance
and irons. As their name suggests, stony meteorites, or stones,

look like ordinary rocks at first glance, but they are sometimes co\ered
with a fusion crust (sec Figure 10-9). This crust is produced by the
r7iomentar\ melting ot the meteorite's outer layers during its fiery descent thrcHigh the atmosphere. When a stony meteorite is cut in two and
polished, tinv Hecks of iron are sometimes found in the rock (see Figure
irons,
10-10).
Although stony meteorites account for nearly 93 percent of all meteon the Earth, stones are the most difficult speci-
oritic material that falls
Inlf'iplaiu'laii;
Figure 10-8
meteor
meteor
vagabonds
pro-
is
duced when a piece of interplanetary rock or

dust strikes the Earth's atmosphere at high
speed. Exceptionally bright meteors, such as tin
one shown in
long exposure (notice the
this
(Com
star trails), are usually called fireballs.

tesy
of Ronald A. Oriti)
mens
to find. If they
for a few years, they

Figure 10-9
[left]
restrial rocks.
stony meteorite
Xmetf-ttiree percent of all meteorites that fall
on the Earth are stones.
Many freshly
ered specimens, like the one
shown
discov-
here, are
coated with dark fusion crusts. This particular

stone fell in Texas.
(From
the collection of
Ronald A. Onti)
Figure 10-10 [right]

polished)
When
meteorites are
iron mixed
i)i
cut
found
stone (cut
and
and poluhed, some
to
stony
contain tiny specks of
the rock. This specimen
was
dis-
go undiscovered and are exposed to the weather

become ahnost indistinguishable from common ter-
Meteorites with a high iron content are
much
easier to
find because they can be located with a metal detector. Consequently,
and stony-iron meteorites dominate most museum collections.

As their name suggests, stony-iron meteorites consist of roughly equal
amounts of rock and iron. Olivine conmionh is the mineral suspended
in the matrix of iron, as in the sample shown in Figure 10-11.
iron
Iron meteorites (see Figure 10-12), or irons, account for nearly 6 percent of the material that falls on the Earth. Iron meteorites may contain
from 10 to 20 percent nickel.
In 1808,
Coimt
Alois von
Widmanstatten discovered a conclusive test

common type of iron meteorite. About
for the authenticity of the most
covered in California. (From the collection of
75 percent of
Ronald A. Onli)
an octahedrite
all
is
iron meteorites are of a type called octahedrites.

cut, polished,
and
briefiy
dipped into
When
a dilute acid sola-
Iiitnlilaiiiltn^ vtliidlmuds
tioii. its uiiic]iie
trvstallinc striicUirc
is
revealed.
These
crystalline designs
are a]ipr()prialelv called Widmanstatten patterns (see Figure 10-lS).

W'idmaiislatten palleiiis coiisiiiutc conclusive prooi ol a meteorite's
authenticity liecause nickel-iron ciystals can grow to lengths of several
the molten metal cools slowly over many millions of
to 10 K per million years during
1
the time the crystals are forming. Widmanstatten patterns can thus never
l)c found in counterfeit meteorites, or "meteorwrongs" as they are hu-
centimeters only
if
years. Ociahedrites cool at rates of
morously
The
called.
existence of Widmanstatten patterns strongly suggests that
some
asteroids were parth molten for a substantial period after their forma-
Furthermore, the size of an octahedrite's parent asteroid can be csinnated b\' calculating how much rock must have insulated the molten
iion-niikel interioi to produce its long-term cooling rate. The results of
such calculations imply that typical meteorites are fragments of parent
111(11.
200 to 400 km in diameter.

The three inain types of meteorites may have come froin different
parts of a parent asteroid in the following manner. As soon as the asteroid had accreted from planetesimals 4i billion years ago, rapid decay of
short-lived radioactive isotopes heated the asteroid's interior to temperatures above the melting point of rock. Over the next few million years,
chemical differentiation occurred. As iron and other heavy elements
sank toward the asteroid's center, they displaced the lighter elements
(such as silicon) upward toward the asteroid's surface. After the asteroid
cooled and its core solidified, interasteroid collisions fragmented the parent body into meteoroids. Iron meteorites are specimens from the asteroid's core, stones being samples of its crust. Stony-irons presumably
come from intermediate regions between the asteroid's core and its crust.
It is clear that meteorites derived from the fragmentation of large asteroids were subjected to substantial processing during the first billion
years of solar-system formation. These meteoritic specimens arc therefore not representative of the primordial material from which the solar
system was originally created. In order to find primordial meteorites, we
must search for specimens that show no evidence of ha\ing been subjected to the metamorphic processes that occurred inside the asteroids.
Such specimens called carbonaceous chondrites do exist, .\bout (i
percent ot all the stones that lall on the Faith are carbonaceous chondrites. Their primordial nature is inferred from their high content of
asteroids
Figun
;,.-
,-
SlOliy-lniin
III,
mini Jul
*//^'A//v /i>< lluni
percent of all meteorites that fall on the Earth.
This particular specimen
is
a variety of stony-
irun called a pallasite. ({'rum the collection of
Ronald
,4.
Orili}
Figure 10-12
[left]
An
iron meteorite
Irons are composed almost entirely of nickeliron minerals.
The surface of a
typical iron
is
covered with thumbprmtlike depressions caused

by ablation
during
its
high-speed descent
through the atmosphere. This specimen
found in Australia. (From

Ronald A. Oritil
patterns
When ml.
a weak acid
..ululion,
uvis
the collection of
Widmanstatten
and etched
poli^lied.
u'ilh
most iron meteorites ex-
hibit interlocking crystals in designs called
Widmanstatten patterns. These patterns appear

only in the type of iron meteorite called an octahedrite. This particular nctahedrite
found in Australia. (From

Ronald A. Oriti)
was
the collection o]
liilnplant'tfiiy
vagabonds
compounds, sometimes including as much as 20 percent water.

Furthermore, carbonaceous chondrites are rich in complex organic compounds. The water and volatile chemicals would have been driven out
and the large organic molecules broken down if these meteorites had
been subjected to any significant heating.
Shortly after midnight on February 8, 1969, the night sky around
Chihuahua, Mexico, was illuminated by a brilliant blue-white light moving across the heavens. The dazzling display was witnessed by hundreds
of people, many of whom thought that the world was coming to an end.
As the light moved across the sky, it exploded in a spectacular, noisy
detonation that dropped thousands of rocks and pebbles over the terrified onlookers. Within hours, teams of scientists were on their way to
collect specimens, from what was collectively called the Allende meteorite
after the locality of the finds.
Perhaps the most significant discovery to come from the Allende meteorite was made by Gerald J. Wasserburg and his colleagues at the California Institute of Technology. They found unmistakable evidence of the
former presence of a radioactive isotope of aluminum, "''Al. This particular isotope has a half-life of only 720,000 years and rapidly decays into
a stable isotope of magnesium, "''Mg.
Detectable amounts of short-lived "'Âl could have been included in
the meteorite only if the radioactive aluminum was created shortly before the formation of the solar system. If only a few hundred million
years had elapsed, virtually all the radioactive aluminum would have
changed into magnesium before the meteorite formed. Astronomers
were therefore faced with evidence of energetic nuclear processes that
occurred in our vicinity roughly 4i billion years ago, aboiu the time the
Sun was born.
One of nature's most violent and spectacular phenomena, called a supernova explosion, occurs during the death of a massive star. As we
shall see in greater detail in Chapter 14, the doomed star blows itself
apart in a cataclysm that hurls matter outward at tremendous speeds.
During this detonation, violent collisions between nuclei produce a host
of radioactive isotopes, including ~''A1. It thus seems inescapable that a
supernova occurred very near the Sun's location 41 billion years ago. In
addition to contaminating the interstellar medium with radioactive ~''A1,
the supernova's shock wave would have coinpressed the interstellar gas
and dust, triggering the birth of the solar system.
Besides telling us aboiU the creation of the solar system, the study of
meteorites may shed light on the origin of life on Earth. Amino acids,
the building blocks of proteins on which terrestrial life is based, are
among the organic compounds found inside carbonaceous chondrites.
Perhaps interstellar organic material falling on Earth played a role in the
appearance of simple organisms on our planet nearly 4 billion years ago.
volatile
A comet
is
a dusty
as
it
chunk
of
vaporized
passes near the Sun
ice that is partly
vast majority of meteorites come from asteroids, the interplanetary particles that produce meteor showers are probably related to
comets. Astronomers who think so note that the particles that produce
meteor showers follow orbits around the Sun that are quite similar to
those of comets.
Asteroids travel around the Sun in roughly circular orbits that are
largely confined to the asteroid belt and to the plane of the ecliptic. In
Although the
sharp contrast, comets travel around the Sun along highly
elliptical orbits
angles to the ecliptic plane. The swarms of particles

that produce meteor showers also follow highly elliptical orbits (see Figinclined at
random
Intrrplam'kn's vaâbunds
Figure 10-14
and
fraginenli
conlinue
Meteoritic swarms
Ruck
dust jr urn "burned-uul" cornels
to circle the
Sun. (a) // the comet m
only recently extinct, the particles are
still
concentrated in a compact swarm. The
tightly
most spectacular meteor showers occur when
Earth happens
the
to
pass through such a
swarm, (b) Over the ages, the

ally
spread out along the
particles
gradu-
elliptical orbit.
This
configuration produces the most predictable
meteor showers because the Earth miut pass
through the e\<enh distributed swarm on each

trip
around
the
Mftforouls spread
aloiis orbit
Sun.
lire 10-14). In tact, some meteor showers are directly associated with
"burned-out" comets.
Nearly a dozen meteor showers can be seen each year. Meteor showers come from various parts of the sky, just as comets are sighted at \arious locations without any correlation with the plane of the ecliptic.
Most astronomers agree that the solid part of a comet, the nucleus, is
essentially a chunk of ices, typically measuring a few kilometers across
(see Figure 10-15). Harvard astronomer Fred L. Whipple, a pioneer in
Figure 10-15
The nucleus of Comet
The polaln-shaped nucleus of Comet
Halley
seen ui the upper
Halley
is
picture
from
lejt
corner oj
the Giotto spacecraft.
this
The nucleus
km in the
and about S km in the
shortest. A dark, circular area f km in diameter is seen on the comet's nucleus. The Sun illuminates the comet from the lower right. Two
bright jets of dust extend 15 km from the nuis
darker than coal and measures 15
longest dimension
cleus
toward the Sun. suggesting that major
emanates from the sunlit

Planck Inshlul fin ,\rriinnmie}
activity
side.
(Alax
Intf'rplatif'la)^
The structure of a comet
Figure 10-16
The
vagabonds
solid pari of
a cornel (the nucleus)
is
10 km in diameter. The coma can In

as large as 100,000 lo 1 million kilometers
across. The hydrogen envelope is typically
10 million kilometers in diameter. A comet's
roughJy
tail
can be as long as
AU, long enough
way from
reach all the
the
Earth
to the
In
Sun.
comet research, coined the description "dirty iceberg" to reflect the fact
that bits and pieces of dust and rocky material are mixed in with the
ices. Frozen ammonia, methane, and water are the primary components
of cometary ices, as indicated by their spectral lines.
As a comet approaches the Sun, solar heat begins to vaporize the ices.
The liberated gases surrounding the icy nucleus soon begin to glow, producing a fuzzy, luminous ball called the coma that can eventually expand
to a million kilometers in diameter. Continued action by the solar wind
and radiation pressure blows these luminous gases outward into a long.
Figure 10-17
Comet Kohoutek and
hydrogen envelope
These
lira
Comet Kohoulek are reproduced

scale. (Left)
The comet
iti
its
phnlngraplis nf
lo ihe
same
visible light. (Right)
This view of the comet in ultraviolet wavelengths reveals a huge hydrogen cloud sur-
rounding
tJie
University;
comet's head. (Johns
Hopkins
Naval Research Laboratory)
lulfijilant'ttn-y vaâhiiud.K
Figure 10-18
Brooks
person
who
The head of Comet

u always named after the
[leftl
cornel
first sights
Comet Brooks
after
iLi
it.
This comet,
discoverer,
the night skies
named
had an
ex-
dominated
ceptionally large, bright coma. It
during October 191
1,
(lick Ohsenmtory]
Figure 10-19 [rightl
Comet Iheya-Seki
This comet,
named
discoverers,
dominated the predawn sky during
late
Its
after
its
two Japanese
October 1965. Although
tail
was
AU
its
coma
co-
teas tin\.
long. (Lick Ohseivaton)
llowing
The
tail.
result
is
one
ol the
most awesome
sights ever visible in
the nighttime sky.
Not visible to the hinnan eye' is the hydrogen envelope, a tenuous

sphere of gas surrounding the comet's nucleus and measuring as much
as 10 million kilometers in diameter. The overall structure of a comet is
diagramed in Figure 10-16. Figure 10-17 shows two views of Comet
Kohoutek: as it appeared to Earth-based observers in 1973, and as photographed by an ultraviolet camera from a rocket. From the ultraviolet
view, astronomers fust discovered the enormous extent of the hydrogen
envelope.
Comets come in a wide range of shapes and sizes. F"or example, the
comet shown in Figure 10-18 had a large, bright coma but a short,
stubby tail. In contrast, the comet seen in Figure 10-19 had an inconspicAU. long enough
uous coma, but its tail had an astonishing length of
to reach all the way from the târth to the Sun.
It has long been known that comet tails always point awav irom the
Sun (see Figure 10-20), regardless of the direction of the (oinet's motion.
In fact, the Sun usually produces two comet tails: an ion tail ,iiid a dust
tail. Ionized atoms (that is, atoms missing one or moie elections) are
swept (liieeth away from the Sun by the solar wind. Micron-si/ed dust
|)ariick's aic blown away from (he comet's coma by radiation pressme.
1
(sometimes called a Type I tail) lan exsome dramatic stnuture that changes from night to night (see Figlue 10-21).
he more amorphous (lust tail (somciinus called a Type II
I'he relatively straight ion tail
hibit
arched.
.Asironomei s dis(o\t'r at least a dozen comets in a typical year. Some
are short-period comets, w^hich circle the Sun in less than 200 years.
Like the famous llalle\'s Comet (recall Figuie 3-13). they appear again
and again at piiclic l.ii)le inliivals. The inajoiilv of comets distoyered
mileach year aic long-period comets, howevei which take 100, 000 to
lion years to com])lele one oibit ol the Sun.
hese comets tia\el along
extremely elong.ited otbits and cc)nsec|uenll\ spend most of theii time at
tail) is typically
orbit and tail of a

wind and radiation presstite
from sunlight blow a comet's dust particles and
ionized atoms away from the Sun. C.onseifuently. the c<imrl\ tad nlieaw jiuint'. inm\
Figure 10-20
comet
jrom
the
The
The
solar
Sun.
distances' of 'lO.OOO to
way
,)().()()()
lo the nearest slai.
AC
iioiii
the
Sun .iboiil
one-llilli ol llu-
InterplancUn~s I'tigabondi
The two tails of Comet

Comet Mrkos dominated the evenitig
Figure 10-21
Mrkos
sky during August 1957. These three views,
taken at two-day intervals, show dramatic
changes in the comet's ion (Type

contrast, the slightly
1) tail.
curved dust (Type
In
II) tail
remained fuzzy and featureless.
(Palomar Obsenialory)
August 22
August 24
August 26
Because astronomers discover long-period comets at a rate of roughly

it is reasonable to suppose that there is an enormous
population of comets out there at 50,000 AU from the Sun. This reservoir of cometary nuclei surrounding the Sun is called the Oort cloud
after the Dutch astronomer )an Oort, who first proposed its existence in
the 1950s. Estimates of the numbers of "dirty icebergs" in the Oort
cloud range from 1 million to more than 100,000 million. Only if there
is such a large reserve of cometary nuclei can we understand why we see
so many long-period comets even though each one takes up to 1 million
years to travel once around its orbit.
Comets cannot survive very many perihelion passages. Eventually, a
comet's ices are completely vaporized and only a swariu of meteoritic
one per month,
A comet's nucleus will disintegrate more

quickh if it happens to pass near the Sun, as do the so-called Sungrazing comets. A comet's nucleus is sometimes even observed to
fragment (see Figure 10-22).
After a comet dies, its remaining dust and rock fragments spread out
to form a meteoritic swarm, a loose collection of debris that continues to
circle the Sun along the comet's orbit (recall Figure 10-14). If the Earth's
orbit happens to pass near or through the swarm, a meteor shower is
dust and pebbles remains.
Figure 10-22
West
The fragmentation of Comet

Sun in
Shortly ajter passing near the
1976. the nucleus of Comet West broke
four
of five photographs
shows the distintegration of the
pieces. This series
clearly
into
comet's nucleus.
(New Mexico
State
seen on the Earth.

it may seem, a total of 300 tons of extraterrestrial
estimated to fall on the Earth each day. The fluffy, lowdensity material from comets burns up in the atmosphere, and only
denser specimens related to asteroids typically reach the ground. Nevertheless, there is evidence that a comet struck the Earth in the recent
As incredible
rock and dust
past.
is
as
lnlt'i/ilaiifl(in x'fiiâbfDith
Figure 10-23
event
Aftermath of the Tunguska
In 1908, a piece
oj
a comet's
tiucletis
struck the Earth's atmosphere over the Tun-
guska region of Siberia. Trees were blown
down for many

around
kilometers in all directions
the impact
site.
(Courtesy of Sovfolo)
On June 30, 1908, a spectacular explosion occurred over the Tunguska region of Siberia. Hundreds of square kilometers of forest were
devastated (see Figure 10-23), and the blast was audible 1000 km away.
The explosion was equivalent to the detonation of a tactical nuclear warhead with the destructive power of several hundred kilotons of TNT.
The most likely explanation of this event is that a small comet (perhaps a 100-m fragment of the short-period Comet Encke) collided with
the Earth. No impact crater was formed, and the trees at "ground zero"
were left standing upright, but they were completely stripped of
branches and leaves. This phenomenon is what would be expected from
a loosely consolidated ball of cometarv ices that vaporized with explosive
force before striking the ground. The Tunguska event is a good example of the kind of devastation that can be wreaked by interplanetar\ debris.
Summary
Bode's law is a numerical sequence that gives tlic distances from the Sun
AU) oi the planets Mere ur\ iluougli I ranus. I'liis "l,iw" inspired nineteenlh-centur\ astronomers to soartli loi .i pi. mil in llie n.ip hciueen the
(in
orbits ol
Mars and Jupiter.
Thou.sands
km,
.
ol
circle the
Ix-ll
.isli-ioids. uilli (ll.nnclcrs
Sun between
tlii'
Iroin a lew kilometers to 1000
orhils nl .M.irs
and |upiui.
Gravitational perlinb.ilioris l)\ Jupiler deplele lertain oibiis uiiliin the asteroid belt. The resulting gaps, called Kirkwood .n.q)s. oiiur at simple
fractions of )upilcr's orbital period.
Jupiter's gravity also captines asteroids in two locations, called

points, along Jupiter's orbit.
Some
asteroids
mo\e
in ellipliial orhils lliat cross
Lagrange
ihe oibils ol M.ns
.iiul
Earth.
.
Small rocks
in
space arc called meteoroids.
Earth's atmosphere,
it
produces a
II
a inetcoroid enters ilie
fiery trail called a
meteor. If part
ot the
hilnjiliiiiiliiis i'/tt^fthiDuh
object survives the
fail, tiie
fragment
tliat
readies
tiie Eartli's
surface
is
called a meteorite.
Meteorites are grouped in three major classes according to their composition: iron, stony-iron, or stony meteorites.
Rare stony meteorites called carbonaceous chondrites may be relatively

unmodified material from the solar nebula. These meteorites often contain organic material and may have played a role in the origin of life on
Earth.
An
may have
asteroid
struck the Earth 63 billion years ago causing the
extinction of the dinosaius
and many other
species.
An
analysis of isotopes in certain meteorites suggests that a nearby supernova explosion triggered the formation of the solar system 45 billion years
ago.
comet
is
chunk of
ices
and rock fragments that generally moves in a

Sun at a great inclination to the plane of
highly elliptical orbit about the

the ecliptic.
As a comet approaches the Sun, its icy nucleus develops a luminous coma
surrounded by a vast hydrogen envelope. An ion tail and a dust tail extend from the comet, pushed away from the Sun by the solar wind and
radiation pressure.
Fragments of "burned-out" comets produce meteor swarms. Millions of

cometary nuclei probably exist in the Oort cloud some 50,000 AU from
the Sun.
Review questions
Why do
you suppose there are many small
asteroids, but only a few very
large asteroids?
2 Can you think of another place in the solar system in which occurs
nomenon similar the Kirkwood gaps in the asteroid belt? Explain.
3 Describe the three main classifications of meteorites.
ferent types of meteorites have originated?
4 What types of surface features would you expect

tograph of an asteroid?
5 Where on Earth might you find large
are not significantly weathered?
How
to see
on
a phe-
might these
a closeup
numbers of stony meteorites
dif-
pho-
that
6 Suppose you found a rock that you suspect might be a meteorite. Describe
some of the things vou could do to see if it were a meteorite or a
"meteorwrong."
7 Explain
8
Advanced questions
What
is
why comets are
generally brighter after passing their perihelion.
the relationship between comets and meteor showers?
9 Since there are different types of meteorites, would you expect different
types of asteroids to exist also? What sorts of observations might an Earthbased astronomer make in order to discover chemical differences between
asteroids?
Some astronomers have recently argued that passage of the solar system
through an interstellar cloud of gas could perturb the Oort cloud, causing
many comets to deviate slightly from their original orbits. What might be the
consequences for Earth?
10
littfflilfiiH-taiy I'tiiâhoiuls
Suppose il were discovered that the asteroid Hermes had been i)erturhed
such a way as to put it on a collision course with Karih. Describe what you
would do to counter such a catastrophe within the framework of present
technology.
11
in
12 From the abundance of craters on the Moon and Mercury, we know that
numerous asteroids and meteoroids struck the inner planets early in the history of the solar system. Is it reasonable to suppose that numerous comets
also pelted the planets 3 to 4 billion years ago? Speculate about the effects of
such a comelarv bombardment, especially with regard to the evolution of the
primordial atmospheres of the terrestrial planets.
For further reading
Chapman,
Falk, S.,
C. "The Nature of Asteroids." Scientific American, Jan. 1975.

and Schramm, D. "Did the Solar System Start with a Bang?" Sky
isf
Telescope,
July 1979, p. 18.
Hutchinson, R. The Search for Our Beginnings. Oxford University Press,

1983. On meteorites and the solar system.
Morrison, D. "Asteroids." Astronomy, June 1976, p. 6.
Seargent, D. Comets: Vagabonds of Space. Doubleday, 1982.
Whipple, F. "The Nature of Comets." Scientific American, Feb. 1974.
1 1
cluster of stars
Bv analyzing
starlight,
The nature
of the stars
an
astronomer can determine such details about a

star as
its
position,
clearly
called
surjaii
and
chemical com-
Irmlicynliirr.
hniiiiiiiul',
lliis
photograph
shows color dijjcreiues in a star cluster
NGC
3293. The reddish
stars are
com-
paratively cool, with surface temperatures
around 3000 K. Such
stars are also quite
luminous and have very large diameters,
100 times as
and blue-white
cally
large as the Sun.
ish
stars
The
typi-
blu-
have much higher sur-
face temperatures (15,000
to
30,000 Kl and
are roughly the same size as the Sun. (.Angln-
Australian Observatory)
Although they appear only as

to be
brilliant pinpoints of light, the stars are
much
known
our Sun. Astronomers

many of the nearer stars. In this chapter we
huge, massive spheres of glowing gas
like
have measured the distances to

also learn about imys in which astronomers determine the luminosity, surface
temperature, mass, and other properties of stars. We make our first acquaintance with the Hertzsprung Russell diagram, which reveals the fundamental
types of stars. Finally,
common
we turn
to the topic
of binary
stars, those surprisingly
systems in which two stars orbit each other. Binary stars are
an im-
portant source of information for the astronomer seeking data about the stars.
The
each appearing as a
of other stars
too faint to be seen with the naked eye, but every star appears only as a
bright point of light. A star is a huge, massive ball of hot gas like our
Sun. held together by its own gravity. We now know that some stars are
larger than our Sun, some smaller, some brighter, some dimmer, some
night sky
is
spangled with thousands of
bright pinpoint of light. Telescopes reveal
stars,
many thousands
hotter, others cooler.
The
quest for information about the mass, luminosity, surface temperstars has been a major occupation of twentieth-century astronomers. A remarkably coinplete picture has emerged in recent times. By understanding the stars, we gain
ature,
and chemical composition of different
insight into
our relationship
scope of space and time.
to the universe
and our place
in the
cosmic
The nature of
Distances to nearby stars are

determined by parallax
the stars
Astioiioniers speciJy the briglitness ol a star by its magnitude according

scheme that dates back to the ancient Greek astronomer liipparchus. The brightest stars Hipparchus saw in the sky he called
to a
first-
magnitude
The dimmest stars he called sixth-magnitude. To stars

ot intermediate brightness he assigned numbers
on a scale of 1 to 6.
In the nineteenth century, techniques were developed for measuring
the amount of light arri\ ing irom a star, and then astronomers
defined
the magnitude scale more precisclv. .Measurements showed
that a firstmagnitude star is about 100 times as bright as a sixth-magnitude star. In
other words, it would take 100 stars of magnitude +(i to "provide as
much light energy as we receive from a single star ol magnitude -i-I.
The magnitude scale was redefined so that a magnitude difference of 5
stars.
corresponds exactly
light energy,
of 100 in the amount of light received. .A

therefore corresponds to a factor of 2.512 in
to a factor
magnitude difference of
because
2.512 X 2.512 X 2.512 x 2.512 x 2.512
For example,
light as
we
it
= 100
takes about 2i third-magnitude stars to provide as

from a single secoiul-magnituile star.
much
receive
Astronomers
extended the magnitude scale to describe the dimmer stars visible through their telescopes. For example, with a good pair
ot binoculars, you can see stars as faint as tenth magnitude.
Through
also
some of the largest telescopes in the world, it is possible to see stars as

dim as magnitude 20. Photography with long exposure times reveals
even dimmer stars.
.Astronomers use negadve numbers to extend the magnitude scale so
that
includes verv bright objects. For example. Sirius the brightest

skv has a magnitude of -li. At its brightest, \enus shines
it
star in the
with a magnitude of -4.

sky. Its
N'enus
Sirius (brightesi star)
(at brightesi)
stars.
Naked eye
that just
limit
I'luto
Sun
is
the brightest object in the
dim in the skv might actuallv be a brilliant star

be extremelv far awa\ In order to determine the
star that looks
happens
to
we must
know how far awav the star is.

measuring stellar distances involves
first
lie most straightforward wa\ of

parallax, which is the apparent displacement of an observed object
caused bv a change in the point from which it is viewed. Parallax occurs
when nearbv objects appear to shift their positions against a distant back1
ground
i.arge telescope
stars.
isual limit)
The
The apparent magnitude
Astronomers dennlr ihr hn^hliiess of

objects in the sk\ by then ap/Mirent magnitiule.
stars visible to the
2).
distance to a star can be determined bv measuring the star's parstar (p) e(|uals one-half of the angle through
which a stars position shifts as tlie Karth moves from one side of its own
allax.
scale
Most
course, the
(photographic limit)
Figure ll-l
Of
-26i
as the observer moves Irom one place to another (see Figure 1 1Stars also exhibit parallax. .As the Karth orbits the Sun, nearbv stars
appear to move back and forth against the background of more distant
large telescope
IF
is
brightness of a star,
at tual
Binocular limit
magnitude
Figure 11-1 illustrates the modern magnitude scale, fhese magnitudes

are pioperly called apparent magnitudes because thev describe howbright an object appears to an tai th-liased observer. Apparent magnitude is a measure of the energy arriving at the Earth.
Apparent magnitudes do not tell us about the actual iMightncss of the
naked eye hare mag-
nitudes in the range + 1 to +6. Photography

through large telescopes can reveal stars as
faint as magnitude +24.
The parallax of a
orbit to the other (see Figure
1 1-;i).
Ihc smaller the angle (/)), the
greater the distance (d) to the star. Astronomers usually denote the
distances to stars in light \ears or parsecs (1 parsec = 3.26 light \eais).
A simple equation relates the parallax of a star to its distance from
Earth. If the angle /; is measured in arc sec then ihe disiaiue d to the
star in parsecs is given bv
,
The
Figure 11-2
Imagnte
Parallax
[left]
ing at some nearby object
tike
it
look-
tree seen
against a distant background siuh as

tains.
If you move from one location
mounto
an-
other, the nearln object will apl)e(ir to shift
location with respect to the distant

scenery. This familiar
iKiliirr oj the star.
phenomenon
its
background
is
called
parallax.
Figure 11-3 [right]

As
Stellar parallax
Sun. a nearby
the Earth mints the
appears
to shift
ground of
its
position
distant stars.
allax of the star)
is
agaunt
star
the back-
The angle p (the parto the angular size
equal
of the radius of the Earth's orbit as seen from

the star.
The smaller
the parallax (p), the
larger the distance (d)
to the star.
P
For example, a star whose parallax is 0.5 arc sec is 2 parsecs from Earth.
The parallax method of determining stellar distances works only for
nearby stars. There are slightly more than 1000 stars within 20 parsecs
of the Earth whose parallaxes have been measured with a high degree of
precision. Most of these nearby stars are invisible to the unaided eye.
The majority of the familiar, bright stars in the nighttime sky are too far
away to exhibit measurable parallax as the Earth orbits the Sun.
star's luminosity can be

determined from its apparent
magnitude and its distance
from Earth
The
absolute magnitude of a star is the apparent magnitude it would

have if it were located at a distance of exactly 10 parsecs from the Earth.
For example, if the Sun were moved to a distance of 10 parsecs from
the Earth, it would have an apparent magnitude of +4.8. Therefore the
absolute magnitude of the Sun is +4.8. The absolute magnitudes of
other stars range from roughly -10 for the brightest to +15 for the
dimmest. The Sun's absolute magnitude is about in the middle of this
range, giving us our first hint that the Sun is an average star.
How do astronomers determine absolute magnitudes of stars? The
farther away a source of light is, the dimmer it appears. The inverse
square law tells us that the apparent brightness of a light source is inversely proportional to the square of the distance between the source
and the observer. For example, double the distance to a light source and
its apparent brightness decreases by \, triple the distance and the brightness decreases by g. Using this law, astronomers have derived a mathematical equation relating three quantities: a star's apparent magnitude
(m), its absolute magnitude (M), and its distance (d) from the Earth. If
you know any two of these quantities (such as apparent magnitude and
The tmlure
oj ihe stars
205
you can calculate the ihird one (such as ahsoluie magnitude).

the apparent magnitude ol a nearby star,
find its distance by measuring its parallax, then calculate its absolute
magnitude.
Absolute magnitude is directly related to luminosity, the amount of
energy escaping from a star's surface each second (usually given in ergs
distance),
Thus astronomers measine
per second). Many scientists prefer to speak of a star's luminosity rather

than its absolute magnitude because luminosity is a measure of the star's
energy output. There is a simple, direct relationship between absolute
magnitude and luminosity, and astronomers can convert from one to the
other as they see fit.
For convenience, stellar luminosities are expressed in multiples of the
Sun's luminosity (L). which is 3.90 x 10-" erg/sec. The brightest stars in
the sky (absolute magnitude = -10) have luminosities of 10" L-,. In
other words, each of these stars has the energy output of a million Suns.
The dimmest stars (absolute magnitude = +15) have luminosities of
10-^ L.
star's color reveals

surface temperature
Figure 11-4
One
its
Temperature and color
Tins
schematic diugtam shim's the relatwmhip be-
tween the color of a star and

perature.
The
its
is
plotted against wave-
lengths (compare Figure 5-3).

visible
wavelengths
which a
is
The range of
The way in
indicated.
star's intensity
curve
mines the dnminnnt color of
surface tem-
intensity of light emitted by three
hypothetical stars
skewed deter-
is
its
when comparing stars in the nighttime

apparent magnitude. More careful examination, even with the naked eye, reveals that stars also have different colors. For example, in the constellation of Orion, you can easily note the
difference between reddish Betelgeuse and bluish Rigel (see Figure 2-2).
A star's color is directly associated with its surface temperature
through relationships such as Wien's law discussed in Chapter 5. The
intensity of light from a cool star peaks at long wa\elengths. so the star
looks red (see Figure 1 1-4). A hot star's intensity curve is skewed toward short wavelengths, making the star look blue (see Figure ll-4f).
The maximum intensity of a star of intermediate temperature (such as
the Sun) occurs near the middle of the visible spectrum, giving the star a
yellow-white color (see Figure ll-4h).
To measure the colors of the stai s accurately, astronoiners have de\cloped a technique called photoelectric photometry. This technitjue uses a
light-measuring device at the focus of a telescope, with a standardized
set of coloted filters. By far, the most commonly used filters are the
three UBV filters. Each filter is transparent in one of three broad wavesky
visible light.
of the
is
first
things you notice
their differences in
length bands: the ultraviolet (U), the blue (B), and the central region (V)
visible spectrum (see Figure
1-5). The transparency of the V filter fairly accuratelv mimics the sensiti\it\ of the human c\e.
of the
This star looks red
Lr\
")ll(l(l
lll.dllll
Wavrlrni;(h(A)
'i,
011(1
l!ll, 11(1(1
')(I(H1
HI, (Hid
V\'avrl<-ni.lll
I"). (111(1
.'(I.
(KM
r>, (11)11
1(1,11(1(1
\V,i\c-lrni;lll
A)-
L!i).
110(1
Tilt'
4000
3000
5000
(
Figure 11-5
To do photometry, the astronomer aims a telescope at a star and

measures the intensity of starlight that passes through each of the filters.
This procedure gives three apparent magnitudes for the star, usually
designated by the capital letters U, B, and V. The astronomer then compares the intensity of starlight in neighboring wavelength bands by subtracting one magnitude from another to form the combinations (U B)
and (B V), which are called the star's color indices. The UBV magnitudes and color indices for several representati\e stars are given in Table
6110(1
Wa\ elength A
nature uj the stars
11-1.
Light transmission through
the JJBV filters
This graph shows the wave-
length ranges over which standardized L. B.
and V fitters are transparent
U filter
is
4000 A,
because
transparent
The
to tight.
to tight
front
3000
TABLE
just beyond the violet
it lies
The B
visible spectrum.
from about 3800
to
from about 5000
to
filter
is
end of
the
The
UBV
magnitudes and color indices of selected stars
(B-
Regulus (a Leo)
1.36
1.25
0.89
-0.11
-0.36
(a Aql)
0.77
0.99
1.07
+0.22
+0.08
Bellatrix {y Ori)
-0.87
Star
name
V)
(U-B)
the
transparent
5500 A, and
6500 A.
11-1
to
a range called the near-uttraviotel
V filter
.Mtair
1.64
1.41
0.54
-0.23
Aihena
(y
Gem)
1.93
1.93
1.96
0.00
+0.03
30.000
Megrez
(5
UMa)
3.31
3.39
3.46
+ 0.08
+ 0.07
27,000
Elnath
(P Tau)
1.65
1.52
1.03
-0.13
-0.49
24.000
21 .(100
18.000 -\
15,000
s
12.000
900(1
\^^ Sun
6000
3000
III
-0
+0.5
3 0.0
+1.0
+1.5
+2.
B - V) Color Index
Figure 11-6
color index
Blackbody temperature versus

The (B - V) color inde.x is tlie
difference between the
star.
If the star
10,000 K,
(B
it is
V) index
is
B and V
magnitudes of
hotter than about
a very bluish star and
is less
cooler than about
than
zero.
10,000 K,
its
If a star
its
(B
is
V) index
The Sun's (B - V) index
is
greater than zero.
is
about 0.62, which corresponds
to
a tempera-
ture of
5800 K.
and V
magnitudes, an astronomer can estimate
After measuring a
the star's suiface temperature
such as
this one.
star's
spectrum
is
a clue to the star's
surface temperature
star's
from a graph
A color index tells you how much brighter or dimmer a star is in one
wavelength band than in another. For example, the (B - V) color index
tells you how much brighter or dimmer a star appears through the B filter than it does through the \' filler.
A color index is important because it tells you about the star's surface
temperature. If a star is very hot, its radiation is skewed toward the
short-wavelength ultraviolet, which makes the star bright through the
U filter, dimmer through the B filter, and dimmest through the V filter.
The star Regulus (see Table 11-1) is an example of this case. If the star
is cool, its radiation peaks at long wavelengths, making the star brightest
through the V filter, dimmer through the B filter, and dimmest through
the U filter. The star Altair is an example.
The graph in Figure 11-6 gives the relationship between the (B - V)
color index and temperature. If you know a star's (B V) color index,
you can use this graph to find the star's surface temperature. For example, the Sun's (B - \') index is 4-0.63, which corresponds to a surface
temperature of 5800 K.
New filter systems have been devised in recent years. Nevertheless,
the UBV filters are still one of the most commonly used set of filters in
modern astronomy.
The
field of stellar spectroscopy was born in the 1860s wiien the Italian
astronomer Angelo Secchi attached a spectroscope to his telescope and
pointed it toward the stars. Secchi observed stellar spectral lines and
made the important discovery that stars can be classified into various
spectral types according to the appearances of their spectra. In those
days, the nature and cause of spectral lines were not well understood.
Astronomers classified each star by assigning it a letter from A through
The nature
oj the stars
207
depending on the strength
P,
of tlie
hvdrogen Bahiier
lines in the star's
spectrum.
When
Niels
Bohr explained the
structure of the hydrogen
atom
in the
early lyOOs (recall Figure 5-11), astronomers realized that the strength of
the lines in a star's spectrum is directly related to the temperature of the
gases in the star's outer layers. Hydrogen is by far the most abundant
element in the universe, accounting for about three-quarters of the mass
of a typical star. Hydrogen lines do not necessarily show up in a star's
spectrum, however. If the star is much hotter than 10,000 K, highenergy photons pouring out of the star's interior easily knock electrons
out of the hydrogen atoms in the star's outer layers, ionizing the gas.
Hydrogen ions have no electrons in their lower energy levels to absorb
photons and produce Balmer lines. Conversely, if the star is much cooler
than 10,000 K, the majority of photons escaping from the star do not
possess enough energv to boost manv electrons up from the ground
state of the hydrogen atoms. These unexcited atoms also fail to produce
Balmer lines. In short, the star must be hot enough to excite the electrons out of the ground state
but not so hot that the atoms are ionized.
A stellar surface temperature of 10,000 K results in the strcmgest Balmer
lines.
prominent
of Balmer lines is a clear indication that a star's surabout 10,000 K. At other temperatures, the spectral
lines of other elements dominate a star's spectrum. For example, around
25,000 K the spectral lines of helium are strong because, at this temperaset
face temperature
ture,
is
photons have enough energy
to excite
helium atoms without
tear-
ing away the electrons altogether.

When a h\clrogen atom is ionized,
its only electron is torn awav and

no absorption lines can be produced. However, an atom of a heavier element has two or more electrons. When one electron is knocked away,
the remaining electrons can take over to produce a new and distinctive
For example, in stars hotter than about 30,000 K

in a helium atom is torn away. The remaining
electron produces a set of spectral lines that is recognizably different
from the lines produced by un-ionized helium. When the spectral lines
of singly ionized helium appear in a star's spectrum, we know that the
star has a surface temperature greater than 30,000 K.
.Astronomers designate an un-ionized atom with a Roman numeral I:
thus H I is neutral hydrogen. A Roman numeral II is used to identifv
an atom with one electron missing, so He II is singlv ionized helium
(He*). Similarly, Si III is doubly ionized silicon (Si**), whose atoms are
each missing two electrons.
In the early 1900s, astronomers at Harvard Observatorv set up the
spectral classification scheme we use today. Many of Secchi's A-through-P
categories were dropped, and the remaining spectral classes were reordered into the temperature sequence OBAFGKM. This sequence has traditionally been memorized as "Oh. Be A Fine Girl, Kiss Me!" The O
stars are the hottest: their surface temperatures arc in excess of
35.000 K. and their spectra show He II and Si 1\'. Ihe M stars are the
coolest; their surface temperatuies of around 3000 K are so cool that
atoms can stick together in molecules such as titanium oxide, whose
spectral lines arc prominent. Figure 11-7 shows representative examples
of each spectral type.
Astronomers have foinid it useful to subdivide the original
OB.AFCIKM sequence further. These finer steps are indicated by the
addition of an integer from
through 9. Thus, for example, we have
F8. F9. GO, Gl, C;2
G9, KO. Kl, K2
The Sun, whose
of spectral
set
lines.
one of the two electrons
The nature of
Figure 11-7
spectra
spectrum
is
seen in the mid-
stars,
OB
The hydrogen
dle of each of these seven strips.

tines are strongest in
'
Type
Principal types of stellar
star's
Star
""^
lines
and
stars exhibit
to
6000 K. The
in the spectrum of
tanium
an
oxide, which
perature
is
can
Aurigae
Cygni
broad, dark bands
M star are caused by
ti-
^^^^^1
tem-
exist only if the
cooler than about
numerous
caused by metals, indicating temperatures
from 4000
\ Cephei
which have sur-
face temperatures of about 10,000 K. The

spectra of
the stars
P Cassiopeia
3000 K.
(Palomar Obsen'ntoiy)
=HH
spectrum
Ca
II)
is
The
tion
is
dominated by
G2
singly ionized metals (especially Fe II
T)
Pegasi
"y
Draconis
a Herculis
and
star.
strength of a particular spectral line depends on both the ionizaexcitation of the atom responsible for that line. Ionization
and the
and excitation of atoms in a gas depend on the temperature of the gas,

as shown in Figure 11-8, allowing us to deduce a star's surface temperature from its spectral type. For example, a star exhibiting strong Ca II
and Fe I lines in its spectrum is a K5 star with a surface temperature
around 4500 K.
Temperature (K)
8onn
6000
1
Spectral type
temperature
The strengths of
lines of
the absorption
various elements are directly related
the temperature of the star's outer layers.
to
He
II
For
and calcium (Fe II and

Ca II), corresponding to a spectral class of G2
and a surface temperature of about 5800 K.
Mi;
singly ionized iron
stars,
tines
are strongest in
whereas stars cooler than about
3500
that there are different

kinds of stars
y/si
^
y^
T.O
Ca>
>< %
11
IIl\
Fll
shoxo absorption caused by titanium oxide.
The Hertzsprung- Russell

diagram demonstrates
Call
H^/^
example, the Sun's spectrum has strong tines of
Note that hydrogen
and
Figure 11-8
Fe
GO
Spectral type
The
first accurate measurements of stellar parallax were made in the

mid-nineteenth century, at about the time when astronomers began observing stellar spectra. Observing techniques improved during the next
half-century, and the spectral types and absolute magnitudes of many
became known.
Around 1905, the Danish astronomer Ejnar Hertzsprung pointed out
that a regular pattern appears when the absolute magnitudes of stars are
stars
The nature of
the stars
plotted against their color indices on a graph. Almost a decade later, the
American astronomer Henrv Norris Russell independently discovered
this regularity in a graph using spectial t\pes instead ot color indices.
Plots of this kind are now known as Hertzsprung Russell diagrams, or
H-R
diagrams.
Figure 1-9 is a typical Hert/sprung-Russell diagram. Each dot represents a star whose absolute magnitude and spectral type have been determined. Bright stars are near the top of the diagram, dim stars near the
bottom. Hot stars (O and B) are toward the left side ol the graph, cool
1
stars
(M) toward the
The most
right.
H-R diagram is that the data points

over the graph but are grouped in several
striking feature of an
are not scattered
randomh
all
The band stretching diagonally across the H-R diagram

represents the majoritN of stars we see in the nighttime sky. This band,
called the main sequence, extends from the hot, bright, bluish stars in
distinct regions.
Supergiants
^.
\.
Figure 11-9
diagram
A Hertzsprung-Russett
An ll-R diatpum
is
a graph on
which the absolute magnitudes of
stars are
plotted against their spectral types.
on
this
diagram represents a
Each dot
White dwarfs
star in the sky
whose absolute magnitude and spectral class
have been determined.
\'ote that the data
points are grouped in specific regions on the
graph.
Tilts
pattern reveals the existence of
different types of stars in
sequence
stars,
while dwarfs.
tlie
sky:
mam-
red giants, supergiants.
1-
and
Spectral class
c;
The nature uf
down
the upper-left corner of the diagram
to the cool,
the stars
dim, reddish
whose properties place it in this

region of the H-R diagram is called a main-sequence star. For example,
the Sun (spectral type G2, absolute magnitude +5) is such a star.
stars in the lower-right corner.
star
Toward the upper-right side of the H-R diagram, there is a second

major grouping of data points. Stars represented bv these points are
both bright and cool. From Stefan's law, we know that a cool object radiates much less light per unit of surface area than a hot object does. In
order to be so bright, these stars must be huge. They are typically 10 to
100 times as large as the Sun and have surface temperatures around
3000 to 4000 K. They are called giants, or red giants, because they appear reddish in the nighttime sky. Aldebaran in the constellation Taurus
and Arcturus in Bootes are examples of red giants that you can easily
see with the naked eye.
A few rare stars are considerably bigger and brighter than typical red
giants. These superluminous stars are appropriately called supergiants.
Betelgeuse in Orion and Antares in Scorpius are examples of supergiants visible in the nighttime sky.
grouping of data points toward the

Hertzsprung Russell diagram. These stars are
hot, dim, and small. They are appropriately called white dwarfs. These
stars, which are roughly the same size as the Earth, can be seen only
Finally, there
is
a third distinct
lower-left corner of the
with the aid of a telescope.

There is another useful way to exhibit an H-R diagram. Instead of
absolute magnitude, luminosity is plotted on the vertical axis of the
H-R
Figure 11-10
An
HR
stellar luminosities are
diagram,
diagram
On
this
graphed
against the surface temperatures of stars. The

locations of
vanous
cated, as
the point representing the Sun.
The
size
is
of a star
and
draw
is
types of stars are indi-
related to
its
surface tem-
perature
luminosity. This relationship
used
the
to
dashed
lines
on
this
indicating stellar radii. The Sun's size
way between
the largest
see in the sky.
and
was
graph
is
mid-
smallest stars
we
20.000
411.000
*
10.000
5000
l\'mporalure (K)
2500
The nature of
lite
ilars
graph, and instead of spectral class, the surface temperature is plotted

on the horizontal axis. The resulting graph is still an H-R diagram, but
the observational quantities (such as spectral t\pe) have been replaced by
calculated cjuantities (such as temperature in kelvins).
Figure 1-10 shows this type of H-R diagrain. Note that the temperature scale on the horizontal axis of the graph increases toward the lell,
because Hertzsprung and Russell drew their original diagrams with O
stars on the left and M stars on the right. (They made this choice because the O stars have the simplest spectra.) Having hot stars toward the
left and cool ones toward the right is a convention that no one has seri1
ously tried to change.
Also shown
Figure 1 1-1(1 are dashed lines that display the radii of

stands for 1 solar radius (that is, half the Sun's
diameter) and is ecjual to 6.96 x 10" km, or roughly half a million miles.
Notice that most red giants are 10 to 100 times as large as the Sun,
whereas white dwarfs are only about too the size of the Sun. Also notice
that most main-sequence stars are roughly the same size as the Sun.
The stars are classified into spectral types on the basis of the most
prominent lines in their spectra. There are minor differences, however,
stars.
Tcrnpci-alutv (R)
Figure 11-11
cotwenieni
to
Luminosity classes
divide the
HR
11
diagram
i.\
into
regions corresponding to luminosity classes.
This breakdown permits finer distinctions be-
tween giants
class
and
supergianls. Luminosity
encompasses the mam-sequence
including the dim red stars called red
dwarfs toward the lower-right side of

the
HR
diagram.
stars,
in
The symbol R;
the spectral patterns of stars having the same spectral tvpe. A

system of luminosity classes was developed in the 1930s, based upon
these minor differences in the patterns of spectral lines. When the luminosity classes are plotted on the H-R diagram (sec Figure 1 1-1 1). they
provide a useful subdivision of the star types in the upper-right half of
the diagram. Luminosity class I includes all the supergiants, and luminosity class X includes the main-sequence stars. The intermediate classes
distinguish giants of various luininosities as indicated in Table 1 1-2.
Astronomers commonly describe a star by combining its spectral type
and its luminosity class into a sort of shorthand description, for example,
calling the Sun a G2V star. This notation supplies a great deal of inlormation about the star. Its spectral type is correlated with its surface temperature, and its luminosity class is correlated with its luminosity. Thus
an astronomer knows immcdiateh that a G2\' star is a main-sequence
star with a linninosit\ of aboiu 1 L- and a surface temperatine of nearly
6000 K. Similarlv. a description of Aldebaran as a K.tIII star lells an astronomer that it is a red giant with a luminositv of aroimd 500 L and a
surface temperature of about 4000 K.
An awareness of fundamentally different types of stars is the first
important lesson to come from the H-R diagram. .\s we shall see in the
following chapters, the different kinds of stars represent different stages
of stellar evolution. We shall learn that a true appreciation of the H-R
diagram requires an uncierstanding of the life cycles of stars: how they
are born, what happens as they mature, and where they go when they
among
die.
TABLE
11-2
Stellar luminosity classes
Luminosity
class
Types of
Supergiants
stars
11
Bright giants
111
Giants
IV
Subgiants
Main sequence
Tlir niihirv
We now know
Binary stars provide

information about
sizes,
lite
stars
temperatures, and luminosities
To complete our picture of the physical properties of stars, we

need only to know their masses. There is, however, no practical and di-
of
masses
stellar
something about the
iij
stars.
rect
way
to
measure the mass of an
isolated star observed in the sky.
Fortunately for astronomers, about half of the visible stars in the

nighttime sky are not isolated individual ones. Instead, they are revealed
by the telescope to be multiple-star systems in which two or more stars
orbit about each other.
By observing
exactly
how
these stars orbit each
other, astronomers can glean important information about the stellar
The binary star Kruger 60
Figure 11-12
About one-half of
stars.
the visible stars are double
This series of photographs shows the
nary star Kruger
60
bi-
Cepheus. The orbital motion of the two stars
about each other
tem
lias
is
apparent. This binary
a period of 44i
years.
sys-
The maximum
angular separation of the stars is about 2i arc

sec; their apparent magnitudes are +9.8 and
+ 11.4.
masses.
In cases where astronomers see the two stars actually orbiting each
other, they are called a visual binary (see Figure 11-12). After
years of patient observation, astronomers can plot the orbit of
about another
in a binary pair (see
The gravitational force between the two stars in a binary star system
keeps them in orbit about each other. This means that details of their
orbital motions can be described using the laws of Newtonian mechanics.
Specifically, Kepler's third law relates the masses of the stars in a binary
1920
1830
1835
..
The
1840
70 Ophiuchi
bmaiy star
astronomers can draw the orbit
orbit of
After plotting the observations of a
over the years,
of one star with respect

star
and
may
size
case.)
to the other.
be regarded as stationary
the orbit
law can be used
to
is
known, Kepler's third
deduce information about
the masses of the stars. This illustration
shows the
orbit
of a faint double star in the
constellation Ophiuchus.
1880
the shape
of the orbit will be the same in either
Once
(Either
star
Figure 11-13).
1908
Figure 11-13
many
one
1885
Period
88 years
Till'
iiiiliiii-
Ike Stan
(if
lo their orbital
form
useful
where
A/,
masses,
size of the orbit.
Newton proved
that a
is
-.". =
Af,
period and the
of this law
and
is
M-, are the masses of the two stars expressed in solar

the orbital period in years, and a is the semimajor axis
(measured in AU) of the elliptical orbit of one star about the other.
I hus, an astronomer can combine data concerning the
orbit of a binary
with Kepler's third law to calculate the sum of the masses of the two
stars.
Each of the
Figure 11-14
Each
star
Star orbits in a binary
a binary system follows an
cal urhit about their
The center of mass

massive of the two
common
is
stars in the binary
system actually moves in an elliptical

mass of the system (see Figure 1-14). This
concept is analogous to two children on a seesaw. In order for the seesaw to balance properly, the center of mass of the two-child system must
be located just above the support point or fulcrum. As you no doubt
know from experience, this center of mass is offset from the midpoint
orbit about the center of
ellipti-
center of mass.
always nearer the more
stars.
between the two children toward the heavier child.

Similarly, the center of mass of a binary system could conceivablv be
determined by placing the two stars at either end of a huge seesaw and
determining where the fulcrum must be placed to balance the seesaw.
The center of mass is always offset toward the more massive star.
In practice, the center of mass of a visual binary is detennined b\
using the background stars as reference points. The separate orbits of
the two stars can then be plotted as in Figure 1 1-14. The center of mass
is located by finding the common focus of the two elliptical orbits.
This
information is needed to calculate the individual masses of the two stars.
Years of careful, patient observation of binaries have vielded the
masses of many stars. As the data accuinulated, an important trend
began to emerge. For main-sequence stars, there is a direct correlation
between mass and luminosity: the more massive the star, the more luminous it is. This mass-luminosity relation can be convenientlv displayed
as a graph (see Figure 11-15). Note that the range of stellar masses extends from about one-twentieth of a solar mass to about 50 solar masses.
The Sun's mass lies in the middle of this range, so we see again that our
star
ordinary and
is
typical.
The mass-luminosity
the
0.05 0.1
0.2.') O.,")
MiLSS
Figure 11-15
relation
")
2'
HI
The mass-tuminosity
Fur mani-\eijiienre
sluts, there
lined correlation beliveen mass anil

the
',
(.1/,.)
mure massive a
star, the
is
liiminiisit\:
mure lumniiins
H-R
diagram
of a star
affects the location of

its
spectral lines
relation demonstrates that the
a progression in
mass as well as
main secjueiue on
and
in lumiiiositv
surface temperature. The hot, bright, bluish stars in the upper-left corner of the H-R diagram (see Figure i 1-9) are the most massive mainsequence stars in the sky. The dim, cool, reddish stars in the lower-left
corner of the H-R diagram are the least massive. The main-sequence
stars of intermediate temperature and luminosiiy have intermecliate
mass. This relationship of mass to the main sequence will play an important role in
The motion
is
our
later discussion
of
stellar evolution.
Many
binary stars are scattered throughout our galaxy, but onlv those
nearby or that have a large separation between the two stars can
be disiinguished as \ isual binaries. Star images in a remote binary are
often blended together to produce a visual image that looks like a single
that are
star.
Spectroscopy piovides evidence that some single-appearing stars are
in fact binaries.
The
lines.
spectral analysis of some stars may yield incongruous spectral

For example, the spectrum of what appears at first to be a single
rill'
iHilutc
iij
the slais
may
Star
star)
include both strong hydrogen lines (characteristic of a type-A

and strong absorption bands of titanium oxide (indicating a type-M
star).
single star could not have the differing physical properties of
these two spectral types, so

binar\ system.
we must conclude
binary star that
is
that this star
detected in this way
is
is
actually a
called a spec-
trum binary.
Spectroscopy also provides important information about the movestars because the wavelength of light is itself affected by motion. The effect of motion on wavelength was first described by the nineteenth-century mathematician Christian Doppler and is therefore called
ments of
the
Doppler
effect.
The Doppler
effect
is
shown schematically
in
Figure 11-16, where the
represent the peaks of waves emitted as the light source moves

along. Notice that light waves from an approaching light source are compressed. Because each successive wave is emitted at a position slightly
closer to you, you see a shorter wavelength than you would if the source
circles
stationary. All the spectral lines in the spectrum of an approaching

source are shifted toward the short-wavelength (blue) end of the spectrum. This phenomenon is called blueshift.
Conversely, light waves from a receding source are stretched and you
see a longer wavelength than you would if the source were stationary.
All the spectral lines in the spectrum of a receding source are shifted
toward the long-wavelength (red) end of the spectrum, producing red-
were
shift.
According to the Doppler effect, the size of the shift of a spectral line
proportional to the speed with which a source of light is moving toward or away from you. The greater the speed, the greater the shift.
The Doppler effect can be expressed in the form of an equation. Suppose that An is the wavelength of a particular spectral line from a source
of light that is not moving. If the source is moving, this particular line is
shifted to a different wavelength A. The size of the wavelength shift is
usually written as AA where AA = A A(|. Christian Doppler proved that
is
Figure 11-16
The Doppler
effect
Liglil
wavelengths are affected by motion between a

light source
and an
observer. Wavelengths
appear shortened (blueshifled)
moving toward
if the
source
is
the observer. Wavelengtlis ap-
pear lengthened (redshifted)
if the source
is
receding from the observer. Motion perpendic-
ular
to
an
observer's tine-of-sight does not
affect wavelength.
AA _
I'
A|)
The nature of
the stars
where i' is the speed of the source measured along the line of sight
between the source and the observer, and c is the speed of light
(3 X 10 '"cm/sec).
A speed determined in this fashion is tailed a radial velocity because
the Doppler effect measures onlv the component of a star's motion parallel to our line of sight, or along the "radius" drawn from the Earth to
the star. Of course, a sizable portion of a star's motion may be perpendicular to our line of sight. This transverse movement, called proper
motion, does not affect wavelengths.
It the orbital speeds of the two stars in a binary are more than a tew
kilometers per second, the Doppler effect can be used to calculate important information about the binar\ even though two separate stars
mav not actualU be observed. Such binaries yield a spectrum in which
two complete sets ot spectral lines shift back and torth. They are called
spectroscopic binary stars. The regular, periodic shifting of the spectral
lines is caused by the orbital motions of the stars as they revolve about
their center of mass.
Figure 11-17 shows two spectra of a spectroscopic binary taken a few
davs apart. In Figure ll-17a, two sets of spectral lines are visible, slightly
offset in opposite directions from the normal positions of these lines.
The star moving toward the Earth has its lines blueshifted; the other star
(moving awav from the Earth) has its lines redshifted. .A few days later,
the stars have progressed along their orbits so that one star is moving
toward the left and the other toward the right. At this point, neither star
has any motion toward or away from the Earth, so there is no Doppler
shifting and both stars yield spectral lines at the same positions. Thus
onlv one set of spectral lines appears in Figure ll-17i.
Significant information about the orbital velocities of the stars in a
spectroscopic binarv can be deduced from measuring shifts in spectral
lines. This information is best displayed as a radial velocity curve
graphing radial velocitv versus time for the binary system (see Figure 118). Radial velocitv is the portion of a star's motion that is directed parallel to the line of sight between the Earth and the star.
In Figure 11-18. note that the waw pattern repeats with a period of
about 15 days, which is the orbital period of the binary. Also note that
.
the entire wavy pattern is displaced upward trom the zero-velocity line
by about 12 km/sec, which is the overall motion of the binary system
aw-ay from the Earth. Superimposed on this overall recessional motion
(an overall redshift of the spectra lines) are the periodic approaches
recessions of the two stars as they orbit about the center of mass.
and
spectroscopic binaries, one of the stars is so dim that its speccannot be detected. The fact that the star is a binar\ is ol)\ ious.
however, because its spectrum shows a single set ot spectral lines ih.it
shift regularh back and forth. Such a single-line spectroscopic binary
viclds less infoi iiiation about its two stars than does a double-line spectroscopic binary like that shown in Figure 11-17.
In
many
tral lines
Figure
back
1-17
spertrnscoptc
shift
spectroscopic binary
hmars
and forth
exhibits spectral lines that
as the two stars rei'olve
about each other. These
turn spectra
show
the
behavior of the spectroscopic binary k Arietis.

(a)
The
stars are
moving
parallel to the line of
sight (one star
approaching Earth, the other
star receding),
producing two
sets oj shifted
spectral Itnei. (b) Both stars are
pendicular
to
our line of
(Lick Obsen'ator\l
sight.
moving
per-
rhr
Stage
Slagc 2
Center
ol*
iHiliiir
1,1
the skir
Stage 4
Stage 3
mass
Figure 11-18
radial velocity curve
Tin-
graph
displays the radial velocity curve uj the
binary
HD
171978. The drawings
the positions of the stars

at
four
selected
and
indicate
their spectra
moments during an
orbital period.
" Time (days)
The orbital speeds of the two stars in a binary are related to the
masses of the stars by Kepler's laws and Newtonian mechanics. However,
the individual masses of the two stars can be determined only if the tilt
oi their orbits is known. The angle of the orbits determines how much
of the true orbital speeds of the stars appears as radial velocity measured
from the Earth.
If the two stars are observed to eclipse each other, their orbits must
be nearly edge-on as viewed from the Earth. As we shall see next, individual stellar masses can be determined if a spectroscopic binary also
happens to be an eclipsing binary star.
Light curves of eclipsing
binaries provide detailed
periodically eclipse each other as seen
information about the stars
binary systems are oriented so that the two stars

from Earth. Such eclipsing binaries can be detected even when the two stars cannot be resolved visually
to be two distinct images in a telescope. The apparent brightness of the
image of the binary dims momentarily each time one star blocks out the
small fraction of
all
other.
of a telescope, an astronofrom binaries very accurately. The data

for an eclipsing binary are most usefully displayed in the form of light
Using a
light-sensitive detector at the focus
mer can measure
light intensity
The nature
i>j
Ihf star.
217
curves such as those shown in Figure 1 1-19. The overall shape of the
light curve for an eclipsing binary reveals at a glance such information as
whether the eclipse is total or partial (compare Figures 1 l-19a and 1 1196).
The light curve of an eclipsing binary provides detailed information

about the motion of the two stars. For example, the time between successive eclipses of the same star is the orbital period of the binary, which is
related bv Kepler's third law to the separation of the two stars. Astronomers can use Newtonian mechanics to analyze such data from the light
curve and calculate the combined masses of the two stars. If spectroscopic data are also available (so that the individual orbital speeds of the
Partial eclipse
^'OOO'^
Orbital prrtod
b Total
eclipse
aSr>QO<QÔp^2r>
Orbilal prriiKl
Tidal dislonion
^-)fC>ij^<^Figure 11-19
Representative light curves
of eclipsing binaries Tin- shiipc nj its light

curi'e laualh rex'eab many details about an
eclipsing binary.
Examples
oj (a) partial
eclipse, (b) total eclipse, (c) tiiial distortion,
and
(d) hot-spot reflect imi are
illustrated here.
d Hoi-spol
reflcciion
Thr nahiic
can be calculated from the Doppler
a/ the \l(ir
then astronomers ha\e

masses of the two stars.
The light curve of a typical eclipsing binary can be analyzed to yield a
surprising amount of detailed information. For example, the duration of
an eclipse depends upon the size of the eclipsing star and the speed with
which it inoves. If the orbital speed is known, the diameter of the star
can be computed from its light curve. Even such details as tidal distortion of each star by the other can be detected in the shape of the light
curve (see Figure ll-19f).
If one of the stars in an eclipsing binary is very hot, its radiation typically creates a hot spot on its cooler companion star. Everv time this hot
spot is exposed to our Earth-based view, we receive a little extra light
energy, which produces a characteristic "bump" on the binary's light
curve (see Figure \\-l9d).
Information about stellar atmospheres can also be derived from light
curves. Suppose that one star of a binary is a white dwarf and the other
star is a bloated red giant. By observing exactly how the light from the
bright white dwarf is gradually cut off as it moves behind the edge of
the red giant during the beginning of an eclipse, astronomers can infer
manv facts about the upper atmosphere of the red giant like its pressure
Stars
enough information
and
effect),
to calculate the individual
density.
Binary stars are fascinating objects. A single star leads a straightforward, birth-to-death existence, but exotic things can ha])])en to stars in
binary systems. One star in a binary pair might evolve rapidly, become a
bloated red giant, and have its outer layers stripped away by the gravitational pull of its companion. The result would be an aging star with its
exposed to our view.

As we shall see in later chapters, manv cinious variations are possible
interior
as the
two
stars in a binary affect
each other's evolution. First, however,

closeup look at an isolated main-sequence star that happens to be only 150 million kilometers from the Earth.
we
Summary
shall take a
Distances to the nearer stars caii l)e determined bv parallax

the apparent
of the star against the background stars while the Earth moves
shift
around
.
its
orbit.
The absolute magnitude of a

when viewed from a distance
is the apparent magnitude it would iia\e

of 10 parsecs. Absolute magnitudes are cal-
star
culated from the star's apparent magnitude and distance.
The luminosit) ol a star

second.
is
the
amount of energ\ escaping from
it
each
Photoelectric photometry measures brightness through such standard filfilters. A color index of a star is the difference between
the brightness values obtained with two different filters.
ters as the
The
UBV
surface temperature of a star can be determined from its color indiK, G. K, oi M), which is based upon
its spectriun.
from its spectral type (O, li. A.

the major patterns of spectral lines in
ces or
The Hertzsprung-Russell (H-R) diagram
is
graph plotting absolute
magnitudes of stars against their spectral types (or, equivalently,

kuninosities against surface temperatiues). It reveals the existence of various types of stars such as main-setjuencc stars, red giants, supergianls,
and white dwarfs.
The nalure of
219
Ihe stars
Binary stars are surprisingly common in the universe. Those that can be
resolved as two distinct star images by an Earth-based telescope are called
visual binaries.
The masses
of the two stars
in a
binary system can be computed from

and orbital dimensions of the system.
measurements of the
orbital period
The mass-luminosity
relation expresses a direct correlation
and luminosity
for
main-sequence
between mass
stars.
spectroscopic binary exhibits periodic shifting of its spectral lines due to

the Doppler effect as the orbits of the stars carry them alternatively toward and away from the Earth.
-An eclipsing binary
the Earth, so that
is
one
one whose
orbits are
viewed nearly edge-on from
star periodically eclipses the other.
Surprisingly detailed information about the stars in an eclipsing binary

can be obtained by studying its light curve.
Review questions
Explain the difference between apparent magnitude and absolute magni-
tude.
2 Why do you suppose that Wien's law and the Stefan-Boltzmann law are
important physical principles for the astronomer?
How
and why
4 What are the

ure a
5
*
is
the spectrum of a star related to

filters,
its
surface temperature?
and how does an astronomer use them
to
meas-
surface temperature?
star's
What
is
UBV
the chemical composition of most stars?
the hottest star listed in Table ll-l? Which is the coolest? List
the stars in Table 11-1 in order of decreasing surface temperature.
Which
is
H-R diagram and sketch the regions occupied by main-sequence

red giants, and white dwarfs. Briefly discuss the different wavs in
which vou could have labeled the axes of vour graph.
7
Draw an
stars,
8 What
apply?
is
the mass luminosit\ relation, and lo what kind of stars does
it
9 Sketch the radial velocity curve of a t)inar\ whose stars arc moving in a
is (a) perpendicular, and (b) parallel to our line of sight.
circular orbit that
Advanced questions
10 Sketch the light curve of an eclipsing binary having high orbital eccentricwhich (a) the major axes are pointed toward the Earth and (b) the
major axes are perpendicular to our line of sight.
ity in
*11 Estimate the mass of a main-sequence star that is 10.000 limes as luminous as the Sun. What is the luminosity of a main-sequence star whose mass
is one-tenth that of the Sun?
12 Why do you suppose that stars of the same spectral type but different
luminosity class exhibit slight differences in their spectra?
13
How
might a
star's rotation affect
the ap]>earaiHc of
its
speclial lines?
The nature of
For further reading
the stun
Ashbrook, J. "Visual Double Stars for the Amateur." Sky ef Telescope,

Nov. 1980, p. 379.
Evans, D., et al. "Measuring Diameters of Stars." Sky <S Telescope, Aug.
1979, p. 130.
Gingerich, O. "A Search for Russell's Original Diagram." Sky
July 1982, p. 36.

Nielsen, A. "E. Hertzsprung
Measurer of Stars." Sky iS

What
about
1968, p. 4.
Page, T., and Page, L. Starlight
It Tells
is'
Telescope,
Telescope. Jan.
the Stars.
Macmillan,
1967.
Phillip, A.,
and Green,
L.
"Henry N.
Russell
and the
H R
Diagram." Sky
306; May 1978, p. 395.

Upgren, A. "New Parallaxes for Old: Coming Improvements in the Distance Scale of the Universe." Mercury, Nov./Dec. 1980, p. 143.
<s'
Telescope,
Apr. 1978,
p.
12
Our
star
The solar corona Tltii sinking vieui of Ihr

Sun's ouln alm(i\jthnr ims prepared from
data supplied hy
satellite.
The
corona, go
llu-
Solar
Maximum
Mission
colors, representing density in Ihr
from purple (densest) to yellow

The densest coronal regions (pur-
(least dense).
ple
and
blue) are located over sunspols.
From
one of these dense repons, a prominent coronal spike extends nearly 2 million kilometers
from
the
Sun. Sei'eral other spikes can also he
seen. Shortly after this picture
was taken, a
solar flare occurred in this field of view

the shape of the corona
in
and
changed dramatically
a matter of minutes. (NASA)
We continue with our study oj stars in this chapter as we learn about the
Sun. We find that the Sun is a typical main-sequence star. We first discuss
the thermonuclear reactions that occur in the core of the Sun and the wa\s
that this energy moves from the core to the suiface. to be radiated into space.
This discussion introduces a theoretical model of the Sun's interior structure.
Next we tui~n to a variety of phenomena obsen'ed on the solar surface. We
find that the 11 -year sunspot cycle is only one aspect of a more general 22year .solar cycle tliat affects many properties of the Sun. Finally, ire mention
some of the new tools that astronomers have begun using to study the Sun
and learn more about the general nature of stars.
The Sun is an average star. Its mass, size, surface temperature, and
chemical composition lie roughh midway hetwccn the extremes exhibited
l)\ olher stars. Unlilic other stars, however, the .Sun is a\ail.il)le lor detailed, closeup exai7iination. Siiidving the .Sun iherelore olTeis excellent
insights into the nature of similar main-sequence stars.
Understanding the .Sun is important to humanity hecause the Siui is
our source of heal and light. Life would not be jiossible on Earth without the energy provided by the Sun. Even a small change in the Sun's
si/e or surface temperature could dramatically alter conditions on the
Earth, either melting the polar caps or producing another Ice Age.
is a commonplace star, it is a dramatic arena where

the fascinating and complicated interaction of matter and
energy on a colossal scale. Beautiful and bewildering phenomena occur
Although the Sun
we can observe
columns of hot gases gush up to the solar surface, interact with the
field, and dissipate energy into the Sun's outer atmosphere. The source of all this energy lies buried at the Sun's center.
as
Sun's magnetic
The Sun's energy

is produced by
thermonuclear reactions
the core of the Sun
in
During the nineteenth century, geologists and biologists found convincing evidence that the Earth must have existed in more or less its present
form for hundreds of millions of years. This fact posed severe problems
for physicists because it seemed impossible to explain how the Sun has
been shining for so long, radiating immense amounts of energy into
space. If the Sun were made of coal, for example, it could burn for only
3000 years.
A key to this dilemma was provided in 1905 by Albert Einstein's special
theory of
that matter
relativity.
One
of the implications of Einstein's theory

to the simple
is
and energy are interconvertible according
equation
In other words, a mass (m) can be converted into an

(E) equivalent to mc^,
where
c is
amount of energy
the speed of light. Because
f is
a large
number and c is huge, a small amount of matter can be converted into

an awesome amount of energy.
Inspired by Einstein's work, astronomers began to wonder if the Sun's
energy output might come from the conversion of matter into energy.
But exactly what kind of mechanism would transform matter into energy?
In the 1920s the British astronomer Arthur Eddington showed that
temperatures near the center of the Sun must be much greater than had
previously been thought. Another British astronomer, Robert Atkinson,
suggested that under such high-temperature conditions near the Sun's
center hydrogen nuclei might fuse to produce helium nuclei in a reaction that would transform a tinv amount of mass into a very large
amount of energy. Because the nuclei fuse together under high temperatures, this process is called thermonuclear fusion.
Recall that the nucleus of a h\clrogen (H) atom consists of a single
proton. The nucleus of a helium atom (He) consists of two protons and
two neutrons. In the nuclear process
4
H^He
two of the four protons from hydrogen are changed into neutrons to
produce a single helium nucleus.
In the conversion of hydrogen into helium, matter is lost because the
ingredients (four hydrogen nuclei) weigh very slightly more than the
product (one helium nucleus). Specifically, the mass lost during this reaction
may
be calculated as follows:
4 hydrogen atoms =
6.693 x 10"-^ g
-1 helium atom = -6.645 x

mass
lost
10 -^ g
0.048 x 10""^ g
223
This lost mass is converted iino energy

equation E = mc'.
in the anioinit
predicted
1)\
tlie
To produce the luminosity of the Sun (3.9 x 10'' erg/sec). 600 million
metric tons of hydrogen must be converted into helium within the Sun
each second. This prodigious rate is possible because the Sun contains a
vast supply of hydrogen

enough to continue the present rate of energy
output for another 5 billion years.
In this process of nuclear fusion, the coinersion ol h\cirogeii into helium at the Sun's center is lalied hydrogen burning even though nothing
is actually "burned" in the conxeniional sense of that word.
The ordinary burning of wood, coal, or any flammable substance is a chemical
process involving only the electrons that orbit the nuclei of atoms. Ther-
monuclear fusion is a far more energetic process that involves violent

between the nuclei of atoms.
Hydrogen burning occurs at a star's core where it is ver\ hot (15 million K in the case of the Sun). Normally, the positive electric charge on
protons is quite effective in keeping the protons far apart, because like
charges repel each other. In the extreme heat of the Sun's center, however, the protons are moving so fast that they can penetrate each other's
electric fields and stick together. In Chapters 14 and 15, we will see that
other thermonuclear reactions (such as helium burning, carbon burning,
and oxygen burning) occur late in the lives of many stars.
collisions
model of the
us how energy
gets from the Sun's center
to
Its
hidden from our view, we can use the

is going on below the solar surface (see
The resulting model of the Sun tells us about the Sun's
Although the Sun's
theoretical
Sun can
interior
is
laws of physics to calculate what
tell
Figure 12-1).
its pressure, temperature, and density.

model of the Sun or of any other stable star, we
the Sun is not undergoing any dramatic changes. The
internal characteristics such as
surface
In developing a
should note that

Sun is not exploding or collapsing, nor
cooling off. The Sun is thus in balance
is it
in
significantly healing
up or
two wa\s: mechanically and
thermally.
Mechanical balance, often called hydrostatic equilibrium, means simis supporting its own weight. Because of gravity, the tre-
ply that a star
Figure 12-1
is
the only star
The Sun's surface
The Sun
whose
can be
sitrjace details
examined through Earth-based
telescopes.
As
tronomers always take great care (with extremely dark filters
tir l>y
image onto a screen)

to their eyes.
Never
to
projecting the
took directly at the
Sun, with or without a
Sun\
avoid severe damagi
telescope.
(Celestron International)
mendous weight of the Sun's outer layers pressing inward from all sides
tries to make the star contract. As gravity compresses the star, however,
gas pressure inside the star increases. The greater the compression, the
higher the internal pressures. Hydrostatic equilibrium is achieved when
the pressure at every depth within the star
port the weight of the overlying layers.
is
exactly sufficient to sup-
Thermal balance, often called thermal equilibrium, means simply that

a star keeps shining. Vast amounts of energy escape from the Sun's surface each second. This energy is constantly resupplied by thermonuclear
fusion in the Sun's core at a steady, persistent rate that maintains ther-
mal equlibrium. But exactly

center to its surface?
how
is
energy transported from the Sun's
Experience teaches us that energy always flows from hot regions to

cooler regions. If you heat one end of a metal bar with a blowtorch, the
other end of the bar eventually becomes warm. This method of energy
transport is called conduction. Conduction varies significantly from one
substance to another (copper is a good heat conductor, wood is not),
depending on the arrangement and interaction of the atoms. Calculations have demonstrated that conditions inside stars like the Sun are not
favorable for conduction, so this means of energy transport is not an efficient one in ordinary stars. (In Chapter 14 we shall see that conduction
is important in very compact stars such as white dwarfs.)
Two other means of energy transport convection and radiative diffusion
are active inside main-sequence stars to move energy from a star's
center to its surface. Convection involves the circulation of gases between hot and cold regions. Just as a hot-air balloon drifts skyward, so
hot gases rise toward a star's surface while cool gases sink back down
toward its center. The net effect of this physical movement of gases is to
transfer heat cnerg\- from the center toward the surface.
In radiative diffusion, photons migrate outward from the star's hot
core, where they are constantly created, toward the cooler surface, where
they escape into space. The path of an indi\idual photon is quite random as it is knocked about between the atoms and electrons inside a
star. In all, it takes nearly a million years for energy created at the Sun's
center to reach the solar surface finally and escape as simlight.

The concepts of hydrostatic equilibriimi, thermal equilibrium, and
energy transport can be expressed in the form of a set of mathematical
equations, that are collectively called the equations of stellar structure.
These equations can be solved to yield the conditions of pressure, tem-
perature, and density that must exist inside a star to maintain equilibrium. Astrophysicists use high-speed computers to solve the equations of
stellar structure, thus developing a detailed theoretical model of the
structine of a star. The astrophysicist begins with astronomical data
about the star's surface such as, for example, that the Sun's surface temperature is 5800 K, its kmiinosity is 3.9 x 10'* erg/sec, and the gas pressiue and density are almost zero. The equations of stellar structure are
used to calculate conditions layer by layer toward the star's center. In

this way, the astrophysicist can discover how temperature, pressure, and
density increase with increasing depth below the star's surface. This information constitutes a stellar model. We have learned through this technique that the temperature at the Sun's center is 15.5 million kelvins and
the pressure is 3.4 x 10" atmospheres.
Figure 12-2 presents a theoretical model of the Sun. Note that the
fraction of luminosity rises to 100 percent at about one-quarter of the
way from the Sim's center to its surface. This fact tells us that the Sun's
225
Figure 12-2
Sun
The Sun's
theoretical
here with âplis thai slum'

temperuliire
the distance
mass,
from
and
surface) equals
how
is
tlie
displayed
deusilt.
lujiihwsilf I'ary with
the Sun's center.
from the Sun's

696,000 km.
radius (the distance

its
model of the
iuleniiil slrurturc
.solar
renter to
0.2
0+
0.6
0.8
Distance from Sun's center (solar radii)
energy production occurs within
xoliime extending out to
:j
solar ra-
dius.
Also note that the fraction of mass rises to nearly 1 00 percent al

about 0.6 solar radius from the Sun's center. Most of the Sun's mass
must therefore be confined to a volume extending only slightly more
than half the way from the Sun's center to its surface. The outer lavers
of the Sun thus contain verv little matter.
From the Suns center out to 0.8 solar radius, energy is transported
by radiactive diffusion, as we have seen. This inner region is therefore
called the radiative zone.
dcnsit\ of matter in
tion
tiie
From
Sun
is
0.8 solar radius to the Sun's surface, the

so low (less than 0.01 g/cm*) that convec-
dominates the energ\ How.
We
tive zone, or convective envelope.
thus say that the Sun has a convecThese aspects of the Sim's internal
structure are sketched in Figure \2-'.'>.

A small but troublesome |)roblem remains lor
siiiiiiists ii\ing to understand the internal structme of the Sun. The thei nioinu lear leadions
thai i)i<)diue the Sim's energy should also release niimeious particles
called neutrinos, which are exiremelv elusive and ver\ diUlcull to delect.
Phvsicisis ha\e used huge t.uiks llllid with leatiing ihiid (CijCll,) to trap
neutrinos because thev convert the chlorine into aigon gas which can be
collided and measured. However, all such experiments delect only about
one-iliii(l ol the ex|)ecied nimiber of neuliiiios (rom the Sun. \'arious
explanations have been pioposed for this uiiexpecied result hui luriher
i
research
is
certainly required.
Figure 12-3 The Sun's internal structure

Thermonuclear reactions occur in the Sun's
core,
which extends
radius
from
to a
the center.
distance of 0.25 solar
Energy from
the core
is
transported outward via radiative diffusion to
a distance of 0.83 solar radius. Convection n

responsible for energy transport in the
Sun
outer layers.
The photosphere is the

lowest of three main layers
in the Sun's atmosphere
Although astronomers often speak of the solar surface, the Sun really
does not have a surface at all. As you move in toward the Sun, you encounter ever more dense gases but no sharp boundary like the surface
of the Earth or Moon. The Sun appears to have a surface (see Figure
12-1) because there is a specific layer in the Sun's atmosphere from
which most of the visible light comes. This layer, which is probably not
more than 500 km thick, is appropriately called the photosphere
("sphere of light").
The photosphere is the lowest of three layers that together constitute
the Sun's atmosphere. Above the photosphere are two additional layers,
the chromosphere and the corona, which are discussed later in this chapter. At visible wavelengths you cannot see through the shimmering gases
of the photosphere, so everything below the photosphere is called the
Sun's interior.
Convection in the Sun's outer layers affects the appearance of the

photosphere. Under good observing conditions with a telescope and
using special dark filters to protect your eyes, you can often see a
blotch v pattern called granulation (see Figure 12-4). Each light-colored
granule measures about 1000 km across and is surrounded by a darkish
boundarv. High-resolution spectroscopv reveals a blueshifting of spectral
lines in the central, bright regions of a granule and a redshifting along
the dark, intergranule boundaries. These Doppler shifts show that hot
gases rise upward in the granules, cool off, spill over the edges of the
granules, and plunge back down into the Sun along the intergranule
boundaries. The difference in brightness between the center and the
edge of a granule corresponds to a temperature drop of 300 K.
Granules are individual convection cells in the Sun's outer layers.
Time-lapse photography shows that granules form, disappear, and then
re-form in cycles lasting several minutes. At any one time, roughly 4 mil-
221
Figure 12-4
Solar granulation
Higli-
leiolutiun jiliolof^Kiphs aj the Sim's sutjnci-
reveal a blotchy patttrn called granulatiun.
Granules, each measuring about
1000 km
across, are convection cells in the Sun's outer

layers.
iSan Fernando Ohserx'aton)
G?
^^^f^
lion granules cover the solar surface.
Depth in photosphc
which visible hi;hi
Figure 12-5
Limb darkening
Each granule occupies an area

roughly equal to Texas and Oklahoma combined (about lO" knr).
The photosphere appears darker around the edge, or limb, of the
Sun than it does toward the center of the solar disk. This phenomenon,
called limb darkening, is caused by photon absorption in the solar almospheie. Photons emitted in warm layers fairlv deep in the solar atmosphere are able to escape into space only along the relatively short
paths directed away from the Sun's center. The photons reaching the
Earth from the edges of the Sun's visible disk come from cooler lavers
nearer the top of the solar atmosphere, because the photons from the
deeper layers are absorbed before the\ are able to escape along the
much longer paths ihev must travel toward the Earth (see Figure 12-5).
Less intense light from the cooler, more shallow layers of the Sun shows
up in photographs as limb darkening.
The photosphere shines with a continuous, nearly perfect blackbodv
spectrum corresponding to its average temperature of about 5800 K (recall Figure 5-4). Because stars behave like blackbodies, manv important
concepts about blackbodies (Wien's law. the Stefan-Boltzmann law. and
others) help us understand the nature of the Sim. (Blackbodv radiation
was discussed in depth in Chapter 5.)
The spectrum of the photosphere (recall Figure 5-5) exhibits numerous absorption lines. As we saw in Chapter 11 (see Figure 11-8). the
spectral lines in a star's spectrum are directlv related to the star's surface
temperature. For example, the nimierous dark lines caused bv singlv
ionized calcium and iion (Ca II and Fe II) in the Sims spectrum are
originates in deep,
sphere.
Hut
warm
the light
Sun's limb originates
layers of the photo-
coming
to us
temperatme of 5800 K.
Above the photosphere, higher regions
consistent with a
Light reach-
ing the Eiirlh from the center oj the solar disk
from
the
higher, cooler layers
of the photosphere. The center of the Sun's

disk therefore appears brighter than the limb.
in the Sun's atmosphere display different spectral lines. Nevertheless, the temperature of the gas can
always be deduced from the strengths of these spectral lines. In this way,
astronomers are able to measine the temperature at various depths in
the
Suns
atntospheie.
sj)eclra in (Ihapter 5.)
Fhc reader may wish
to review the subject
of
The chromosphere
is located between the
photosphere and the Sun's
outermost atmosphere
Immediately above the photosphere is a 2000-km-thick layer called the

color"), which is the second of the three
major levels in the Sun's atmosphere. When the Moon blocks out the
photosphere during a total eclipse, the chromosphere is visible as a pinkish strip around the edge of the dark Moon.
The spectrum of the chromosphere is dominated by emission lines.
The characteristic pinkish color of the chromosphere is caused by the
Balmer line H. wbich is the strongest emission line in the red region of
the chromosphere's spectrum. The blue end of the chromosphere's spectrum is dominated by two bright emission lines called the H and K lines
of ionized calciiun.
The photosphere emits almost no light at the wavelengths of H and
the calcium H and K lines. The photospheric spectrum has broad, dark
absorption lines at these wavelengths. The chromosphere, however, is
especially bright at these wavelengths. Astronomers can thus study details
of the chromosphere bv viewing the Sun through special filters that are
transparent to light only at the wavelengths of H or the calcium lines.
Figure 12-6 is a photograph of the chromosphere taken through an
H filter. Note the many dark, brushlike spikes that protrude upward.
These spikes, called spicules, are jets of gas surging up out of the Sun.
A typical spicule rises at the rate of 20 km/sec, reaches a height of abotit
7000 km, then collapses and fades away aftei a few minutes. At any one
time, roughly 300,000 spicules exist, covering a few percent of the Sun's
chromosphere ("sphere of
surface.
Spicules are generally located on the boundaries between large, organized cells called supergranules. Detailed observations of the solar surface in the 1960s revealed the existence of these supergranules, which
Figure 12-6
Spicules
chromosphere
thai risf
up mto warmer
outer atmosphere.
ble in this
many
details
and
the
Spicules are jels of cool gas

regiotis
Numerous
of Ike Swi's
spicules are
I'isi-
photograph, which also shows
of the chromosphere. Spicules
an
located along the irregularly shaped boundaries
between supergranules.
(NOAO)
'
229
are about 30,000 km in diameter and contain many hundreds ot ordinary granules. Gases rise upward in the middle of a supergranule and
move
horizontally
outward toward
its
edge, where they descend back
into the Sun.
The corona
The outermost
is
the outermost layer of

the Sun's atmosphere
10,000
8000
c:
iiona
le
hi""
1.1
Hlon
Chromospherf
Phut .sph. r,'^''
'
'
Figure 12-7
Suii'i
atmusphne luu
photosphere
many
up
to
500 km
an
exleiids to
2000 km above
jutting
Thr
The
three dislmcl luyen.
roughly
is
chromosphere
'
The solar atmosphere

thick.
The
altitude of about
the photosphere, with spicules
10.000 km. The corona extends
and
millions of kilometers out into space
merges with the \olar
u'lncl.
Adapted
/mm
John A. Eddy)
Figure 12-8
traordman
The solar corona
pliiiliigmph
xms taken
/lying 10.000 /eel above

total solar eclipse o/
Numerous
I lii\ ex
pom
February 26. 1979.
.streamers are visible, extending
of 4 million kilometers
above the solar surface. (Los Alamos
to distances
Scientific Laboratory)
jel
MonUitm during
the
region of the Sun's atmosphere is called the corona. It

extends fiom the top ol the chroinosplieie out to a distance of several
million kilometers, where it merges into the solar wind. (.As noted at the
end of Chapter 6, the solar wind consists lA high-speed protons and
electrons constantly e.scaping from the Sun). The three layers of the
Sun's atmosphere are sketched schematically in Figure 12-7.
The total amount of visible light emitted by the solar corona is comparable to the brightness of the Moon at full moon
only about onemiilionih as bright as the photosphere. The corona can thus be seen only
when the photosphere is blocked out during a total eclipse or in a specially designed telescope called a coronagraph. Figure 12-8 is an exceptionally detailed photograph of the corona. Numeious coronal streamers
are visible, extending outward to a distance of 6 solar radii.
.Around 1940, astronomers realized that the spectrum of the Sun's
corona contains the emission lines of a number of highlv i(jnized elements. For example, there is a prominent green line caused by the presence of Fe XIV (iron atoms each stripped of 13 electrons). Extremely
high temperatures are required to strip that many electrons from atoms,
so it was clear that the corona must be very hot. It is now known that
coronal temperatures are in the range of 1 to 2 million kelvins.
Recent observations from space show that the Sun's corona exhibits
complicated structure and activity. For example, the huge, bubblelike disturbance seen in Figure 12-9 is called a solar transient. These shcjrt-lived
protuberances erupt suddenly and expand lapidK outward through the
corona. Solar transients piobably occur as often as once a day, but they
were unknown until the Siui was examined with a coronagraph carried
aloft by Skylab.
Figure 12-9
solar transient
During
lln
early 1970s, astronauts on Skylab discovereil
huge bubbles erupting outward through
the
corona. This solar transient was photographril
1973. The leading edge of the bubble

moved outward from the Sun at a speed
500
kmlsec. (NASA and the High
of
in
Altitude Obseivaton)
Figure 12-10
coronal hole in
This X-ray picture of the
Sun wcu
Skylab astronauts in 1973.
rays
taken b\
huge, dark,
boot-shaped coronal hole dominates
this
view of the inner corona. Numerous bright

points are also visible.
(NASA and Harvard
College Observatory)
X-ray photographs of the corona were also obtained during the SkyThe corona has temperatures in the million-kelvin range,
and thus it should be shining brightly at X-ray wavelengths of roughly
30 A (recall Wien's law). The Skvlab pictures reveal a verv blotchv. irregular inner corona. There are large dark regions, called coronal holes,
which are nearlv devoid of the hot, glowing gases. Man\ astronomers
suspect that coronal holes are the main corridors through which particles
of the solar wind escape from the Sun.
X-ray photographs also reveal numerous bright spots that are hotter
than the surrounding corona. Temperatures in these bright points occasionally reach 4 million kelvins. Many of the bright coronal hot spots
seen in X-ray pictures such as Figure 12-10 are located over sunspots.
lab missions.
Sunspots are only one of
Superimposed on the
many phenomena
of
associated
with the 22-year solar cycle
phenomena
atmosphere are a host

shaped dark
Figure 12-11) are the most easily recog-
basic structure of the solar
that vary with a 22-year period. Irregularly
regions called sunspots (sec

nized of these phenomena because the\ occur in the photosphere, the
most visible part of the Sun's atmosphere. The yery dark central core of
a sunspot is called the umbra; it usually is surrounded by a less-dark
border called the penumbra. .Although they \ary greatly in size, typical
sunspots measure a few tens of thousands of kilometers across.
On rare occasions, a sunspot group will be so large that it can be seen
with the naked eye (always be sure to use special dark filters or other
means to protect your eyes when obserying the Sun). .Ancient Chinese
astronomers recorded such sightings 2()()() years ago. (;alileo was the first
person to examine sunspots in detail with a telescope which, of course,
gives a
much
better view. In fact. Galileo discovered that he could deterrate by following sunspots as they nio\ed across
mine the Sun's rotation
the solar disk (see Figure 12-12). He found that the Sun rotates once in
about four weeks. A typical sunspot group lasts about two months, so it
can be followed for two solar rotations.
Careful observations demonstrate that the Sun does not rotate as a
rigid body: the equatorial regions rotate more rapidly than the polar
regions. .A sunspot near the solar equator takes 25 days to go once
around the Sun. However, at 30 north or south of the equator a sunspot takes 27i days to complete a rotation. The rotation period at 75
north or south of the equator is about 33 days, and at the poles it may
be as long as 35 days. This phenomenon is called differential rotation,
because different parts of the Sun rotate at slighth different rates.
Observations over many years reveal that the number of sunspots
changes in a periodic fashion. In some years there are many sunspots. in
others, almost none. This pattern is called the sunspot cycle. .As shown
in Figure 12-13, the average number of sunspots \aries with a period of
Figure 12-11
tofriafih n'd-s liiken
al
an
sunspot group
Ity
This plm
a hnUmm-hnrne
altilude oj toujihiy
50.00U
ft.
Irlruiijir
\nt,-
featherlike appenrance of the penumbrae.

notice the granulation
III,
Ah"
the surrounding,
undisturbed photosphere. (Project Straluscope:
Princeton University)
Figure 12-12
sennng
day
the
The Sun's rotation
to the next,
rotates once in
rial regions
By
nh-
same group of mn.spols from one

Galileo foutiil that the
about four
of the
what faster than
Sun
iveeks.
Sun
194 7 March 3
^| II
March 31
^B
1<)47 April
Tlw equato-
actually rotate someOvercast March 4
the polar regions. This series
of photographs shows the same sunspol group
(Mount Wihon and
over li solar rotations.
Las Campanas Observatories)
L-
^^Êk.
March 5
t-
'
March 6
'IK
A^Bk i^
April 3
>
April 2
March
^^Bk
*^
March
^^Ht
March 8
April 4
April 5
'mm
JH^^
')
.April
6
'
-.
March 10
^|^^
.tpril 7
March 11
;Î^HIl
.-ipril
March 12
^1^^
April
>
jH
//Ik
March 13
March 14
^^^^^
.-ipril
11
March 15
il^^
.ipril
12
March lb
U^^
.April
13
^ÎB^
April 10
'
/ 'j|
^'
M
/M
time of exceptionally many sunspots is called a sunsuch as occurred in 1959, 1970, and late 1980. The Sun
was almost devoid of sunspots (called times of sunspot minimum) in
about
spot
1 1
years.
maximum,
1954, 1965,
and 1976.
American astronomer George
Ellery Hale made the important discovery that sunspots are associated with intense magnetic
fields on the Sun. When Hale focused a spectroscope on sunlight coming
from a simspot, he found that each spectral line in the normal solar
In 1908, the
Figure 12-13
The sunspot cycle The

the Sun vanes with a
number of sunspots on
period of about
1 1 years.
Large numbers of
sunspoLs were seen in 1959, 1970,

1
and
late
980. Exceptionally few sunspots were ob-
served in 1954, 1965,

sunspot
sunspot
minimum
maximum
is
in
and 1976. The
due
in
1992.
next
1987. the next
1860
1870
1880
1890
1900
1910
1920
Year
19:i0
1940
1950
1960
1970
1980
233
spectrum is flanked by additional, closely spaced spectral lines not usually observed (see Figure 12-14). This "splitting" of a single spectral line
into two or more lines is called the Zeeman effect after the Dutch physicist Pieter Zeeman, who first observed it in 1896 in his laboratory. Zee-
man showed
that the spectral lines split
when
the light source
is
inside
an intense magnetic field. The more intense the magnetic field, the
wider is the separation of the split lines. The splitting of the Fe 1 spectral line in Figure 12-14 into three lines corresponds to a magnetic field
roughly 5000 times more intense than the Earth's natural magnetic field.
Hale"s discovery demonstrates that sunspots are areas where a concentrated magnetic field protrudes through the hot gases of the photosphere. Because of the temperature, many atoms in the photosphere arc
ionized so that the photosphere is a mixture of electric charges (ions and
electrons). This blend, technically called a plasma, is an extremely good
conductor of electricity that interacts vigorously with magnetic fields.
Specifically, a magnetic field restricts and constrains the motions of a
plasma. The intense magnetic field in a sunspot greatly inhibits the natural convective motions of the gases. Energy cannot fiow freely upward
from the Sun's convective zone, and the gases in this region of the photosphere cool off. Temperatures in a sunspot are in fact tvpicallv 4()00 to
4500 K, or more than 1000 K cooler than in the surrounding, undisturbed photosphere. Because of this lower temperature, sunspots look
dark in contrast to their brighter surroundings.
A host of exotic phenomena occur around and above sunspots as a
direct result of their intense magnetic fields. Huge, arching columns of
gas called prominences often appear above sunspot regions (see Figure
12-15). Some prominences hang suspended for days above the solar surface while others blast material outward from the Sun at speeds of
roughly 1000 km/sec.
The most violent, eruptive events on the Sun, called solar flares,
occur in complex simspot groups. During a solar flare, temperatures in a
compact region soar to 5 million kelvins. Vast quantities of particles and
radiation are blasted out into space. The fiare is usuallv over within 20
minutes. When these ejected particles arrive at the Earth a dav or so
later, they interfere with radio communications and produce beautiful,
shimmering lights called aurorae in the night sky. Most astronomers suspect that prominences and flares involve concentrated portions of the
Sun's magnetic field.
Figure 12-14
Zeeman
splitting by
sunspot's magnetic field

(Irmim
iirriiw llir
\iiti\jiiil
toward which the
slit iij
(a) llir black Inn-
iniluntes the InrtilKiii
the spectroscope
was
aimed, (b) In the resulting spectrogram, one

lute in the middle oj the
spill into three
normal solar
spectruii
components. The separatuoi
between the three lines rorresponth

netic field roughly
5000
to
mag
time\ stronger than
the Earth's natural magnetic field.
(MOAO)
Figure 12-15
this
A huge
prominence
prominence arches above
the solar surface in
Skylab photograph taken on December
1973. The radiation thai exposed

is
from
singly ionized helium
(He
this
1 9.
picture
II) at a
wavelength of 304 A, corresponding
to a
tem-
perature of about 50.000 K. (Naval
R esea rch La bo ra to n)
Figure 12-16
sunspot group
ture
disl>lii\^
III,-
A magnetogram of a
Thi'. arlificiallt
nitciiît\
and
magnclti jirhl n-.\n, lated -with
group.
One
side nj
colored
pn-
folarity of the
a large sunspot
a ttpual sunspot group has
one magnetic polarity and the other side has

the opposite polarity.
(NOAO)
^^^
Astronomers can construct
artificial
pictures called
magnetograms
that
display the magnetic fields in the solar atmosphere by combining two
wavelengths on either side of a magnetically split

in Figure 12-16 shows a large sunspot
group. Dark blue indicates the area of the photosphere covered by north
magnetic polarity and yellow designates the area covered by south polar-
photographs taken
spectral line.
at
The magnetogram
Many sunspot groups are said to be bipolar, meaning that they have
roughly comparable areas covered by north and south magnetic polarities. The simspots on the side of the group toward which the Sun is rotating are called the preceding members of the group. The remaining
spots, which follow behind, are called the following members.
After years of studying the solar magnetic field. Hale was able to
piece together a remarkable magnetic description of the solar cycle. First
of all. Hale discovered that the preceding spots of all simspot groups in
one solar hemisphere have the same magnetic polarity. We now know
same
hemisphere in which the

hemisphere of the Suns north
magnetic pole, the preceding members of all sunspot groups have north
magnetic polarity. In the south magnetic hemisphere, all the preceding
members have south magnetic polarity.
In addition. Hale found that the polarity pattern completely reverses
itself every
1 years. For instance, the hemisphere that has a north magnetic polarity at one solar maximum has a south magnetic polarity at the
next solar maximum. For this reason astronomers prefer to speak of a
that this polarity
group
is
is
the
as that of the
located. In other words, in the
solar cycle with a period of 22 years, rather than the sunspot cycle having a period of 1 1 years.
In 1960, the .American astronomer Horace Babcock proposed a deseems to account for many aspects of the 22-year solar
scription that
Babcock's scenario, called a magnetic-dynamo model, employs two

of the Sun: its diffei eiitial rotation and its convective
envelope. Differential rotation causes the magnetic field in the photosphere to become wrapped around the Sun as shown in Figure 12-17.
Convection in the photosphere causes the concentrated magnetic field to
c\cle.
basic properties
become
and kinks erupt through the solar surface. Sunspots

where the magnetic field protrudes through the photosphere. Careful examination of how the magnetic field in Figure 12-17
is oriented reveals that the preceding members of a sunspot group
should indeed have the same luagnetic polarity as that of the hemisphere in which they are located.
There are many things that the magnetic-dynamo model fails to explain. Most embarrassing to astronomers is a basic lack of understanding
of the physics of sunspots. We still do not know what holds a sunspot
together week after week. Our best calculations predict that a sunspot
should break up and disperse as soon as it forms.
Our understanding of sunspots is further confounded hv irregularities
appear
Figure 12-17
Babcock's magnetic dynamo
posiible partial explanaliun jor the sunspul
cycle involves the
around
spots
the
Sun
wrapping of a magnetic field

Sun-
by differential rotation.
appear iihere the concentrated magnetic
field has hrnh-n Ihnms^h Ihr solar sntiarr.
tangled,
at locations
example, the overall reversal of the Sun"s magnetic

often piecemeal and haphazard. One pole mav reverse polarity
long before the other so that for several weeks the Sun may have two
in the solar cycle. F'or
field
is
north poles but no south pole at all. To make matters worse, there is
strong historical evidence that all traces of sunspots and the sunspot
cycle have vanished for many years. For example, virtually no sunspots
were seen from 1645 through 1715. Similar sunspot-free periods apparently occurred at irregular intervals in earlier times.
Solar seismology and new

satellites are among the
latest tools for solar research
Figure 12-18
The Solar
Maximum
SMM
Mission spacecraft The

spacecraft
was specifically designed to study solar flares
during the most recent sunspot maximum. It
was launched
10
montlis.
in
Two
1980 and
the satellite above the

the
seeti
repairing
open cargo-bay doors of
Space Shuttle. (NASA)
Summary
recorded data for
astronauts are
Puzzles in solar research have prompted astronomers to question many

long-standing ideas about the Sun. For example, the magnetic field that
produces simspots might be bits and pieces of ancient magnetism
trapped deep inside the Sun as it formed 45 billion years ago. To search
for such a magnetic field or to study other deeply buried phenomena,
astronomers need a way to probe the solar interior.
Geologists are able to study the Earth's interior structure by using a
seismograph to record vibrations during earthquakes. Although there are
no true sunquakes, the Siui does vibrate at a variety of frequencies,
somewhat like a ringing bell. After these vibrations were first noticed in
1960, sensitive Doppler-shift measurements revealed that portions of the
Sun's surface move up and down by about 10 km every 5 minutes.
Slower vibrations with periods ranging from 20 minutes to nearly an
hour were discovered in the 1970s. More recently, a variety of longperiod oscillations has been foimd. The possibility of detecting extremely
slow pulsations has inspired astronomers who call themselves helioseismologists to set up telescopes at the South Pole, where the Sun can be
observed continuously for many days. This new field of solar research is
called solar seismology.
No one knows why the Sun vibrates. The five-minute oscillations are
similar to sound waves but are more complicated. Segments of the Sun's
surface rise and fall as the waves move back and forth in the Sun's convective zone. Slower oscillations, such as the 30-minute vibrations, are
believed to originate very deep in the Sim. Many helioseismologists are
hopeful that observations of these long-period oscillations will yield detailed information about the solar interior.
Some solar astronomers are probing the Sun's core, and others have
concentrated on observations of the solar atmosphere from Skylab and
Earth-orbiting satellites. During the 1970s, a series of satellites called the
Orbiting Solar Observatories provided a wealth of data about the Sun at
ultraviolet. X-ray, and gamma-ray wavelengths. In 1980, a satellite called
the Solar Maximum Mission spacecraft (see Figure 12-18) was launched
to study solar fiares. Unfortunately, the system that keeps the satellite
pointed toward the Sun failed after only 10 months of operation. In
1984, NASA scientists used the Space Shuttle to repair the
spacecraft so that its instruments can continue to send back spectra and pictiues of the most violent phenomena occurring on the surface of our
SMM
star.
The
Sun's energy is produced by the thermonuclear process called hydrogen burning, in which four hydrogen nuclei combine to produce a single
helium nucleus and release energy.
The energy
released in a thermonuclear reaction comes from the conversion of matter into energy according to Einstein's equation E = mc~.
Thermonuclear reactions occur only at very high temperatures and hydrogen burning occurs only at temperatures of more than about 8 million
kelvins.
stellar iiiodel
is
calculations based
of a star's interior derived from

The solar model suggests that
core that extends from the Sun's center to
a theoretical description
upon
the laws of physics.
hydrogen burning occurs
in a
about 0.25 solar radius.
Throughout most of the Sun's interior, energy moves outward from the
tore by radiative diffusion. In the Sun's outer layers, energy is transported to the Sun's surface by convection.
The
visible surface
tom
of the solar
of the Sun is a layer calUfi the photosphere at the botatmosphere. The gases in this layer shine with almostperfect blackbody radiation, and convection produces features called granules there.
Above the photospheie
is a layer of cooler and less dense gases called the

chromosphere. Jets of gas called spicules extend upward from the photosphere into the chromosphere along the boundaries of supergranules.
The outermost layer of thin gases in the solar atmosphere is called the
corona, which blends into the solar wind at great distances from the Sun.
The gases of the corona are very hot but at low density. Solar transients,
coronal streamers, and coronal holes are other features associated with the
corona.
Surface features on the Sun vary periodicallv in a 22-year solar cycle.
Sunspots are cool regions pioduced by local concentrations of the Sun's

magnetic field. The average number of sunspots increases and decreases
in a
regular 11 -year cycle.
solar flare
is
a brief eruption
(jf
hot. ionized gases
liom
sunspot
group.
The magnetic-dynamo model suggests that many features of the solar

cycle are caused by the effects of differential rotation and convection on
the Sun's magnetic field.
Space missions to observe the Sun from outside the Earth's atmosphere
and studies of solar seismology are expected to yield further understanding of the Sun (and hence of stellar processes in general) in the next few
decades.
Review questions
Describe the dangers
in
attempting to olxscrve the Sun.
How
have astron-
omers circumvented these problems?

2 Give an everyday example of hvdrostaiic
example of thermal equilibrium.
3(iive
some everyday examples
ecjiiilibrium.
of (()n<lu( lion,
comec
Give an everyday
lion,
and ladiative
diffusion.
4 What are the three main layers

you best observe them?
5 When
Kxplain.
6 What
7 Why
do you think the next
is
a stellar
of the Sim's
siins|)oi
model? What does
atmospheie. and bow would
maxiiiHun .nid miinminii
a stellar
model
tell
will
occin?
us .ibout a star?
the solar cycle said to have a period of 22 years. e\en though the
sunspot cycle is only 1 1 years long?
is
8 What is hydrogen burning? Win does Incliogen Inirmng

around the Sun's center?
occiii
onh
9 How do astronomers detect the presence of a magnetic

such as those in the solar photosphere?
Advanced questions
10
What would happen
if
field in
hot gases,
the Siui were not in a state of h\clrostatic or ther-
mal equilibrium?
"11
Assuming
that the current rate of
hydrogen burning
in the
Sun remains
constant, what fraction of the Sun's mass will be converted into helium over
the next 5 billion vears? How will this affect the chemical composition of the
Sun?
12 Discuss some of the difficulties of correlating solar activitv with changes in

the terrestrial climate.
13 Describe some of the advantages and disadvantages of observing the Sun
from both space, and the South Pole. What kinds of phenomena and issues
do solar astronc mers want to explore from Earth-orbiting and Antarctic observatories?
For further reading
Eddy,
J.
"The Case of the Missing Sunspots."
Scientific
American,
May
1977.
Eddv,
J.
A Xew
Sun.
NASA
SP-402. 1979.
superb, lavishly illuslrated in-
troduction to the Sun.
Frazier, K.
Our Turbulent Sun. Prentice-Hall, 1983.
"Helioseismologv." Scientific American, Sept. 1985.

J. Cornell and P. Gorenstein, eds., Astronomy from Space. MIT Press, 1983.
Mitton, S. Daytime Star. Scribners, 1 98 1.
Noves, R. The Sun, Our Star. Harvard University Press, 1982.
Robinson, L. "The Disquieting Sun: How Big, How Steady?" Sky i^ Telescope, Apr. 1982, p. 354.
Wolfson, R. "The Active Solar Corona." Scientifc American, Feb. 1983.
Leibacher,
J. et al.
Levine, R.
"The New Sun." In
13
Reflection
and emission nebulae
The
The
lives of stars
livn
hhiish iihjirts (Inward ihr upper left side of ihis
plioldgiaph) are re/leetinn nebulae surround-
ing Iwo young, hot main-sequence

stellar dust
around
these
hluuh
light.
reflects their
stars.
Inin
two stars efficienth

Several smaller
flection nebulae are scattered
around
re-
the large
reddish patch of ionized hydrogen gas. Dust
mixed with the gas dilutes the intense red cwm

sion of the hydrogen atoms with a soft bluish
haze. (Anglo-Auslralian Obsen'atoiy)
.slurs and calculations of stellar models have led

a theory of .stellar evolution. In this chapter we learn hou' protostars form
in cold clouds of interstellar dtist and gas, then evolve into main-.secptence
A.slrviwmcrs' ubsen'utiuus oj
to
stars.
We
.see
that the spiral
can trigger the
amis of a galaxy and
birth of stars in these giant
e.\plosio)ts of supernovae
gas clouds. \'exl we turn to the
evolution of stars after they leave the main sequence. We find that stars hecome red giants after their core hydrogen is exhausted, and iir di.scuss the
thermonuclear process of helium burning at the centers of aging stars.

Throughout, we .see how H-R diagrams help
understand many
details of stellar evolution.
The legend of the Ephemera tells of a race of remarkable insects that

once inhabited a great forest. These noble creatines were blessed with
great intelligence, yet cursed with tragically
shon
life
spans of
less
than a
day.
To
the F.phcniera. the forest
seemed eternal and unchanging. Manv
generations lived out their briet lives without ever noticing anv alteration
in the surrounding foliage. Nevertheless, careful observation and reasoning led some Kphemera to postulate that the forest is not static. Thev
began to suspect that small gieen shoots grow to become huge trees and
that mature trees eventually die, tojiple ovei and litter the forest with
rotting logs, enriching the soil for future trees. Although they were un,
Till' lives
able to witness this transformation personally, the
of Stan
Ephemera became
aware of life cycles stretching over awesome periods of many years.

An analogous situation faces astronomers. Gazing across the galaxy,
we can see a vast array of stars and nebulae. The heavens seem eternal
and unchanging at first; what we see is virtually indistinguishable from
what our ancestors saw. This seeming permanence is an illusion, however. Stars emit vast amounts of radiation, which must produce changes
that cause the stars to evolve. For careful observation and calculation,
astronomers have assembled a theory of stellar evolution, which describes how stars are born in great interstellar clouds of gas and dust.
They mature, grow old, and eventually blow themselves apart in death
throes that enrich the interstellar gas for future stellar generations. The
stars seem unchanging only because major stages in their lives last for
millions or billions of years.
Protostars form
dark nebulae
Figure 13-1
in cold,
The Horsehead Nebula
[left]
from
Diisl giaiiis block the light
ground
nebulosity whose
excited b\ ultraviolet radiation
massive
stars.
The nebula
at a distance of roughly
Earth. The bright star
the back-
glowing gases are
is
from young,
located in Orion
1600
light years
to the left
from
of center
is
Alnitak, the easternmost star in the "belt" of
Orion. (Royal Observatuiy. Edinburgh)
Figure lB-2 [right]
A dark
nebula
dark nebula, called Barnard 86.

Sagittarius. It
simply because
beyond
it.
The
is
it
is
This
located in
visible in this jihotograph
blocks out light
from
the stars
cluster of bluish stars to the left
of the dark nebula
is
NGC
Australian Ob.sernatory)
6520. (Anglo-
For many years, astronomers have suspected that stars are born in cold,
dark clouds of interstellar gas because the temperature of a gas is directly related to the average speed of its atoms and molecules. If an interstellar cloud is warm, its atoms are moving about so rapidly that there
is no chance for a protostar to condense from the agitated gases. If the
cloud's temperature is low, however, its atoms are moving slowlv enough
to allow denser portions of the cloud to contract gravitationally into
clumps that collapse to form new stars.
Many of these cold clouds are scattered across the Milky Way. In
some cases, they appear as dark regions silhouetted against a glowing
backgroimd nebulosity, such as the famous Horsehead Nebula in Figure
13-1. In other cases, they appear as dark blobs that obscure the backgroimd stars (see Figure 13-2). These dark nebulae are often called Barnard objects, after the American astronomer E. E. Barnard, who first
discovered them aroimd 1900. Some 200 relatively small and round dark
nebidae are called Bok globules, after the Dutch-American astronomer
Bart Bok, who called attention to them in the 1940s.
A typical Bok globule measures 1 to 2 light years across and has a
mass between 20 and 200 solar masses. The chemical composition by
mass of these dark clouds is the standard "cosmic abundance" of about
75 percent hydrogen, 23 percent helium, and 2 percent heavier elements
The twn
uj stars
(recall
Table 6-4). Infrared observations demonstrate that the internal

from 5 to 15 K. This range is so cold thai
dense regions inside the globule conlraci giavitatioiiallv, and ilie glohiiii
gradually coalesces into lumps called protostars.
tcniperatitres of globules are
Protostars evolve into young
main-sequence stars
As early as the 1950s, astrophysicists began calculating the structure and

evolutionary history of protostars. At first, a protostai- is a cool blob of
gas several times larger than the solar system. Low pressures inside the
protostar are incapable of supporting all this cool gas, so the protostar
begins to contract. As it does so. gravitational energv is converted into
thermal energy, which causes the gases to heat up and start glowing.
Alter only a few thousand years of gravitational contraction, the surface
temperature has reached 2000 to 3000 R. At this point, the protostar is
still quite large, so its glowing gases produce substantial luminosity.
After
only 1000 years of contraction, a protostar of 1 solar mass would be 20
times larger in diameter and 100 times brighter than the Sun.
Astrophysicists use high-speed computers and the equations of stellar
structure described in Chapter 12 to calculate the conditions inside a
contracting protostar. The results tell how the protostar's luminosity and
surface temperature change at various stages during its contraction. With
we can graph the evolutionary track of the protostar

Hertzsprung-Russeil diagram.
information,
this
on
When a protostar begins to shine at \isible wavelengths, it is both

luminous and cool. Evolutionarv tracks of protostars thus begin near the
upper-right corner of the H-R diagram (see Figure 13-3). Continued
gravitational contraction causes the protostar to shift rapidiv
away from
region of the diagram, however. A protostar more massive than

three Suns becomes hotter without much change in overall luminosil\
The evolutionary tracks of massive protostars thus traverse the H-R diathis
gram
horizontally,
from
right to
left.
Less massive protostars
as they contract, so their luminosities decrease
temperatures
dim
slightlv
.somewhat as their surface
rise.
protostar continues to shrink uniil tiie temperature at its center

reaches a few million degrees, when Indrogen burning begins. As we
Chapter 12, this thernionuclear process releases enormous

amounts of energy. The outpouring of energy from hvdiogen burning
.saw in
creates conditions inside the protostar that finally halt
its
gravitational
and thermal e<iuilibrium are eventuallv estabis born. It is at this stage that the wandering
evolutionary track ends on the main sequence, as seen in Figure 13-3.
We now know that the main sequence represents relativelv voung
stars inside of which hydrogen burning has onh "recentlv" begun. This
state is a very stable one for most stars. For example, the Sun will remain on or near the main sequence, quietly burning hvdrogeii at its
contraction. Hydrostatic
lished
4(1. (HIO
.'(1.(10(1
Figure 13-3
1(1.000
5000
IViniKralurc (K)
-.
Pre-main-sequence
evolutionary tracks
The evolulmnan Imcks
of seven stars having different masses are
shown
thk
in
H-R
diagram. The dashed
lines
indicate the stage reached after the indicated
number of years of
tracks terminate
et'otutwn. Note that all
on the main sequence at
(mints agreeing with the massluminositv

relation.
lations
(Based on
Ij\
I.
Ihen)
.Uellur
model calcu-
and a stable
star
core, lor a total of 10 billion years.
Note that the evolutionary tracks in Figure 13-3 end at locations along
the main sequence that agree with the mass-luminosity relation (recall
Figure 1-15). The most massive stars are the most luminous, and the
least massive stars are the least luminous. Protostars less massive than
about 0.08 solar mas.ses never manage to develop the nccessarv pressures
1
and tem|HMatures
to start
hvdrogen burning
protostars instead contract to
beiome
at their cores. Fhese small

pl.uuilikc objects. On ilie oilier
Thr
lii'es
of stars
hand, protostars with masses greater than about 100 solar masses
(100 A/q) rapidly develop such extremely high temperatures that radiation pressure becomes the dominant force supporting the star against
Such stars are very unstable and very uncommon.

evolutionary tracks of protostars begin in the red-giant region of
the H-R diagram, but they are not red giants. .An H-R diagram such as
Figure 1 1-9 shows where stars spend mo.<it of their lives, and protostars
gravitational collapse.
The
a tiny fraction of their existence in the red-giant region. A

protostar takes only 10,000 years to become a main-sequence
spend only
15
Mq
star; a
ing at
its
that
is
it
Mq
protostar takes a few million years to ignite hvdrogen burnBy astronomical standards, these are such brief intervals
core.
most unusual
to discover a protostar that
is
in
its
earliest stages
of formation.
We
also are unlikely to observe the birth of a star at visible wave-
lengths because
surrounding globule or
pearance because of interstellar dust. (National

Aerospace Laboraton of the Netherlands:
cloud shields the

emitted bv the
protostar is absorbed by interstellar dust in the siurounding cocoon nebula, which becomes heated to a few iiundrcd kelvins. The wanned dust
reradiates the energy at infrared wa\elengths. Infrared observations can
thus reveal what is going on inside a "stellar nursery."
Figure 13-4 shows two views of a stellar nursery taken at infrared and
visible wavelengths. The dark cloud near the center of Figure 13-4i contains roughly 1000 solar masses of hydrogen and is a site of active star
formation. This same region appears very bright in the infrared view,
Figure 13-4a. Interstellar dust in the cloud absorbs nearly all the visible
light emitted by newborn stars, but infrared radiation passes easily
through the obscuring material. Recent infrared observations reveal a
cluster of 20 new stars embedded deep within the dust cloud. Astronomers are understandably excited by the ease with which infrared obser-
Rovil Obsen'aton-. Edinburgh)
vations probe these clouds.
Formation of new stars near
Figure 13-4
p Ophiuchi
(a) -4 u'ide-a)igle
infrared view
of the region near the star p Ophiuchi. covering nearly 13 X 13 at a wavelength of
100
fjL.
The while rectangle
indicates the
smaller region covered by photograph

ble wavelengths, (b)
The white
20 newborn
are probably
less
than
at visi-
circle indicates
the location of at leant
stars that
millon years old.
The
bluish star near the top of this view at visible
wavelengths
est star in
Many
is
p Ophiuchi.
Anleres, the bright-
Scorpius, appears at the lower
stars in this region
left.
have a fuzzy ap-
its
protostar from our view.
The
vast
interstellar
amount of
visible light
The
Young star clusters

regions
in H
of stars
lii'es
We can see from the evolutionary tracics of protostars in Figure 13-3 that
the most massive stars are the first to form. The more massive the prolostar. the sooner it (lf\elo|)s the necessary central pressures and temper-
are found
II
atures to ignite hydrogen hurning. These massive main-sequence stars
and B) are also the hottest and most luminous stars in

the sky. Their surface temperatures are typically 15.000 to 35,000 K.
(spectral types
meaning
that they emit vast (juantities of ultraviolet radiation.

energetic ultraviolet photons from newborn massive stars easily
ionize the surrounding hydrogen gas. For example, radiation from an
The
05
can knock electrons from hydrogen atoms in a volume roughly

This ionization has a dramatic effect on the globule or dark nebula in which a cluster of stars is forming. While some
hydrogen atoms are being knocked apart bv ultra\iolet photons, other
hvdrogen atoms are being reassembled as some of the free ])i()tons and
electrons manage to get back together. During this recombination of
hydrogen atoms, the captured electrons cascade downward through the
atom's energy levels toward the ground state. These downward quantum
jumps release numerous photons, many visible wavelengths. The nebula
star
1000
light vears across.
is the transition from h = 3 to

6563 A in the red portion of the
spectrum. The nebulosity around a newborn star cluster thus
begins to glow. Particularly prominent

n
2,
visible
which produces H photons
at
shines with a distinctive reddish hue.

Figure 13-5 shows one of these emission nebulae. Because these neb-
II reulae are predominantly ionized hvdrogen. thcN are also called

and B stars near the core of
The collection of a few hot. bright
gions.
the nel)ula that produces the ionizing ultraviolet radiation

OB association.
is
ailed
an
Observations of indixidual stars in a young cluster \icld finther information about stars in their infancy. Figure 13-6 shows a beautiful emission nebula surrounding the cluster called NGC: 2264. Bv observing each
star in the cluster and measuring its magnitude and coloi astronomers
,
Figure 13-5
An
H II
region
shape, ihis emission nebula
Xehula.
MI6
It
is
Hkciuv
o/
il^
called the F.af;le
surrounds the star cluster called
in the constellation oj Serpens at a dis-
tance of 65(H) light years from Earth. Sei'euil

blight, hot
and B
stars are responsible foi
the ionizing radiation that causes the gases to
glow. (Anglo-Australian Observatory)
The
Figure 13-6
H II
young star
lives
of stars
This
cluster
region in the constellation of Monoceros
NGC
contains a young
2264. It is located about 2600 light years
from Earth and contains numerous stars thai
star cluster called
are about
to
begin hydri)ge)i hiirnnii^ ni
llicii
(Anglo-Auslralian Observatory}
cores.
can deduce the
star's
luminosity and surface temperature.
The
data for
the stars in the cluster can then be plotted on an H-R diagram, as

shown in Figure 13-7. Note that the hottest stars, with surface temperatures around 20,000 K, are on the main sequence. These stars are the
all
whose radiation
is causing the surthan about 10,000 K have not

yet quite arrived at the main sequence, however. These less massive stars
are in the final stages of pre-main-sequence contraction and are just
now starting to have thermonuclear reactions at their centers. The locations of data points on Figure 13-7 suggest that the cluster has an age of
only about 2 million years.
Spectroscopic observations of the cooler stars in NGC 2264 show that
many are vigorously ejecting gas, a very common phenomenon in most
stars just before they reach the main sequence. Such gas-ejecting stars
are called T Tauri stars because the first example was discovered in the
constellation of Taurus. Some astronomers suggest that the onset of hy-
rapidly evolving, massive, hot stars
rounding gases
to glow.
The
stars cooler
preceded by vigorous chromospheric activity, with

and fiares that propel the star's outermost layers back
out into space. In fact, an infant star going through its T Tauri stage
can lose as much as 0.4 solar masses before it settles down on the main
drogen burning
enormous
is
spicules
sequence.
During these
final
spasmodic stages
40,000
20,000
10,000
Figure 13-7
2264 Each
An H-R diagram
dot plotted on this
represents a star in
nosity
and
NGC
2500
5000
- Temperature
(K)
of NGC
diagiam
HR
2264 whose
lumi-
surface temperature have been de-
termined. Note that most of the cool, low-mass

stars
have not yet arrived at
quence.
The
the
main
se-
star cluster probably started form-
ing only 2 million years ago. (Based on observations by Merle Walker)
in the birth
may become
of a
star,
material in
and start to glow.

Such activity is probably responsible for the so-called Herbig-Haro objects, named after astronomers George Herbig and Guillermo Haro who
began discovering them in the 1940s. As shown in Figure 13-8, HerbigHaro objects consist of several bright knots. From year to year, these
clumps of glowing gases change slightly in size, shape, and brightness.
These alterations are presumably caused by fiuctuations in nearby infant
stars as they undergo their final stages of pre main-sequence contraction. Herbig-Haro objects are indeed often found in the vicinity of T
the surrounding cocoon nebula
Tauri stars.
Sporadic
excited
activity and mass loss can continue after a star has arrived
on the main sequence. For example. Figure 13-9 shows a young star
cluster called the Pleiades in the constellation of Taurus that is easily visible to the unaided eye. All the stars in this picture are on the main se-
The
Figure 13-8
[left]
Heibig Hiiro
glowing gas
Herbig-Haro
objects
objects consist nj bright knots oj
that vary slightly over the years.
They probably are the
sites
of active star for-
mation and are typically found in the vicinity

of
Tauri
stars. (Lick
Figure 13-9 [rightj
410
Obsen'atory)
is
The Pleiades
This ope
light years
about 100 million years
old.
(U.S.
of stars
The cluster's age is presumably about 100

how long it takes for the least massive stars
quence.
million years because
that
finally to
is
gen burning
Naval
One
in their cores.
begin hydro-
of the brighter stars in the cluster
is
still
experiencing minor outbursts.
Note the
This glow
from Earth and can

easily be seen with the naked eye. The cluster
lias a diameter of about 7 light years and is
cluster
Iwe.s
is
distinctly bluish color of the nebulosity
called a reflection nebula because
it
is
around the
Pleiades.
caused by starlight
reflected from dust grains in the interstellar material. The nebulosity is

blue for the same reason Earth's daytime sky is blue: particles whether
in a nebula or in the air
scatter short-wavelength light much more effi-
wavelength radiation. Blue light is bounced around

and reflected back toward us much more readily than is light of other
ciently than longer
colors.
Obsen'atory)
In many respects, the Pleiades cluster is a mature version of the H II

Tauri activity
regions seen in Figures 13-5 and 13-6. Many years of
and vigorous stellar winds have widely dispersed the surrounding gases
so that they
tually
no longer glow
become
as
an emission nebula. These gases
will
even-
so widely scattered that the reflection nebulosity will also
fade from view.
A loose collection of stars such as the Pleiades or NGC 2264 is called

an open cluster or a galactic cluster. Such clusters possess barely
enough mass to hold themselves together by gravity. A star moving
faster than the average speed will occasionally escape, or "evaporate,"
from a cluster. By the time the stars are a few billion years old, thcv
may be so widely sejiarated that a duster no longei truly exists. If the
group of stars is gravitationally unboimd from the beginning that is, if
the stars are moving away from one another so rapidly that their gravitathe apparent cluster that initially
tional forces cannot pull them back
existed
Star birth begins
molecular clouds
in
giant
is
called
simph
a stellar association.
regions are stellar nurseries, but where do

This question was finally answered in the
197()s when radio astronomers Ijcgan discovering enormous douds of
gas scattered about our galaxy.
.AstrcMiomers agree that
the
II
regions
II
come from?
The
lives
of stars
As we saw in Chapter 6, hydrogen is by far the most abundant element in the universe. In the cold depths of interstellar space, hydrogen
atoms combine to form hydrogen molecules (Ho). A molecule vibrates
and rotates at specific frequencies dictated by the laws of quantum mechanics. As a molecule goes from one vibrational or rotational state to
another, it emits or absorbs a photon, just as an atom emits or absorbs a
photon when an electron jumps from one energy level to another. Many
molecules emit photons with wavelengths of a few millimerecent years, observations with radio telescopes tuned to wave-
interstellar
ters. In
lengths in this range have greatly increased our knowledge of the interstellar
medium, or
interstellar matter.
Although hydrogen molecules are scattered abundantly across space,

thev are difficult to detect. The hydrogen molecule consists of two atoms
of equal mass joined together, and such molecules do not efficiently emit
photons at radio frequencies. Radio astronomers can more easily detect
asymmetric molecules such as carbon monoxide (CO), which consists of
two atoms of unequal mass joined together. Carbon monoxide emits
photons at a wavelength of 2.6 mm, corresponding to a transition between two rates of rotation of the molecule.
The existence of carbon monoxide is especiallv useful in probing the
interstellar medium. Astronomers have reason to believe that the ratio of
hvdrogen to carbon monoxide is reasonably constant in space: about
lO.OOO Ho molecules for every CO molecule. Consequently, wherever
astronomers detect strong emission from CO, they know that hydrogen
gas must be abundant.
In mapping the locations of CO emission, astronomers soon realized
that vast amounts of hydrogen are concentrated in giant molecular
clouds. These clouds have masses in the range of 10' to 2 x 10'' solar
masses and diameters that range from 20 to 80 parsecs. Inside one of
these clouds, the density is about 200 hydrogen molecules per cubic centimeter. Astronomers estimate that our galaxy contains about 5000 of
these
enormous
Manv
clouds.
own and the one shown in Figure 13-10.

have spiral arms, which are huge, arching lanes of glowing gas and
stars. In Chapter 16, when we discuss our galaxy, we shall learn that
these spiral arms are caused by compression waves that squeeze the interstellar gases through which they pass. When one of these waves passes
through a giant molecular cloud, it compresses the cloud and vigorous
star formation begins in the densest regions. Massive stars, which are the
Figure 13-10
spiral galaxy
As compres-
sional waves pass through the molecular

clouds, they trigger vigorous star birth. This
process produces numerous
H II
regions
and
and B stars that outline the spiof

ral aims. Each pinkish speck along the spiral
clusters
arms of
this
galaxy
is
an
H II
region. (Anglo-
galaxies, including oiu"
Thf
The Orion Nebula
Figure 13-11
famous
naked
II region
can be seen wilh
e\e. 1 1 is located
1600
lives
247
of stars
This
llie
light years
from
Earth and has a diameter of roughh 16 light

years. The mass of this nelmlosity u estimated
300
Four
bright, massive stars
at the center of the nebula
produce the ultravi-
at
solar masses.
olet light that
four
causes the gases to glow. These
stars, called the
from each other
fry
Trapezium, are separated
only 0.13 light year.
(Anglo-Australian Obsen<ator\)
first to form, emit ultraviolet light that soon ionizes the surrounding
hydrogen and an H II region is born.
The Orion nebula (see Figure 13-11). like manv H II regions, is a
small, bright "hot spot" in an enormous molecular cloud. Four hot. massive O and B stars at the heart of the Orion nebula are responsible for
the ionizing radiation that causes the surrounding gases to glow.
OB
The
association at the core of the H II region affects the rest of
the giant molecular cloud. X'igorous stellar winds, along with ionizing
ultraviolet radiation from the
and B stars, carve out a cavitv in the
Much
of this outflow is supersonic, which creates a shock wave

where the outer edge of the expanding H II region impinges on the rest
of the giant molecular cloud. The shock wave compresses the hydrogen
gas through which it passes, stimulating a new round of star birth. Compact H II regions and infrared sources are often found immediatelv behind the shock wave. These new-born H II regions soon form their own
O and B stars, which then start carving out a new cavit\ deeper into
the giant molecular cloud. Meanwhile, the older O and B stars are left
behind and begin to disperse (Figure 13-12). In this wav. an OB association "eats into" a giant molecular cloud, "spitting out" stars in its wake.
Recent radio observations of giant molecular clouds suggest that it
takes only 10' years for one of our galaxy's spiral arms to pass through
cloud.
Older
subassociation
Figure 13-12
association
and B
Old
subassociation
The evolution of an OB
jrom young
produces a shock wave that
i'ltraviolel radiation
stars
compresses gas further into the molecular cloud
and
stimulates
the cloud.
new
star formation deeper into
Meanwhile, older
hind. (Adapted
stars are left be-
from C. Lada.
and B. Elmegreen)
L. Blitz,
Compact
II
region
and infrared sources
The
lives
of stars
molecular cloud. By that time, the bright, massive O and B stars

have rushed through their brief lives and faded from view. With no
more ultraviolet light to ionize hydrogen atoms, the beautiful emission
nebulae also fade and disappear. In other words, by the time a spiral
arm leaves a giant molecular cloud, all the dramatic activity associated
with star birth has come to an end.
The compression wave associated with a spiral arm continues to revolve about the galaxy, however, and soon it encounters new giant molecular clouds, which triggers a new round of star birth. We can see spiral arms because they are outlined by the glowing H II regions that
surround newborn O and B stars. In many galaxies, including our own,
star birth is an ongoing process intimately related to the majestic, pinwheel-like rotation of spiral arms.
a giant
Star birth
is
Presumably, any mechanism that compresses interstellar clouds can trigger the birth of stars. As we shall see in detail in Chapter 14, a massive
star can end its life with a violent detonation called a supernova explosion. In a matter of seconds, the doomed star is blown apart and its
outer layers are blasted outward into space at speeds of several thousand
kilometers per second.
also triggered
by supernova explosions
that compress the
interstellar
medium
Astronomers find many nebulosities across the sky that are the shredded funeral shrouds of these dead stars. Such nebulae (like the Cygnus
Loop shown in Figure 13-13) are called supernova remnants. Many su-
Figure 13-13
supernova remnant
This
remarkable nebula called the Cygtiui Loup

is
the
remnant of
ii
supernova explosion
that occurred almul 2(1.000 years ago.
expanding spherical
lias
shell
of gas
a diameter of about 120
(Palomar Observatory)
The
now
light years.
Till'
Inn
of slnrs
pernova remnants liave a distinctly arched appearance, as would be expected for an expanding shell of gas. This wall of gas is typically still
nio\ing away iroin the dead star at supersonic speeds, lis passage
through the surrounding interstellar medium excites the atoms, causing
the gases to glow.
Supersonic inotion is always accompanied by a shock wave that abruptly compresses the gas through which it passes. If the expanding
supernova remnant encounters an interstellar cloud, it can

squeeze the cloud, stimulating star birth. As we learned in Chapter 10,
there is strong evidence that the Sun was created in this fashion.
Our understanding of star birth has improved dramatically in recent
years, primarily because of infrared and millimeter-wavelength observations. Nevertheless, many mysteries remain. For example, astronomers
had generally assumed that there must be a lot of interstellar dust
shielding a stellar nursery to protect it from the disruptive effects of external sources of ultraviolet light, which is what we seem to find in our
own galaxy. However, in a neighboring galaxy called the Large Magelshell of a
Cloud there are yoimg OB associations with virtually no dust. Does

the process of star birth differ slightly from one galaxy to another?
lanic
Another problem is that different methods of star birth tend to produce different percentages of different kinds of stars. Specifically, the
passage of a spiral arm through a giant molecular cloud tends to produce an abundance of massive O and B stars. In contrast, the shock
wave from a supernova seems to produce fewer O and B stars and manv
more of the less massive A, F, G, and K stars. We do not know win this
is
so.
In spite of unanswered c]ucstions,
it
is
now
clear that star birth in-
mechanisms on a colossal scale that we have just begun to appreciate

from the deaths of massive stars to the rotation of the entire galaxy. The study of cold, dark stellar nurseries will certainly be an active
and exciting area of astronomical research for manv years to come.
volves
When
core hydrogen burning
ceases, a main-sequence star

becomes a red giant
main-sequence
star
is
young
star
whose radiated energy comes from
its core. A mainsequence star is in thermal equilibriuin, with energy liberated in its core
balanced by energy radiated from its surface. Eventualh however, all
the hydrogen in the core of the star will be used up. Core hydrogen
burning then must cease, with dramatic effects upon the star's eiiuilib-
the thermonuclear process of hydrogen burning in
liiuii,
We
structure,
and evolution.
(]uicklv builds up the necessary

core tem|)eratine and pressure for hydiogen burning to comnience, so
that the most massive protostars are the fust to become main-sequence
have seen that a massive protoslar
(O and B) stars. Furthermore, the more massive main-sec|uence stars are

the most luminous stars; this rapid emission of energy must correspond
to a rapid depletion of hydrogen in the core of the massive stars. Fhus,
even though a massive O or B star contains much more hvdrogen fuel
than a less massive main-sequence star, it consumes its hydrogen far
inore rapidly. The main-sequence lifetime of a massive star thus is considerably shorter than that of a less massive star.
Hydrogen binning has continued
in the Sun's core for the past 5 bilchemical com|)osilion was loughlv 75 percent hydrogen and 2.") percent helium (plus a smatteiing of heavv elements). The continued fusion of hvdrogen into helium in the Sun's core,
however, has dramatically altered the core composition. Indeed, there is
now more helium than hydrogen at the Sun's center.
lion years. Initially, the Sun's
The
TABLE
13-1
lives
of stars
Main-sequence lifetimes
Time on
Surface
Mass
(Mo)
temperature
Luminosity
main sequence
(K)
(io)
(10* years)
25
35,000
80,000
15
30,000
10,000
15
11,000
60
500
1.5
7,000
3,000
10,000
1.0
6,000
0.75
5,000
0.5
0.50
4,000
0.03
Enough hydrogen remains
in the Sun's core for
15,000
200,000
another 5
billion
years of core hydrogen burning. That brings the Sun's total lifetime
the main sequence to 10 billion years. Table 13-1 shows
on
how long other
exhaust the supplies of hydrogen in their cores. Note that

high-mass stars gobble up their hydrogen fuel in only a few million
years, whereas low-mass stars take hundreds of billions of years to accomplish the same thing.
As the supply of hydrogen at a star's center dwindles, the star begins
to have difficulty supporting the weight of its outer layers. This enormous weight pressing inward from all sides compresses the star's core
slightly. The compressed gases become warmer, allowing hydrogen burning to move outward from the core. In other words, during its final
years on the main sequence a star makes a final attempt to maintain
thermal equilibrium by enlarging its hydrogen-burning region. There is
still plenty of fresh hydrogen surroimding the star's center. By tapping
this supply, the star manages to eke out a few million more years on the
main sequence.
Finally, all the hydrogen in the core of an aging main- sequence star
is used up, and hydrogen burning ceases in the core. Hydrogen burning
does continue, however, in a thin spherical shell surrounding the core.
This shell hydrogen burning initially occurs only in the hot region just
outside the core, where the hydrogen fuel has not yet been exhausted.
Because no thermonuclear reactions are now producing energy in the
core, the heat flowing out of it is no longer replaced. The core gradually
contracts, converting gravitational energy into thermal energy to maintain thermal equilibrium. As the hydrogen-burning shell slowly works its
way outward from the original core, more helium is added to the core,
causing further core contraction and heating.
As the core shrinks and becomes hotter, its increased energy output
causes the star's outer layers to expand, increasing the overall size of the
star. The diameter of the star may increase by a factor of ten. And as
the star's outer atmosphere expands farther and farther into space, its
gases cool. Soon the temperature of the star's bloated surface has fallen
to about 3500 K and its gases glow with a reddish hue. Such stars are
appropriately called red giants.
In the Sun's case, it will take about 5 billion years more for it to finish
converting hydrogen into heliimi at its core. As the Sun's core contracts,
its atmosphere will expand to encompass Mercury, then Venus, and finally our own planet. The red-giant Sun will swell to a diameter of 2
AU, and its surface temperature will decline. As a full-fledged red giant
stars take to
Thf
Figure 13-14
giant
Today,
of stars
251
The Sun today and as a red

llie
Suti's etiergy
is
pruducai
a hydrogen-hunung core whose dtameler
about 300.000
lives
km When
the
iii
Sun becomes a
it will draw
red giant, in some 5 billion years,

its
energy from a hydrogen-burning shell
rounding a compact helium-rich
core.
.<,ui
The
helium core will have a diameter of onl\

10,000 km.
(see Figure 13-14), our star will shine with the brightness of a hundred
Suns. While the inner planets are vaporized, the thick atmospheres of
the outer planets will boil away to reveal tiny, rockv cores. Thus, in its
later years the
it
Helium burning begins at the

center of a red giant
since
its
aging Sun
will
destroy the planets that have accompanied
birth.
Helium is the "ash" of hydrogen burning. When a star first becomes a

red giant, its hydrogen-burning shell surrounds a small, compact core of
almost pure helium. In a moderately low-mass red giant (like the Sun 5
billion years from now), the dense helium core is about the same size as
the Earth and the star's bloated surface has roughly the same diameter
as the Earth's orbit.
At
first,
giant.
star,
no thermcjnuclcar reactions occur
The hydrogen-burning
adding mass
to the
slicll
in the
continues to
helium core of a red
move outward
in the
helium core, which slowlv contracts, thereby
forcing the star's central tern|)erature to climb. Kinallv,
when
the central
temperature reaches 100 million kelvins, helium burning is ignited at

the star's center. This new thermonuclear re.u tion occurs in two siei)s.
First, two helium nuclei combine. Then a thiid helium nucleus is added
to this combination, resulting in a carbon luicleus:
3" He
'-c
gamma-ray photon (y). Some of the carbon created

process can fuse with an additional helium nucleus to produce
with the release of a

in this
oxygen:
'-C
-t-
^He -> "'O + y
Thus, both carbon and oxygen make up the "ash
"
ol
helium burning.
The
The
lives
uf stars
and oxygen by helium burning releases

time since leaving the main sequence, the
aging star again has a central energy source. This source, properly called
core helium burning because of its central location, establishes thermal
ec)uilibrium, thereby preventing any further gravitational contraction of
the star's core. A mature red giant burns helium in its core for about 5
to 20 percent as long as the time it spent burning hydrogen as a mainsequence star. For example, in the distant future the Sun will consume
helium in its core for about 1 billion years.
The way in which heliimi burning begins at a red giant's center depends on the mass of the star. For high-mass stars (those with masses
greater than 2 Mr,), helium burning begins gradualh' as temperatures in
the star's core finally reach 100 million degrees. In low-mass (less than
2 Mq) stars, however, helium bin ning begins explosively and suddenly,
an event called the helium flash. The helium fiash occurs because of
unusual conditions that develop in the core of a low-mass star on its way
to becoming a red giant.
gamma
creation of carbon
rays.
For the
first
way most gases do. If the gas is

the gas expands, it cools down. This behavior
serves as a "safety valve," to ensure that the star does not explode. For
example, if energy production overheats the star's core, the core exUsually, the gas inside a star acts the
compressed,
it
heats up;
if
pands, cooling the gases and slowing the rate of thermonuclear reactions. If too little energy is being created to support the star's overlying
layers, the core becomes compressed and the increased temperatines
speed up the thermonuclear reactions to increase the energy output.

In a low-mass red giant, the core must undergo considerable gravitational compression to drive temperatures high enough to ignite helium
burning. At the extreme pressures and temperatures deep inside the
star, atoms are completely torn apart into nuclei and electrons. As a result, under these conditions, the usual "safety valve" between gas pressure, temperature, and density does not operate. Instead, the electrons
are so closely crowded together that they obey a law of quantum mechanics called the Pauli exclusion principle. This principle, formulated
in 1925 by the Austrian physicist Wolfgang Pauli, is analogous to the
idea that you can't have two things in the same place at the same time.
Further compression of the star's core would be like trying to squeeze
one electron inside another.

Just before the onset of helium burning, the electrons in the core of a
low-mass star are so closely crowded together that any further compression would violate the Pauli exclusion principle. Because the electrons
cannot be squeezed any closer together, they produce a powerful presresists finther core contraction.
This phenomenon, whereby closely packed particles resist compression
in accord with the Pauli exclusion principle, is called degeneracy. Astronomers say that the heliiun-rich core of a low-mass red giant is degenerate and is supported by degenerate-electron pressure.
When the temperature in the core of a low-mass red giant reaches the
high level required for helium burning, gamma-ray photons begin to be
released by this reaction. This release of energy heats the core, but since
the usual "safety valve" is not operating, the core does not expand. Instead, the increased temperature causes more helium to burn which
sure that
The result is a run-away process called the

however, the temperature becomes so high that degeneracy is removed and the star's core expands, thereby terminating
the helium flash. These events occur extremelv rapidly and the helium
flash is over in only a few seconds.
heats the core
helium
still
further.
flash. Soon,
The
Evolutionary tracks on the

reveal the ages
"
-t>
'
'
i
t
"
\__^^^
>^\ ^-\
Zero- age
main sequence "''^VN^
'J
>>
hvdroiJen burning
Figure 13-15
evolution
5000
10,000
Ibm
Temperature (K)
Post main-sequence
The n'olulionary
are shoivii on
mass
20,000
-
HR
this
tracks of six stars

diagram. In the high-
helium burning ignites whrre
stars, core
the n'olutionnrs tracks
make a sharp tuni in

The ei'uA/q and
the red-giant region of the diagram.
lutionary tracks for low-ma.ss stars (1
2 Mcj) are shown

the helium
(Adapted from
Figure 13-16
only
up
to the points
I.
Ilien)
globular cluster
lar clutter
contains a
Jew hundred thousand
years
glohu
spherical clu.\ler that typically
particular cluster, call .Ml

con.stellation /
where
occurs at their centers.
/7n.i/i
3,
stars.
This
located in the
Hercules roughly 25.01)0 light
from Earth.
enlightening to follow the post-niain-scqiicncc evolution of mature

on a Hert/sprnng-Riissell diagram (see Figme
3), The zero-age main sequence (or ZAMS) is the
location on the ii-R diagram where stars In st begin core Indrogen
i)in'ning. In snbsetjiient vears. the evoliiiionarv tracks sUjwIv inch awav
Irom the ZAMS as the hydrogen-binning core grows in search of fresh
fuel. Ihe dashed line on Figure 13-15 shows the locations of the stellar
models when all the core hydrogen has been consumed.
After the cessation of core hydrogen burning, the points representing
stars move rapidly toward the red giant region of the H-R diagram.
During this transition, the star's core contracts, while its outer layers expand in response to the increased energy output of the star's hydrogenburning shell. In this way, the stars become red giants.
Soon alter a star becomes a red giant, helium burning begins in its
core. The evolutionary track of a high-mass star then wanders back and
forth in the red-giant region as the star adjusts to its new energy soince.
The e\'olutionar\ tracks of two low-mass stars also appear in Figure 13I."). However, these two tracks are shown onlv to the point where the
helium flash occurs. Although the helium flash releases a sudden flood
ol energy, the effects of this outburst take thousands of vears to reach
the star's surface.
Examples of red gianis and post-helium-flash stars are found in old
star clusters, called globular clusters because of their spherical shape. .\
typical globular cluster, like the one shown in Figure 13- Hi, contains up
to I million stars in a volume less than 100 parsecs across. .Astronomers
know that such clusters are old because thev contain no high-mass mainsequence stars. If you measure the luminositv and sinface temperatiue
of many stars in a globular cluster and plot the data on an H-R diagram as shown in Figure 13-17, you find that the upper half of the
main sequence is missing. All the high-mass main-.sequence stars have
evolved long ago into red giants, leaving behind onlv low-mass, slowlv
evolving stars that still have core hvdrogen burning.
Fhe H-R diagram of a globular cluster typicallv shows a horizontal
grouping of stars in the upper-left to upper-center portion of the dia-
KM
10^
40,000
253
stars b\ plotting their cvoliiiionar\ tracks
of star clusters
\(\
of stars
II is
H-R diagram
\{^\
lives
(i'.S.
S'aval Ohsenuttoryt
The
gram. As shown
lives
iij
stars
Figure 13-17, these stars form a horizontal row at

Lq to 100 Lq. Called horizontal-branch
stars, they are believed to be post-helium-flash low-inass stars. In years
to come, these stars will move back toward the red-giant region as both
core helium burning and shell hydrogen burning continue to devour
in
luminosities in the range of 50
ed
fuel.
An H-R diagram
of a cluster can be used to determine the age of the

diagram for a very young cluster (recall Figure 13-7), the
entire main sequence is intact. As a cluster gets older, however, stars
begin to leave the main sequence. The high-mass, high-luminosity stars
are the first to becoine red giants, and thus the main sequence starts to
burn down like a candle. As the years pass, the main sequence gets
shorter and shorter. The top of the surviving portion of the main sequence is called the turnoff point. Stars at the turnoff point are just now
exhausting the hydrogen in their cores, and their main-sequence lifetime
(recall Table 13-1) is equal to the age of the cluster. For example, in the
case of the cluster M3 (see Figure 13-17), 0.8 Mq stars have just left the
main sequence, so the cluster's age is roughly 15 billion years.
Data for several star clusters are plotted on Figure 13-18. All the
young clusters (those with their main sequences still intact) are open
clusters in the plane of our galaxy, where star formation is an ongoing
process. Stars in these young clusters are said to be metal-rich because
their spectra contain many prominent spectral lines of heavy eleinents.
This material came from dead stars that long ago exploded, enriching
the interstellar gases with the heavy elements formed in their cores.
Young, metal-rich stars are commonly called population I stars. The Sun
cluster. In the
Turn off
puuil
/-;'V^*-
Main
sequence
8000
10.111)0
(iOOO
(K)
Tcnipei.j
Figure 13-17
An H-R diagram
globular cluster
M3
Each dot on
of the
graph
this
represents a star in the globular cluster
M3
whose luminosity and surface temperature have

been determined. Note that the upper half of
the
main sequence
is
missing.
branch stars are believed
to
The horizontal
be low-mass stars
that recently experienced the helium flash in

their cores
ing and
from H.
and now
L.
exhibit core
helium burn-
hydrogen burning. (.Adapted
shell
Johnson and A. R. Sandage)
is
Figure 13-18
The shaded bands
composite
H-R
star.
diagram
indicate where data from
various star clusters Jail on the
The age of a
such a
cluster
H-R
diagram.
can be estimated from the
location of the turnoff point where the cluster's
most massive stars are just leaving the main
sequence for the red-giant region.
40.000 20,000
10,000
5000
2.500
Temperature (K)
Most of the oldest clusters are globular clusters. Globular clusters are
generally located outside the plane of our galaxy. Their spectra showonly weak lines of heavy elements. These ancient stars are therefore said
to be
that
They were created long ago from interstellar gases

been substantially enriched with heavy elements. Old,
are also called population II stars.
metal-poor.
had not
yet
metal-poor stars
The
Red giants
mass loss
typically
show
lives
of stars
Rccl-giain stars are so ciiornioiis that iluir bloated outer layers constantly
mass loss is tiiiite significant (see Fig-
leak gases into space. At times, this
ure 13-19).
Mass loss can be detected spectroscopically. Escaping gases coming

toward us exhibit narrow absorption lines that are slightly blueshifted.
According to the Dopjiler effect, this small shift toward shorter wayelengths corresponds to a speed of 10 km/sec. This yalue is typical of the
expansion yelocities with which gases leayc the tenuous outer layers of
solar
red giants. A typical mass-loss rate for a red giant is roughly 10
masses per year.
Betelgeuse in the constellation of Orion is a good example of a red
giant experiencing mass loss. Betelgeuse is 470 light years away and has
a diameter roughly equal to the diameter of Mars's orbit. Recent spectroscopic observations show that this star is losing mass at the rate of 1.7 x
'
10~' solar masses per year and is surrounded by a huge circumstellar

is expanding at 10 km/sec. These escaping gases ha\c been
detected out to distances of 10,000 AV from the star. Consequently, the
shell that
circuinstellar shell has an overall diameter of 3 light year.

Supergiant stars (that is, stars brighter than 10' Suns) are involved in
mass loss throughout most of their existence and have mass-loss rates
comparable to those of red giants. In the next chapter, we will see that
dying stars eject vast quantities of material into space. Nevertheless, mass
loss from supergiants and red giants accounts for roughly one-fifth of all
the matter returned by stars to the interstellar medium.
expanding
Figure
1319
mass-loss star
Old
become red giatib whose bloated outer
pheres shed matter into space. This star
ing matter at a high rate and
is
^/</'>
attnoiis los-
surrounded by
a reflection nebula caused by starlight reflected
from dust grains. These dust grains may have
condensed from material shed by the

typical red giant
per year, and
can
many
lose
star.
are .surrounded by
cir-
cum.\tellar shells of matter they liave shed.
{Anglo-Australian Observatory)
solar massr^
The Iwes of slars
Many mature
After core helium burning begins, mature stars wander across the middle of the H-R diagram. Figure 13-15 shows the evolutionary tracks of
high-mass stars crisscrossing the H-R diagram. Post-helium-flash low-
stars pulsate
mass stars on the horizontal branch also cross the middle of the HR
diagram as they return to the red-giant region.
During these transitions across the HR diagram, a star can become
unstable and pulsate. In fact, there is a region on the H-R diagram between the main sequence and the red-giant branch that is called the instability strip (see Figure 13-20). When a star passes through this region
on its evolutionary track, the star pulsates. As it pulsates, its brightness
10*^
Instability
varies periodically.
10^
10'^
Cepheid
variables^
/
/
j__
The
first
pulsating variable was discovered in 1784 by
Cephei, regularly varies

period of 5.4 days. It is
/ Lyrae
/ -y-RR
John Good-
amateur astronomer. This star, S

in apparent magnitude from 4.4 to 3.7, with a
the prototype of an important class of pulsating
ricke, a nineteen-year-old English
stars called
Cepheid variables, or simply Cepheids.
A Cepheid
'^
variable is recognizable by the characteristic way in which

output varies: rapid brightening followed by gradual dimming.
This behavior is most easily displayed in the form of a light curve,
MainX
which is a graph of a star's brightness plotted against time. The light
sequence
curve of 5 Cephei is shown in Figure 13-21.
L
A Cepheid variable brightens and fades because of a cyclic expansion
and contraction of the whole star. This behavior is deducible from spec10-2
^
troscopic observations. Spectral lines in the spectrum of 6 Cephei shift
40.000 20,000
10,000
5000
250
back and forth with the same 5.4-day period as that of the magnitude
-
Temperature (K)
variations. According to the Doppler effect, these shifts mean that the
Figure 13-20 The instability strip The
star's surface is alternately approaching and receding from us.
o
c
X*-'
its
instabitily strip uccupies
main sequence and

diagram.
region along
unstable
light
HR
variables
its
and
a region between the
the red-giant
branch on the
When
Cepheid variable pulsates, the star's surface oscillates up and

During these cyclical expansions and contractions,
the star's gases alternately heat up and cool down. The combined effects
of both changing size and surface temperature produce the characteristic
light curve of a Cepheid variable, such as that shown in Figure 13-21.
Cepheid variables are very important to astronomers because there is
a direct relationship between a Cepheid's period and its average luminosity. Dim Cepheid variables pulsate rapidly, have periods of 1 to 2 days,
and have an average brightness of a few hundred Suns. The most luminous Cepheids are the slowest variables, with have periods of 100 days
and average brightnesses equal to 10,000 Suns. This connection between
period and brightness is called the period luminosity relation. As we
shall see in Chapter 17, this relationship played an important role in determining the overall size and structure of the universe.
Details of a Cepheid's pulsation depend on the abundance of heavy
elements in its atmosphere. The average luminosity of metal-rich Cephe-
down
star passing through this
evolutionaiy track becomes
pulsates.
like a spring.
is roughly four times greater than the average limiinosity of metalpoor Cepheids having the same period. Thus there are two classes: Type
I Cepheids, which are the brighter, population I, metal-rich stars; and
Type II Cepheids, which are the dimmer, population II, metal-poor
stars. The period-luminosity relation for both types of variables is shown
ids
T.mc
(davs)
Figure 13-22.
evolutionary tracks of mature, high-mass stars pass back and
forth through the upper end of the instability strip on the HR
diagram. These stars become Cepheid variables. Low-mass, post
helium-flash stars pass through the lower end of the instability strip as
they move in the horizontal branch along their evolutionary tracks.
These stars become RR Lyrae variables, named after the prototype in
in
Figure 13-21
The
Cepheid variable
cun'e
oj
light curve
of a
This graph shows the light
8 Cephei. which
is
important class of variable
the prototype of
stars.
an
All Cepheid
variables exhibit the same type of periodic

changes in luminosity as shown here: rapid
brightening followed by gradual dimming.
The
Tile
lii'ei
of
stars
RR Lyrac variables all have periods shorter

have roughly the same average brightness as stars
the conslellaiion of Lyra.
than one day and
all
on the horizontal branch.

Stellar pulsations can in rare cases be quite substantial. Sonietimes the
expansion velocity exceeds the star's escape velocity. ,so the star's outer
layers are ejected completely, .^s
we
shall see in the
next chapter, signifi-
cant mass ejection accompanies the death of stars in a sometimes violent

process that renews and enriches the interstellar medium for future generations of stars.
Variable stars have also played a crucial role in the development of
modern astronomv.
used
RR
In
Chapter 16 we
Lyrae variables
in
1917
in a
shall see that
Harlow Shapley
bold attempt to determine our
lo-
Chapter 17 we shall learn how Edwin Hubble used Cepheid variables in 1924 to determine the distances to nearby
galaxies. Even today the period-luminosity relation is one of the most
important and reliable tools that astronomers have for measuring intergalactic distances. Such measurements are a first step in determining the
age and size of the observable universe.
cation within our galaxy. In
\\;nn\
Figure 13-22
relation
(.l,,vs)
The period luminosity
The period of a Cepheid variable n
directly related to
Metal-rich (Type
its
I)
average luminosity.
Cepheicb are brighter than
the metal-poor (Tipe II) Cepheids.
Summary
Enormous cold clouds of gas, called gianl molecular clouds, are scattered
about our galaxy. Star formation begins wfien gravitational allradion
causes a protostar to coalesce within a giant molecular cloud.
As a protostar contracts, its gases begin to glow. When its core temperature becomes high enough to begin hvdrogen burning, the protostar becomes a main-sequence star.
The most
massive protoslars are the
first to
bec(jme main-sequence stars
(O and B stars). They emit strong ultraviolet radiation that ionizes hydrogen in the surrounding cloud, creating reddish emission nebulae called H
II
regions.
and stellar winds from the OB association at the core

region create shock waves that compresses the gas cloud to
trigger formation of more protostars. Supernova explosions also compress
gas clouds and trigger star formation.
Ultraviolet radiation
of an
II
In the final stages of pre-main-sequeiue contraction, as thermonuclear

reactions are about to begin in the core of a protostar. the star mav undergo vigorous chromospheric activilv that ejects large amounts of matter
into space.
Such gas-ejecting
collection of
stars are called
f lauri stars.
newborn stars is called an open cluster, or galactic cluster.

moving star will escajie, or "'evaporate," from such
Occasionally, a rapidiv
a cluster.
massive a star, the shorter its main-sequence lifetime. The Sun

has been a iiiain-se(]uence star for about 5 billion vears and should remain
billion vears.
so for about another
The more
;")
Core hydrogen burning ceases when indrogen is exhausted in liie core

a main-sequence star, leaving a core of nearlv pure iieliiim suirnuiuled
a shell where hydrogen burning continues.
When
the central temperature of a red giant reaches about 100 million

thermonuclear process of helium burning begins there. I his
kelvins. the
process converts helium to caibon and oxvgen.
more massive red gi.iiit. In Imni liuining Ix'gins giadu.ilK. In

massive red gianl. it begins suddiiiK in .i process called a luiium
In a
of
h\
.i
less
ll.isli.
Thf
lives oj ilars
The age of
a stellar cluster can be estimated by plotting its stars on an

diagram. The upper portion of the main sequence will be missing
because more massive main-sequence stars have become red giants.
H R
young population
Relatively
stars are metal-rich; ancient
population
II
stars are metal-poor.
Red
undergo extensive mass loss, sometimes producing circumstelof ejected material around the stars.
giants
lar shells
When
a star's evolutionary track carries it through a region called the inthe

diagram, the star becomes unstable and begins
stability strip in
H-R
to pulsate.
Cepheid variables are high-mass pulsating variables having a regular relationship between period of pulsation and luminosity. RR Lyrae variables
are low-mass pulsating variables with short periods.
Review questions
Why
are low temperatures necessary for protostars to form inside dark
nebulae?
2 Explain why thermonuclear reactions occiu' only
sequence star and never on its surface.
What
the
an "evolutionary track" and

diagram?
is
H-R
4 Why do you suppose the

main-sequence stars?
5
On
in
at the
what way can
vast majority of the stars
it
we
center of a main-
help us interpret
see in the sky are
what ground are astronomers able to sav that the Sun has about
remaining in its main-sequence stage?
5 bil-
lion years
What does
it
mean when an astronomer

H R diagram?
says that a star
"moves" from one
place to another on an
7
What
are Cepheid variables and
how
are they related to the instability
strip?
8 Suppose that an oxygen nucleus were fused with a helium nucleus. What
element would be formed? Look up the relative abundance of this element
and cominent on whether such a process is likely.
Advanced questions
9 If you took a spectrum of a reflection nebula, would you see absorption

emission lines, or no lines? Explain your answer and explain how the
spectrum demonstrates that the light is reflected from nearby stars.
lines,
10 Speculate on why a shock wave from a supernova seems to produce relafew high-mass O and B stars compared to the lower-mass A, F, G, and
tively
stars.
11
How
would you distinguish a newly formed protostar from a red giant

H R diagram?
in
view of their identical location on the
12
What do you
think would
giant molecular cloud?
happen
Do you
if the solar system passed through a

think that the Earth has ever passed through
such clouds?
13 Speculate about the possibility ol life forms and biological processes occurring in giant molecular clouds. In what ways might conditions in giant molecular clouds favor or hinder biological evolution?
'
The
For further reading
lives of stars
259
'
'
"Giant Molecula.-(iloud Co,,,plcxcs in .he Galaxv/- ^VW/,/;;

American. Apr. 1982.
Bok, B. "Early Phases of .Star Formation." Sky is
Telescope. .Apr. 1981.
Blitz. L.
p.
284.
Cohen, M. "Stellar Formation." Astronomy, Sept. 1979.

p. 66.
Johnson, B. "Red Giant Stars." Astronomy. Dec. 1976.
p. 26.
Kaufmann, W. Stars ami Xebulas. W. H. Freeman and Company,
1979.
Lada, C. "Energetic Outnows from Young Stars."
Scientific American lulv
'
1982.
Reipurth, B. "Bok Globules." Mercuiy. Mar./Apr.

1984,
Robinson, L. "Orion's Stellar Nursery." Sky &= Telescope
Wyckoff,
S.
p. 50,
Nov
198''*
"Red Giants: The Inside Scoop." Mercun. Jan./Feb. 1979.

p
'
7.
Zeilik.
M. "The Birth
'
of Massive Stars." Scientific American, Apr. 1978.
'
14 The deaths of stars
Dying
planetary nebula
their outer layers.
half
its
stars often ejecl
low-mass star can
lose
mass in a comparatively gentle process
that produces a planetary nebula.
of about 100.000
the size of the
and
is
roughly one-tenth
Sun. Ultraviolet radiation from
the hot stellar core causes the
gases to glow.
The exposed
has a surface temperature
stellar core typically
The greenish
surrounding
color of this plane
The
taiy nebula
comes from oxygen
tral star in
thu nebula has exhausted
nuclear fuel and
is
contracting
ions.
to
all
cenits
become a
white dwarf. (Lick Observatory)
After
its
old age as a red giant, a star approaches the end of
chapter, ive learn that stars eject significant
its life.
amounts of matter
In
this
into space as
they die. We find that low-mass stars eject their outer layers relatively gently,
producing planetaiy nebulae, whereas high-mass stars explode violently as
supernovae. The core of a dead low-mass star contracts to become a white
dwarf, whereas the corpse of a high-mass star can become a neutron star. In
examining these white dwarfs and neutron stars, we discover that there are
upper limits to the masses that these stellar corpses can have. Finally, we discuss the various fascinating forms in which neutron stars have been observed,
such as pulsars, pulsating X-ray sources, and bursters.
From infancy through adulthood, a star leads a fairly placid life with
hydrogen burning in its core. As old age approaches, however, the star
takes on a schizophrenic character with a compressed core and a bloated
atmosphere, and becomes a red giant. Its erratic and fitful behavior becomes even more pronounced as the star devours its remaining nuclear
fuels and begins to die.
The deaths
Low-mass
stars die by gently
oj ilar
chapter, carbon and oxvgcn are the "ashes" ol hcAiler the helium flash in a low-mass led jîant. substantial
amoimts ol these two elements betîn to accumulate at the star's center as
a lesult ol core helium burning. F.ventuallv. all the helium at the center
ot a low-mass star is used up and coi e helium l)urning ceases. Bv this
lime, thermonuclear reactions have also ceased in the hydrogen-burning
As wc saw
ejecting their outer layers,
in the last
liinn Ijurning.
creating planetary nebulae
shell.
As
a result, the star's core begins to contract.
This compression
carbon-oxygen core, and

around the core. This process
heals the helium-rich gases suriounding ihe
soon heliimi burning begins in a thin shell

is called shell helium burning. The star's internal structure now consists
of a helium-burning shell inside the former hydrogen-burning shell, all
within a volume roughlv the size of the Earth (see Figure 14-1).
When shell hydrogen burning first began inside this star, the outpouring of energv caused it to expand and become a red giant. Astronomers sav that the star ascended the red-giant branch on the H-R diagram. Then came the heliiuti Hash, and the star shifted over to the
horizontal branch. But now. with shell helium burning, a renewed outpouring of energy causes the star to expand again. It ascends the redgiant branch for a second and final lime to become a red supergiant.
Such stars typically have diameters as big as the orbit of .Mars and shine
with the brightness of 10.000 Suns.
ergy loss cannot live much longer.
The
star with this furious rate of en-
is signaled by inslabililies that develop in

with the helium flash discussed in the previis again a sudden outpouring of energv. but the details are quite different. The helium flash occurred because the star's
core was degenerate. However, the helium-burning shell is not compressed to a high enough densitv to be degenerate. Instead, a heliumshell flash occurs because the shell is thin. The flash ends only after the
star's
impending death
helium-burning
ous chapter, there
its
helium-burning
star's
shell. -As
shell
becomes
thick
enough
to relieve the pressure
of the
outer layers.
During the flash, the helium-shell's energy output jumps from 100
Sims lo roughlv 100. 000 Suns in a rapid series of brief bursts called
thermal pulses, fhese bursts are separated by relatively quiet intervals
lasting about ;M)0.000 years.
During one of these periods of thermal oscillation, the dying star's

outer layers can separate completely from the carbon-oxygen core. .As
the ejected material expands into si)a(e. dust grains condense out ol the
Figure 14-1
mass star
The structure of an old low-
\iat
Ihe
end uf
ils
Itje.
a low-
nuns star becomes a red superânl, with n

diameter almost
nihil /
tmrnnig
a.\
Mrns. The
shell
and
large
star's
n.(
the diameter uf tin
dormant hydrogen-
active heliiim-lmrntng shell
are contained within a volume roughly the

of the Earth.
size
The dealhi of
The planetary nebula
Figure 14-2
7293
This beautiful
star.
NGC
object, often called the
Helix Nebula, covers an area of the sky equal

to
half the full moon. The star that ejected
these gases
shell.
is
seen at the center of the glowini;
The greenish
nitrogen
from oxygen
color comes
pink and red comes from
ions, while the
and hydrogen
ions.
This nebula
in the constellation of Aquarius, about

tight years
is
400
from Earth. (Anglo-
The planetary nebula
Figure 14-3
6302
At
the
end
NGC
of their lives, low-mass slurs
shed much of their mass. Although most cases

involve a fairly symmetrical ejection of gases,
there are
many examples
expanded unevenly.
nebula such as
here)
NGC
is
An
NGC
in
which the gas has
irregular planetary
6302
in
Scorpius (shown
the result. Spectroscopic observatioiu of
6302
indicate that the gases are
moving
toward us at 400 km/sec, indicating a ptnlicnlarly violent initial ejection.
(Anglo-
cooling gases. Radiation pressure from the star's hot, burned-out core
acts on the specks of dust to continue propelling them outward, and the
star sheds its outer layers altogether. A star can lose more than one-half
of
its
As
mass in
a dying
this fashion.
star ejects
its
ultraviolet radiation intense
outer layers,
enough
its
hot core
to ionize the
ejected gases. This ionization causes the gases to
is
exposed, emitting
expanding shell of
glow, producing a so-
called planetary nebula, or "planetary." Planetary nebulae

to
do with
planets.
have nothing
This unfortunate term was introduced in the eight-
eenth centiny when these glowing objects were thought to look like distant planets when viewed through small telescopes.
Many planetary nebulae, such as the one shown in Figure 14-2, have
a distinctly spherical appearance from the symmetrical way in which the
gases were ejected. In other cases, as in Figure 14-3, the rate of expansion is not the same in all directions, and the resulting nebula takes on
an hourglass or dumbbell appearance.
Planetary nebulae are quite common: astronomers estimate there to
be 20,000 to 50,000 in our galaxy alone. Spectroscopic observations of
these nebulae show bright emission lines of ionized hydrogen, oxygen,
and nitrogen. From the Doppler shifts of these lines, astronomers conclude that the expanding shell of gas is moving outward from the dying
star with speeds of 10 to 30 km/sec. A typical planetary nebula has a diameter of roughly 1 light year, which means that it must have begun
expanding about 10,000 years ago.
The deaths
of slat
By astronomical standards,
a planetary nebula is a very short-lived

After about 50. ()()() years, the nebula has spread over distances so
far from the cooling; central star that its nebulosity simply fades from
view. The gases then mingle and mix with the surrounding interstellar
medium. .Astronomers estiinate that all the planetary nebulae in the galaxy return a total of 5
to the interstellar medium each year. This
amount is about 15 percent of all the matter expelled by all sorts of stars
entity.
Mq
each year. Because of this significant contribution, planetary nebulae

must plav an important role in the evolution of the galaxy as a wfioie.
The burned-out core

of a low-mass star cools
and contracts to become
a white dwarf
Mq never develop the necessary central

pressures or temperatures to permit thermonuclear reactions using carbon or oxygen as fuel. Instead, mass ejection strips off the outer layers
Stars less massive than about 3
from the carbon-oxygen
core, which simply cools off.

evolutionary tracks of three burned-out stellar cores are shown in
Figure 14-4. The initial red supergiants had masses between 0.8
and
The
Mq
Mq. During
between 25
and 60 percent of their matter. During tlie ejection phase, the outward
appearance of these stars changes rapidly and thev race along their evolutionary tracks across the H R diagram, sometimes executing loops that
correspond to stellar pulses. Finallv. as the ejected nebulae fade and the
3.0
their final spasms, these dying stars eject
evolutionary tracks take a sharp turn downward

toward the white-dwarf region of the diagram.
There is no possibility of igniting any additional nuclear fuels inside
one of these dead stars, so the crushing weight of gases pressing inward
from all sides severely compresses the stellar corpse. The density inside
stellar cores cool, their
\V-
III"
"'
c;
1
10-
:i
^"\
^>""- -^
'
\^."X;
"
N..
Irack
Ira. k
N^*
>
si-qutlKiX
'
+
+
+
10-^
+ ^
+ +
10-^
1
1.000.000
:iiio,oiio
fo.nim
iiin.ouii
lVM.|i.i.m.i.-
Figure 14-4
"*"!
iii.oiio
juiio
(K)
Evolution from red superThe evnhtltvnan
giants to white dwarfs

tracks
(ij
three luiu-masi red .sufiergtanti are
ihoiim as they eject planetary nebulae. The

table gives the extent of
the three cases.
The
masi tou in each
dots on this
graph
oj
repre-
Evolutionary
track
Figure 14-4
sent the central stars of planetary nebulae
whose surface temperatures and luminositws
The
crosses are luhite
dwarfs for which similar data
from B. Paczynski)
exist.
Supergiant
Mass of
mass
(Mo)
ejected nebula
(Mo)
White dwarf
mass
(Mo)
.\
3.0
1.8
1.2
h.uk B
1.5
0.7
0.8
0.8
0.2
0.6
liave been determined.
on
.i(
(Adapted
Track

The deaths of stars
the dead star skyrockets until the electrons are so closely packed that
they become degenerate. The degenerate-electron pressure is strong
enough to support the star, so the gravitational contraction halts and the
star is roughly the same size as the Earth. Such a star is called a white
dwarf.
The
density of matter in one of these Earth-sized stellar corpses
typically 10'' g/cm'\ In other
Figure 14-5
companion
Sirius
and
its
while-dwarf
Sinus, the brightest appearing
star in the sky,
secondary star
is
is
actually a double star.
The
a white dwarf, seen here at
the "five-o'clock" position, in the glare of Sirius.
The
spikes
and
rays
around
Sirius are
created by optical effects within the telescope.
(Courtesy of R. B. Minion)
High-mass stars die violently

by blowing themselves apart
in supernova explosions
is
words, a teaspoonful of white-dwarf matter
brought to Earth would weigh as much as a truck.

Calculations by Subrahmanya Chandrasekhar at the University of Chicago prove that there is an upper limit to the mass that a white dwarf
can have. This maximum mass, called the Chandrasekhar limit, is equal
to 1.4 Mq. Above this limit, degenerate-electron pressure cannot support
the weight of the star's matter pressing inward from all sides. Thus all
white dwarfs must have masses less than 1.4 Mq.
Several hundred white dwarfs are scattered across the sky. All are too
faint to be seen with the naked eye. One of the first white dwarfs to be
discovered is a companion to the bright star Sirius. The binary nature of
Sirius was first deduced in 1844 by the German astronomer Friedrich
Bessel, who noticed that the star was moving back and forth slightly, as
if orbited by an unseen object. This companion, called Sirius B (see Figure 14-5), was first glimpsed in 1863. Recent satellite observations at ulwhere white dwarfs emit most of their light
traviolet wavelengths
demonstrate that the surface temperature of Sirius B is about 30,000 K.
As a white dwarf cools, both its luminosity and its surface temperature
decline. As billions of years pass, white dwarfs get dimmer and dimmer
as their surface temperatures drop toward absolute zero. This condition
will be the final fate of our Sun: a cold, dark, dense sphere of degenerate gases rich in oxygen and carbon, about the size of the Earth.
their lives very differently from low-mass stars. A

capable of igniting a host of additional thermonuclear
reactions in its core. The more massive a star is, the greater the temperatures that can be achieved in its core by gravitational compression. High
temperatures are required to initiate nuclear reactions involving heavy
nuclei because the large electric charges of these nuclei exert strong
forces tending to keep the nuclei apart. Only at the great speeds associated with high temperatures are the nuclei traveling fast enough to pen-
High-mass
stars
high-mass star
end
is
and fuse together.

and oxygen are the "ashes" of helium burning.
When helium burning is finished at the core of a star, gravitational compression drives the star's central temperature up to 600 million kelvins,
and carbon burning begins. This thermonuclear process produces elements such as neon and magnesium.
If a star is massive enough to drive its central temperature to 1 billion
kelvins, neon burning begins. This process uses up the neon accumulated from carbon burning, further increasing the concentrations of oxygen and magnesium in the star's core.
etrate each other's repulsive electric fields
Recall that carbon
If the central
temperature of the
star reaches
about
1.5 billion kelvins,
oxygen burning begins. The principal product of oxygen burning is sulfur. As the star consumes increasingly heavier nuclei, thermonuclear reactions produce many different elements. For instance, oxygen burning
also produces isotopes of silicon, phosphorus, and more magnesium.
When a given nuclear fuel is exhausted in the core of a massive star,

gravitational contraction to ever higher densities drives up the star's cen-
The deaths
tif
stars
temperature, thereby igniting the "ash"
ol ilie prexious burning

thermonuclear reaction occurs with increasing rapidity, lor example, detailed cik ulaiions lor a 'Jfi
star demonstrate
that carbon burning occurs tor 600 years, neon burning for
year, but
oxygen burning tor only 6 months.
After halt' a year of core oxygen binning in a 25 M,-, star, a final
gravitational compression forces the central temperature up toward 3 billicjn kelvins, when silicon burning begins. This thermonuclear process
proceeds so turiously that the entire core supply ot silicon in a 25 Mq
star is used up in one day.
Silicon burning involves many hundreds ot nuclear reactions. The
major final product of this process is a stable isotope of iron, ^''Fe. Because ''Te does not act as a fuel for any further thermonuclear reactions, this point is where the secjuence ot burning stages ends. In order
tor an element to be a thermonuclear fuel, energy must be given off
when its nuclei collide and fuse. This energy comes from packing the
neutrons and protons together more tightly in the ash nuclei than in the
tuel nuclei. The 56 protons and neutrons inside the iron nuclei are so
tightly bound together that no further energy can be extracted by fusing
tral
stage. F.ach successive
more nuclei with iron.

The build-up of an inert,
still
iron-rich core signals the impending \ ioleni

Surrounding this iron core, successive layers ot
shell burning consume the star's remaining reserves of fuel (see Figure
14-6). Fhe entire energy-producing region of the star is contained in a
volume as big as the Earth, contiasted with the star's enormously bloated
atmosphere, which is nearly as big as the orbit ot Jupiter.
Because iron does not "burn," the electrons in the core must now support the star's outer layers by the brute strength of degeneracy pressure
alone. Soon, however, the continued deposition of fresh iron tiom the
silicon-burning shell causes the core's mass to exceed the Chandrasekhar
limit. Electron degeneracy suddenly becomes unable to support the star's
enormous weight, and the core begins to collapse.
death of a massive
Figure 14-6
mass star
iruiss
The structure of an old high-
\eiir Ike mil nj l^
/;/('.
rj
lui^h-
star beromes a red superguiut almost
big as the orbit of Jupiter.
comes from
six concentric
The
star's
bunting
a,\
energy
shells, all
contained within a valume rniighlt the same

size as the Earth.
star.
The
tlealks
of stars
mass greater than 10 Mq is capable of developing an

stage will exceed the Chandrasekhar limit, triggering a rapid series of cataclysms that will tear the star apart in a few seconds. Let us see how this happens in the death of a 25 Mq star.
Any
star with a
iron core that at
some
degenerate-electron pressure fails when the density inside the

1 billion g/cm^. The core immediately begins to collapse. Central temperatures promptly soar to almost inconceivable
heights. In roughly a tenth of a second, the temperature exceeds 5 billion kelvins. Photons associated with this intense heat have so much energy that the\ begin to break up the iron nuclei in a process called
photodisintegration.
Within another tenth of a second, as densities continue to climb, the
electrons are forced to combine with protons to produce neutrons in a
process called neutronization. At about \ second after the collapse begins, the density in the core reaches 4 x 10'^ g/cm^, equaling the density
with which neutrons and protons are packed together inside nuclei.
Matter at nuclear density is virtually incompressible. Thus, when the
neutron-rich material of the core reaches this density, it suddenly becomes quite stiff, and the core collapse comes to an abrupt halt.
At this critical stage, the star's unsupported inner regions are plunging inward at speeds of 10 to 15 percent of the speed of light. As this
material crashes onto the now-rigid core, enormous temperatures and
pressures develop, causing the falling material to bounce. In a fraction
of a second, a wave of matter begins to move back out toward the star's
surface. This wave accelerates rapidly as it encounters less and less resistance, and soon it becomes a shock wave. After a few days, this shock
wave reaches the star's surface, by which time the star's outer layers have
begun to lift away from the core. In this way, the star becomes a supernova.
This particular description of the death of a 25
star is based on
detailed computer calculations of an evolving stellar model. The key
stages in such a star's evolution are summarized in Table 14-1.
The final stages in the evolution of other massive stars probably folFirst,
iron core reaches
low similar scenarios, although the details may be different. This particular 25 M- star ejects 24 M-. leaving behind a 1 A/q corpse called a neutron star. Under slightly different conditions, a massive star might blow
itself completely apart, leaving no corpse at all. For example, the bounce
and subsequent shock wave might develop at the center of the star
TABLE
a 25
14-1
Evolutionary stages of
Mrry star
Stage
Temperature
Density
Duration
(K)
(g/cm^)
of stage
7x10*
years
700
5x10'
years
Hydrogen burning
4 X 10'
Helium burning
Carbon burning
6x
X 10^
10"
X 10'
Neon burning
1.2
X 10
4 X 10*
Oxygen burning
1.5
X 10"
10'
2.7
600 years
1
X 10"
X 10"
Core collapse
5.4 X 10"
X 10"
Core bounce
2.3 X 10'
4 X
Explosive
About 10"
Varies
Silicon
burning
year
6 months
lO'-*
day
second
Milliseconds
10 seconds

267
Till- (Inilhs iij slats
the sui lace of the iieutronized core. Alternatively, with a

initial conditions inside the star, a massive burned-out
core might gravitaiionally collapse to form a black hole. This exotic object is discussed in the next chapter.
ratlu-i
than
ai
diflerent set of
The death
of a massive star
can produce a
supernova remnant
layers of a massive dying star are blasted into space, the

luminosity suddenly increases by a factor of lO", equivalent to a
jump of 20 magnitudes in brightness. An outburst of this enormity is
called a supernova. For a few days following the explosion, the star can
As the outer
star's
shine as brightlv as an entire galaxy. .Astronomers often find supernovae

in distant galaxies (see Figure 14-7). Indeed, most of our understanding
of supernovae comes from obserx ing outbursts
Astronomers
also find
in
remote
manv supernova remnants
galaxies.
scattered across the
example is the Veil Nel)ula seen in Figure 14-8. The

outer layers were blasted into space with such violence
that they are still traveling at supersonic speeds through the interstellar
medium. As this expanding shell of gas plows through the interstellar
medium, it collides with interstellar atoms, which excites the gas and
skies.
beautiful
doomed
star's
causes
to glow.
it
Many supernova remnants are quite large and cover sizable fractions
of the sky. The largest is the Gum Nebula it has a diameter of 60
named after (lolin Gum. who first noticed its taint glowing wisps on
photographs of the southern sky (see Figure 14-9). It is also called the
Vela supernova remnant; because it is centered around the constellation
of Vela.
The nebula looks so big because it is so close. Its near side is only
about 300 light years from Earth. Studies of the nebula's expansion rate
Figure 14-7
supernova
Sometime
during 1940. a supernova exploded

galaxy
NGC
Berenices,
4725
(a)
the
in the constellation of
The galaxy
hefc^re
Coma
the outburst.
By the time this photograph was taken in

1941, the supernova (indicated by arrow) had
(b)
faded from
its
maximum
brightness.
(Mount
Wilson and La Campanas Observatories)
The dealhs
The
Figure 14-8
Veil
Nebula
ThU
oj slnrs
nebu-
a portion of the Cygnus Loop (see Fii;ure 13-13), which is the remnant of a superlosity is
nova
20,000 years
that exploded about
The distance
to the
nebula
is
ago.
about 1600
light
and the overall diameter of the loop is

about 70 light years. (Palomar Observatory)
years,
The
Figure 14-9
Nebula
is
Gum
Nebula
the largest knuirn
The
Gum
supernova rem-
nant, spanning 60 of the sky.
The
nearest
portions of this expanding nebula are only
300
light years
from
explosioti occurred
the Earth.
about
The supernova
1,000 years ago.
The supernova remnant now has a diameter of

about 2300 light years. Only the central regions of the nebula are shown here. (Royal
Obset-vatoiy,
Edinburgh)
suggest that the supernova exploded around 9000 bc. It could have
been witnessed by people then living in such places as Egypt and India.
At maximum brilliancy, the exploding star reached an apparent magni-
tude of -10, equal to the brightness of the
Moon
at first-quarter.
Many supernova remnants
are virtually invisible at optical wavelengths. Nevertheless, as the expanding gases collide with the interstellar
medium they do radiate energy at a wide range of wavelengths, from X
rays through radio waves. For example. Figure 14-10 shows both X-ray
and radio images of the supernova remnant Cassiopeia A. Optical photographs showing this part of the sky reveal only a few small, faint wisps.
Radio searches for supernova remnants are more fruitful than optical
searches. Only two dozen supernova remnants have been foimd on photographic plates, but more than 100 remnants have been discovered by
radio astronomers.
From
the expansion rate of the nebulosity in Cassiopeia A, astrono-
mers conclude that the supernova explosion occurred about 300 years
ago. Although telescopes were in wide use by the late 1600s, no one saw
The deaths
Figure 14-10
remuitnh
Cassiopeia
.such as C.assiopeta
strong sources of
(a)
An
rays
A
A
and
radio waves.
X-ray picture of "Cos
Einstein Observatory, (b)
nova explosion
3(11)
that
A"
taken by the
corresponding
radio image produced by the
curred
Superiitiva
are typically
produced
VLA. The superthis
nebula oc-
years agii, about 10.0(1 light years
from Earth. (Smithsonian

Very Large Array)
Institution
and
the
/ \lan
the outburst. In fact, llie last supernova seen in our galaxy, in 1604, was
observed by Johannes Kepler. In 1572, Tycho Brahe also recorded the
sudden appearance of an exceptionally bright star in the sky. To llnd
any other accounts of supernova explosions, we must dehe into ancient
astronomical records that are almost a thousand years old.
At first glance, this apparent lack of nearby supernovas may seem
puzzling. Astronomers have seen more than 600 supernovae in distant
galaxies. From this frequency, it is reasonable to suppose that in a galaxy
such as our own wc should have about five supernovae per century.
Where
are they?
Vigorous stellar evolution is occurring primarilv in the disk of oiu'

galaxy where, for example, the spiral arms sweep ihrcmgh gianl molecular clouds. Our galaxy's disk is therefore the place where massive stars
are horn and where supernovae explode. This region of oiu' galaxv is so
filled with interstellar gas and dust, however, that we simply carmot see
very far into space in the directions occupied by the Milky Way. In other
words, supernovae probably do erupt every few years in remote parts of
our galaxy, but their detonations aie hidden from our \iew b\ intervening interstellai' debris.
Pulsars are rapidly rotating

neutron stars with intense
magnetic
fields
The neutron was discovered duting laboiatoi) expeiimenis in HKVi.

Within a year, two astronomers had predicted the existence of neutron
stars. Inspired by the realization that white dwarfs are supported by degenerate electron |)ressin'e, Fritz Zwicky at the Clalilornia Institute of
Technology and his colleague Walter Baade at Moimi Wilson Observatory proposed that a bighlv compact ball of m-utrons could simil.irK- produce a powerful |3ressine. This degenerate-neutron pressure, like the
degeneiate-ele( troll jiressine, loiild also siippoit a stellar corpse, |)erhaps
even more massi\e than the (Ihandrasekhar limit allows. "With all reserve," Zwicky and Baade theorized, "we advance the view thai superno-
vae represent the transition from ordinary stars into neutron stars,
wfiich in their final stages consist of extremely closely |jac ked neutrons."
In other words, there could be .it least two t\i)es of stellai corpses: white
dwarfs and neutron
stars.
Tlie deaths of stars
This proposal was politely ignored by most scientists for years. After
a neutron star must be a rather weird object. In order to transform
protons and electrons into neutrons, the density in the star would have
to equal to nuclear density, 10'"* g/cm^. A thimbleful of neutron-star
matter brought back to Earth would weigh 100 million tons. Furthermore, an object compacted to nuclear density would be very small. A
1 Mq neutron star would have a diameter of only 30 km, about the
same size as San Francisco or Manhattan. The surface gravity on one of
these neutron stars would be so strong that the escape velocity would
equal one-half the speed of light. All these conditions seemed so outrageous that few astronomers paid any serious attention to the subject of
neutron stars until 1968.
As a young graduate student at Cambridge University, Jocelyn Bell
had spent many months assisting in the construction of an array of radio
antennas covering 41 acres in the English countryside. By the fall of
1967 the instrument was completed, and Bell and her colleagues began
all,
detecting radio emissions from various celestial sources. In November,
while scrutinizing data from the
new
telescope. Bell noticed that the an-
tenna had detected regular "beeps" from one particular location in the
sky. Careful repetition of the observations demonstrated that the radio
pulses were arriving with a regular period of 1.3373011 sec (see Figure
14-11).
The
regularity of this pulsating radio source was so striking that the
Cambridge team suspected
might be detecting signals from an

was soon discarded as several
more of these pulsating radio sources, or pulsars, were discovered across
the sky. In all cases, the periods were extremely regular, from about ?
sec for the fastest to about 14 sec for the slowest.
When the discovery of pulsars was officially announced in early 1968,
astronomers around the world began proposing all sorts of possible ex-
advanced
that they
alien civilization. This possibility
planations for these regular radio pulsations.
Figure 14-11
pulsar
sity
recording of the first
This chart recording shows the inten-
of radio emission from the first pulsar, dis-
covered by British astronomers in 1967. Note

that
some pulses are weak and
others are
strong. Nevertheless, the spacing between the
pulses
is
exactly
1.3373011
from Antony Hewish)
sec.
(Adapted
Many of
these theories
were bizarre, and arguments raged for months. However, by late 1968,
all the controversy was laid to rest with the discovery of a pulsar in the
middle of the Crab Nebula.
In AD 1054, Chinese astronomers recorded the appearance of a supernova (they called it a "guest star") in the constellation of Taurus. When
we turn a telescope toward this location, we find the Crab Nebula shown
in Figure 14-12. This object looks like an exploded star and is a supernova remnant. The pulsar at the center of the Crab Nebula is called the
Crab pulsar.
The Crab pulsar is one of the fastest pulsars ever discovered. Its period is 0.033 sec, which means that it beeps 30 times each second. The
fact that a pulsar is located in a supernova remnant tells us that pulsars
probably are associated with dead stars.
Before the discovery of pulsars, most astronomers believed all stellar
corpses to be white dwarfs. There seemed to be a sufficient number of
white dwarfs in the sky to account for all the stars that have died since
our galaxy was formed. So it was generally assumed that all dying stars
somehow manage to eject enough matter so that their corpses do not
exceed the Chandrasekhar limit.
The deaths of stars
Figure 14-12
The Crab Nebula This

named for the armlike ajj-
beautijul nebula,
pearance of its fUamentan/ strutlure. is the

remnant of a supernova seen in AD 1054. Th,
distance to the nebula
years,
and
its
min) corresponds
7
tr\
10
is
about
6000
light
present angular size (4 b\ 6

to
light years.
an
linear dimensions of alxml
(Palomar Olaen'atom
These conservative opinions were dashed widi the discovery of the

Crab pulsar. It was immediaiely apparent to astrophysicists that white
dwarfs are too big and bulky to produce 30 signals per second. Calculations demonstrated that a white dwarf can neither rotate that fast nor
vibrate that fast. The Crab pulsar clearly indicates that the stellar corpse
at the center of the Crab nebula is much smaller and more compact than
a white dwarf. Astronomers realized that they would have to face the
prospect of neutron stars seriouslv.
As we have seen, a neutron
would be small and dense. It should

most of them do so leisurely. For example, our Sun takes nearly one month to rotate once
about its axis. A collapsing star speeds up as its size shrinks just as an ice
skater doing a pirouette speeds up when she pulls in her arms. This
phenomenon is a direct consequence of a law of physics called the conservation of angular momentum which holds that the total amount of
star
also be rotating rapidly. All stars rotate, but
angular momentum in a svstcm remains constant. An ordinary star rotating once a month would be spinning faster than once a second if compressed to the size of a neutron star.
In addition to having ra])id lotation, we expect a neutron star to have
an intense magnetic field. It is probably safe to say that everv star has a
magnetic field of some strength. In an average star, like oin- Sun, the
strength is ivpicallv quite low because the magnetic field is spread out
over millions upon millions of square kilometers of the star's surface.
However, if a star of solar dimensions collapses down to a neutron star,
its magnetic field becomes very compact and its strength increases bv a
factor of a billion.
Finallv,
we would expect the axis of rotation of a typical neutron star

some angle to the magnetic axis coiniecting the north
to be inclined at
and south magnetic poles (see Kigine l-4-i:<). fhe combinaiion ol a powand lapid rotation operate like a giant electric gen-
erful magnetic field
erator to create intense electric fields near the star's surface. At its surface, there are plenty of protons and electrons because the pressures are
loo low to combine them into neutrons. The powerful electric fields acting on these charged particles cause them to flow out from the neutron
Thr dmtks of
Figure 14-1$
iiar
rotating, magnetized
neutron star
II
u reasonable
a neutron star
is
rotating rapidly
to
suppose that
and
possesses
a powerful magnetic field. Charged particles

are accelerated near the star's magnetic poles
and produce two

radiation.
around
As
oppositely directed
the star rotates, the
the sky. If the
the path of the beams,
beams of
beams sweep
Earth happens
we
see
to lie in
a pulsar.
polar regions along the curved magnetic field, as sketched in Figure 14-13. As the particles stream along the curved field, they are accelerated and emit energy. The end result is two very thin beams of radiation pouring out of the neutron star's north and south magnetic polar
Star's
regions.
Figure 14-14
pliotograptis
and
The Crab pulsar
show
the
Crab pulsar
These two
in
its
"on"
"off" states. Like the radio pulses, these
visual flashes have a period of 0.033
sec.
These pictures of the center of the Crab Nebula can be used to identify the neutron star in
Figure 14-12. (Lick Obserrmtory)
A rotating, magnetized neutron star is somewhat like a lighthouse

beacon. As the star rotates, its beams of radiation sweep around the sky.
If the Earth happens to be located in the right direction, a brief flash
can be observed each time the beam sweeps past our line of sight.
When this scenario was first proposed, in the late 1960s, a team of
astronomers at the University of Arizona began wondering if pulsars
emit flashes at wavelengths other than radio frequencies. After aiming a
telescope at the center of the Crab Nebula, they used a spinning disk
with a slit to "chop," or interrupt, the incoming beam of light at rapid
intervals. To their surprise, they found that one of the stars at the center
of the nebula is actually flashing on and off 30 times each second (see
Figure 14-14).
During the 1970s, radio astronomers discovered about 300 more pulsars scattered across the sky. Presumably, each one is the neutron-star
corpse of a massive extinct star. In the early 1980s, astronomers even
found two pulsars with periods of cjnly a few milliseconds. One of them
is blinking on and off at visible wavelengths, like the Crab pulsar. Also
visibly flashing is the Vela pulsar at the core of the Gum Nebula (see
Figure 14-9). The Vela pulsar, with a period of 0.089 sec, is the slowest
pulsar ever detected at visible wavelengths.
The Crab pulsar is one of the youngest pulsars, its creation having
been observed some 900 years ago. Of course, a few supernovae have
been seen since then (such as those noted by Tycho Brahe and Johannes
Kepler), but no pulsars are found at these locations. Perhaps the stellar
corpses are not neutron stars, or perhaps their beams simply do not
sweep past the Earth. There must be many pulsars that we will never
discover because they are oriented at unfavorable angles.
The
273
deallis oj itars
The Vela pulsar, like the Crab pulsar, is also quite youiif^, its creation
having occurred roughly 1,000 years ago. We may conclude that pulsars
slow down as they get older and that only the very youngest pulsars are
energetic enough to emit optical flashes along with their radio pulses.
1
Pulsating X-ray sources

are neutron stars
in
close
binary systems
Figure 14-15
Ulturu
w(L\
(I
Uhuru (Explorer 42}

smalt
satellite desigyied to detect
astronomical sources of
rays.
years of jhiwless operation,
than
300
across
it
Oiirtng three
observed more
different X-ray-emiltm)^ objects
tlie sky.
(NASA)
During the 1960s, astronomers obtained tantalizing X-ray views of the

and balloon flights that momentarily lifted X-ray
detectors above the Earth's atmosphere. A number of strong X-ray
sources were discovered and each was then named after the constellatic^n
in which it is located. For example, Scorpius X-1 is the first X-ray source
found in the constellation of Scorpius.
Astronomers were so intrigued by these preliminary discoveries that
they built and launched Explorer 42, an X-ray- detecting satellite that
could make observations 24 hours a day (see Figure 14-15). The satellite
was launched in 1970 from Kenya to place it in an equatorial orbit. In
recognition of the hospitality of the Kenyan people, Explorer 42 was
renamed Uhuru, which means "freedom" in Swahili.
Uhuru gave us our first comprehensive look at the X-ray sky. As the
satellite slowly rotated, its X-ray detectors swept across the heavens. Each
time an X-ray source came into view, signals were transmitted to receiving stations on the ground. Before its battery and transmitter failed in
early 1973, Uhuru had succeeded in locating 339 X-ray sources.
The discovery of pulsars was still fresh in everyone's inind when the
Uhuru team discovered X-ray pulses coming from Centaurus X-3 in
early 1971. Figure 14-16 shows data from one sweep of Uhuru's detectors across Centaurus X-3. The pulses have a regular period of 4.84 sec.
A few months later, similar pulses were discovered coming from a
source called Hercules X-1 that has a period of 1.24 sec. Because the
periods of these two X-ray sources are so short, astronomers began to
suspect that they had found rapidly rotating neutron stars.
It soon became clear, howexer, that svstems such as Centaurus X-3
and Hercules X-1 are not ordinary pulsars like the Crab or Vela pidsars.
Periodically, every 2.087 days, Centaurus X-3 turns off for almost 12
hours. This fact suggests that Centaurus X-3 is an eclipsing binary and
sky during brief rocket
The deaths of
Figure 14-16
Centaurus X-3
sity
of
X-ray pulses from

This graph shows the inten-
rays detected by
X-3 moved
Vhuru
as
Centaurus
across the satellite's field of view.
The
successive pulses are separated by
sec.
The gradual vanalion
pulses
from
stars
left to right is
4.84
in the height of the
a result of the
changing orientation of Uhuru's X-ray detectors toward the source as the satellite rotates.
(Adapted from R. Giacconi and colleagues)
Time
that
it
panion
The
(seconds)-
takes nearly 12 hours for the X-ray source to pass behind
its
com-
star.
case for the binary nature of Hercules X-1
is
even more compel-
has an "off" state corresponding to a 6-hour eclipse every 1.7

days, and careful timing of the X-ray pulses shows a periodic Doppler
shifting every 1.7 days. This information is direct evidence of orbital
ling. It
motion about a companion

us, its pulses
source
is
star.
When
receding from us, slightly
the X-ray source
is
approaching
than 1.24 sec. When the

more than 1 .24 sec elapses between
are separated by slightly
less
the pulses.
Careful optical searches around the location of Hercules X-1 soon

revealed a dim star named HZ Herculis. The magnitude of this star varies
between 13 and 15 and
riod
is
exactly the
omers conclude
same
that
HZ
has a period of 1.7 days. Because this peX-ray source, astronHerculis is the companion star aroimd which
it
as the orbital period of the
Hercules X-1 is orbiting.

Putting all the pieces together, astronomers have now realized that
systems such as Centaurus X-3 and Hercules X-1 are examples of double
stars in which one of them is a neutron star. All these binaries have very
short orbital periods. The distance between the ordinary star and its
neutron star must therefore be very small. This proximity puts the neutron star in a position to be able to capture gases escaping from the ordinary companion star.
In the mid- 1800s, the French mathematician Edward Roche pointed
out that you can draw a figure-eight curve around the two stars in a binary to portray the gravitational clomain of each star. If gases should
happen to leak over the figure-eight-shaped surface surrounding one of
the stars, this material is no longer boimd to the star and is free to escape into space. This figure-eight curve is often called the critical surface, and each half of the curve is called a Roche lobe.
To explain pulsating X-ray sources such as Centaurus X-3 or Hercules
X-1, astronomers assume that the ordinary star either fills or nearly fills
its Roche lobe. Either way, matter escapes from the star. This mass loss
occurs by direct "Roche-lobe overflow" if the star fills its lobe, as in the
case of Hercules X-1, or it occurs by a stellar wind if the star's surface
lies just inside its lobe, as with Centaurus X-3 (see Figure 14-17). A typical rate of mass loss from the ordinary star is roughly 10"^ solar masses
per year.
As with an ordinary pulsar, the neutron star in a pulsating X-ray
source rotates rapidly and has a powerful magnetic field inclined to the
axis of rotation (recall Figiue 14-13). Because of its strong gravity, the
neutron star easily captures much of the gas escaping from the companion star. As the gas falls toward the neutron star, the magnetic field fun-
The deaths of star
Figure
1417 A model
X-ray source
star
IS
of a pulsating
(ins escuptng
frum an unlnicin
The in-
capliired by the neutron star.
falting gas
is
funneled down onto
tron star's magnetic poles where

star with
enough energy
the neustrikes
it
to create
lli,
two A'-rni
emitting hot spots. As the neutron star spm^.
beams of X rays from the hot
around
the
spots
sweep
.sky.
nels the
incoming matter down onto
regions.
The
star's gravity
is
its
noith and south magnetic polar
so strong that the gas
is
traveling at nearlv
half the speed of light bv the time it crashes onto the star's surface. This
violent impact creates hot spots at both poles with temperatures of about
lO'* K. so these hot spots emit abundant X rays with a luminositv
roughly 100.000 times brighter than the Sun. As the neutron star rothe beams of X rays from the polar caps sweep around the sky. If
the Earth happens to be in the path of one of the two beams, we can
observe a pulsating X-ray source. The pulse period is thus equal to the
neutron star's rotation period. For example, the neutron star in Hercules
X-I is spinning at the rate of once evcrv 1.24 sec, equal to its pulse petates,
riod.
There
is
an interesting side effect
tured by the neutron
star's gravitv
in this basic scenario.
mav go
Cases cap-
into orbit about the neutron
The
result is a rotating disk of material called an accretion disk, as

Figure 14-17.
Accretion disks have been detecicd in many close binarv svsiems
where mass transfer is occurring between the two stars. Under certain
circumstances, some bizarre things can happen. With ordinary pulsating
X-ray sources such as Hercules X-I, the rate at which gas falls onto the
neutron star is fairly low. This material falls onto the neutron star from
the inner edge of the accretion disc at a rale low enough to allow the
star.
shown
in
resulting
ra\s to escape.
companion
star is dumping vast amounts of material onto the

however, the resulting energv caimot easiK escape. Tremendous pressures build up in the gases crowding down onto the neutron star. These pressures meet strong resistance in the plane of the accretion disk, from which newly arrived gases are constantlv spiraling in
toward the neutron star. The "path of least resistance" is perpendicular
to the plane of the accretion disk. The gas pressure around the neutron
If
the
neutron
star,
Thr
Star
is
(Imllis oj star
thus relieved by material squirting out along this perpendicular

The result can be two powerful beams of high-velocity hot gas,
direction.
shown in Figure 14-18. This explanation apparently accoimts for the

weird star called SS 433.
In the autumn of 1978, Bruce Margon and his colleagues at UCLA
were observing SS 433, which had been noted for strong emission lines
in its spectrum. To everyone's surprise, the spectrum of SS 433 contained several complete sets of spectral lines. One set was very redshifted
from its usual wavelengths, the other set was comparably blueshifted.
Soinehow, SS 433 is "coming and going at the same time." To make matters even more puzzling, the wavelengths of these redshifted and
as
blueshifted lines change dramatically from one night to the next.
Astronomers had never seen anything like this, and soon inany were
observing SS 433. By mid- 1979, it was clear that the system's redshifted
and blueshifted lines are actually moving back and forth across the spectruin of SS 433 with a period of 164 days. Astrophysicists were quick to
point out that the two sets of spectral lines could be caused by two oppositely directed jets of gas, one tilted toward us and the other away from
us. Furthermore, the 164-day variation could be explained by a precession of the two jets. As the two jets circle about the sky every 164 days,
we
see a periodic variation in the
Doppler
come together
shift.
model sketched
in Figure 14explain the large redshifts and blueshifts discovered by Margon,

gas in the two oppositely directed jets must have a speed of 78,000 km/
All these features
18.
in
the
To
roughly one-quarter the speed of light. In addition, the accretion

must be tilted with respect to the orbital plane of the two stars of
the binary system. Just as the tilt of the Earth's axis with respect to the
plane of the ecliptic causes the Earth to precess, the tilt of the accretion
disk results in the 164-day precession of the two jets.
sec,
disk
A model
Figure 14-18
jrum a normal
.star
of SS 433
Gas
captured into an accrt-
lion disk about a neutron star.
Two
high-
speed, oppositely directed jets of gas are ejected
Jrom
the faces of the disk.
tilted,
causes the jets
Kvl
liecame
the gravitational pull of the
da\s.
In
the dish
normal
slai
precess with a period /
Tlif dealhi uf stars
Figure 14-19
Four views of SS 433
These jour radio views, taken

shviii jets oj
light
277
gas exiniiting out
in earlt I'JSI,
to one-sixth oj
year on either side of SS 433. Three-
quarters oj the radio emission comes
SS 433
Itself (the
from
red central blob), which
is
located at the center of a supernova remnant
13,000
light years
from Earth. (Ven
Large Array)
Kigiire 14-19 shows loin high-resolution radio views ot SS 43;^. Note

the two oppositely directed appendages emerging from the central
source. As we shall see in Chapter 18, many quasars and peculiar galaxies
have a similar radio structure, though on a
sars are incredibly far
nificance of SS
right in
Explosive thermonuclear
processes on white dwarfs
and neutron stars produce
novae and bursters
433 may be
our own
Low-mass
away and are thus
celestial
that
it
much
larger scale. Qi'<i-
difficult to studv.
The
real sig-
gives us a miniature qiiasailike object
backvard.
more common than high-mass stars, so white

common than neutron stars. With all the bizarre
stars arc far
dwarfs are far more
and fascinating phenomena associated with neutron stars, you might be

wondering if white dwarfs do anvthing more dramatic than simpU cool
oil.
riu- answer (klinilels is ves, OciasionallN, some star in the sk\ suddenly brightens by a factor of 10''. I his phenomenon is called a nova
(not to be coidused with a supernova, which in\()l\es a much greatei increase in brightness). Novae are fairlv connnon. Their abrupt rise in
brightness is lollowed by a gradual decline that m.i\ ^tl(t(ll lot si\eial
months oi moie (see Figures 14-20 and 14-21).
Painstaking observations of many novae stronglv suggest that all novae
are members of dose binary systems ccmtaining a white dwarf. (Gradual
mass transfer from the ordinary companion star (which presumablv fills
its Roche lobe) deposits fresh hydrogen onto the white dwarf. Because of
the strong gravity, this hydrogen becomes compacted into a dense layer
covering the hot surface of the while dwarf. As more gas is deposited,
the temperatim- in the h\clrogen layer increases. Finally, when the temperature reaches alxmi 10' K, hvdrogen binning ignites throughout the
The dmiks of
Figure 14-20
Nova
Herculis 1934
two pKtures show a irvva

brightness as a magnitude
two months
tude
and year
peak
+3
star,
and
in
which they appeared.
1 +8
^ +10
_.-
-1-.
Nova Cygni 1975

-J
'..
-5
Sept.
Sept.
24
Time
Figure 14-21
Cygni 1975
a nova
tliat
Cygnus
in
The
(days)
11
Oct.
14
light curve
of Nova
This graph shows the hisloiy of

blazed forth in the corutellation of
September 1975. The rapid
rise
magnitude followed by a gradual decline

characteristic of all novae.
that
embroiling the white dwarfs surface in a thermonuclear holocaust
we
see as a nova.
similar
phenomenon
is
occurs with neutron
stars.
Beginning
in late
1975, astronomers analyzing data from X-ray satellites realized that their
instruments had detected sudden, powerful bursts of X rays from certain

The record of a typical burst is shown in Figure 1422. The source emits X rays at a constant low level, then suddenly, without warning, there is an abrupt increase, followed by a more gradual
decline. Typically, an entire burst lasts for only 20 sec. Sources that behave in this fashion are called bursters. Several dozen of them have been
discovered, most located toward the center of our galaxy.
Bursters, like novae, are believed to involve close binaries experiencing mass transfer. With a burster, however, the stellar corpse is a neutron star rather than a white dwarf. Gases escaping from the ordinary
companion star fall onto the neutron star. The energy released as this
gas crashes down onto the neutron star's surface produces the low-level
X rays that are continuouslv emitted by the burster.
Most of the gas falling onto the neutron star is hydrogen, which becomes compressed against the hot surface of the star by the star's powerful surface gravity. In fact, temperatures and pressures in this accreting
layer are so high that the arriving hydrogen is promptly converted into
helium by the hvdrogen-burning process. Constant hydrogen burning
soon produces a layer of helium that covers the entire neutron star.
Finally, when the helium layer is about 1 m thick, helium burning
ignites explosively, and we observe a sudden burst of X rays. In other
words, explosive hvdrogen burning on a white dwarf produces a nova,
but explosive helium burning on a neutron star produces a burster. In
both cases, the burning is explosive because the fuel is so strongly compressed against the star's surface that it is degenerate, like the star itself
As we saw with the helium flash inside red giants, ignition of a degenerate thermonuclear fuel always involves a sudden thermal runaway because the usual "safety valve" between temperature and pressure does
not operate.
objects in the sky.
(Lick Obsei-vatoij)
1 +2
& +4
E +6
layer,
(b)
when it had faded to magniare named after the constel-
later,
+ 12. Xovae
lation
These
(a) shvill\ ajler
stars

The deaths of
Figure 14-22
rays
star
from a burster A
burster emits a constant tow intensity of
rays
interspersed with occasional powerful bursts of
This particular burst
rays.
is
typical. It
was
recorded on September 28. 1973. b\ an X-ra\

telescope
on board the Astronomical Nether-
lands Satellite while the telescope was pointed
toward the globular cluster A'GC 662-?.
About one-third of
all
known
located in globular clusters. (Adapted
Walter
i''lpwUV,rJVw'>vJ%|[Wl</vJV^^
bursters are
from
Tin,.- (.
Lewm)
is an upper limit to the mass of a white dwarf, there is

an upper limit to the mass of a neutron star. Above this limit, degenerate-neutron pressure cannot support the overpowering weight of
the star's matter pressing inward from all sides. The Chandrasekhar
limit for a white dwarf is 1.4 A/q and the corresponding upper limit for
a neutron star is roughly 3 M.
Before the discovery of pulsars, most astronomers believed all dead
stars to be white dwarfs. Dying stars were thought to eject enough material somehow so that their corpses would be below the Chandrasekhar
limit. Obviously, this idea proved incorrect. Inspired b\ this lesson, astronomers soon began wondering what might happen if a dving massive
star failed to eject enough matter to get below the upper limit for a neutron star. For example, what might a 5 ;V/rr stellar corpse be like?
The gravity associated with a neutron star is so strong that the escape
velocity is roughly one-half the speed of light. With a stellar corpse
greater than 3 yV/, there is so inuch inatter crushed into such a small
volume that the escape velocity exceeds the speed of light. Because nothing can travel faster than light, nothing
not even light
can leave the
dead star. The star therefore disappears from the universe, its powerful
gravity leaving a hole in the fabric of space and time. The discovery of
neutron stars thus inspired astrophysicists to examine seriously one of
the most bizarre and fantastic objects ever predicted by modern science
Just as there
also
the black hole.
Summary
low-mass
star
becomes a red giant when
a horizontal-branch star
Thermal
shell
hvdrogen burning begins,

begins, and a red su-
when core helium burning
pergiant when the heiitun

ing begins.
in
he (ore
is
exhausted and
shell
helium burn-
pulses in the helium-burning shell can eject the star's outer layfrom the hot carbon-oxygen core ionizes and
ers. Ultraviolet radiation
excites the ejected gases,
producing a planetary nebula.
The burned-out
size
core of a low-mass star becomes a dense sphere about the

of the Earth called a white dwarf.
high-mass star undergoes a sequence of iherniotuulear reactions in its

These are carbon burning, neon burning, owgeii burning, and silicon burning. The star evenni.illv cle\elops an iron-rich core.
core.
high-mass star can die in a supernova explosion that ejects most of the
matter into space at very high speeds. If the core of the star survives the explosion, it may become a neutron star or even a black hole.
star's
A neutron
neutrons
is a very dense stellar corpse consisting of closelv jiacked

sphere roughly 30 km in diameter.
star
in a
Till' (leulhs oj
pulsar
is
a source of periodic radio pulses.
It is
item
caused by a rapidly roEnergy pours out of
tating neutron star with a powerful magnetic field.
the north and south polar regions of the neutron star in intense
that sweep around the sky.
Some X-ray
be neutron
beams
sources exhibit regular pulses. These objects are thought to

systems with ordinary stars.
stars in close binary
Material from the ordinary star in a binary pair can fall onto the surface
its coinpanion white dwarf or neutron star to produce a surface layer
in which thermonuclear reactions can occur.
of
Explosive hydrogen burning may occur in the surface layer of a companion white dwarf, producing the sudden increase in luminosity that we call
a nova.
Explosive helium burning may occur in the surface layer of a companion
neutron star, producing the sudden increase in X-ray radiation called a
burster.
Review questions
1 Why is the temperature in a

nuclear reactions can occur?
On
low as
mately
Where
3
star's
core so important in determining which
H-R
diagram, sketch the evolutionary track that the Sun will folmain sequence and becomes a white dwarf. Approxihow much mass will the Sun have when it becomes a white dwarf?
will the rest of the mass have gone?
an
it
leaves the
Why do
you suppose that

Sun?
all
the white dwarfs
known
to
astronomers are
relatively close to the
4 Why do astronomers suggest that a supernova visible to the naked eye

long overdue?
5
Why
What
that radio searches for supernovae remnants have

than optical searches?
do you suppose
been more
6
is
is
fruitful
the difference between a nova
and
a supernova?
7 During the weeks immediately following the discovery of the first pulsar,
one explanation for them was that the pulses are signals from an extraterrestrial civilization. Why do you suppose astronomers discarded this idea?
8 Describe what radio pulsars. X-ray pulsars, and bursters have in common.
are they different manifestations of the same type of astronomical ob-
How
ject?
Advanced questions
9 Suppose you wanted to determine the age of a planetary nebula. What

observations would you make, and how would you use the resulting data?
10 What reasons can you think of to explain why the rate of expansion of
the gas shells in some planetary nebulae is nonuniform?
11 What kinds of stars would you monitor if you wished to observe a supernova explosion from its very beginning? Look up tabulated lists of the
brightest and nearest stars. Which, if any, of these stars are possible supernova candidates? Explain.
12 To determine accurately the period of a pulsar, astronomers must take

the Earth's orbital motion about the Sun into account. Explain why.
The
deiUlis of itiirs
13 Suppose thai you discover a small glowing disk of light while seaic liing
the sky with a telescope. How would you obscrvationally decide it this object
is a planetary nebula? Could yoin- object be something else? Kxplain.
14 Compare novae and bursters. What do thev ha\c

wavs are thev different?
For further reading
"How
Bethe, H., and Brown, G.

can, May 1985.
in
common?
In what
a .Supernova Explodes." Scientific Ameri-
Clark, D. Superstars. McGraw-Hill, 1984. .4 book on supemuvae.

Greenstein, G. Frozen Stars: Of PuLars. Black Holes, and the Fate of Stars.
Freundlich Books, 1984.
"New Bursts in Astronomv." .Mercun. Sept./Oci. 1977, p. 6.
Grindlay,
J.
Article on bursters.
Helfand, D. "Pulsars." Mercui-^. May/June 1977, j). 2.

Kaler, J. "Planelarv Nebulae and Stellar Evoluti(jn." .Mercun.
]ul\ .\ug.
1981, p. 114.
Margon, B. "The Bizarre Spectrum of SS 433." Scientijic .\meric<in. Oct.
1980.
Reddv.
F.
"Supernovae:
a Challenge." Sk\ is Telescope. Dec. 1983, p.
Still
485.
Wallerstein, G.,
and Woltf,
1981, p. 44.
Weiler, K. "A New Look
1979, p. 414.
at
S.
"The Next Supeinova?" .Mercun. Mar./Apr.
Supernova Remnants." Sky
iS Telescope.
Nov.
15
An
Black holes
accretion disk around a black hole
This false color image shows in cross-sectwn
how matter
hole.
distributes itself
around a black
"supercomputer" was used
to solve the
complicated relativistic equations that govern

the infall of matter
toward
the black hole at
where the two
the center of this diagram,
broad angles point. These calculations proved

that a fat accretion disk resembling a
nut
is
more
stable than
resembling Saturn's rings.
Many
astrophysicists
believe that supercomputers present
revolutionary
dough-
a thin accretion disk
way of doing
a new and
research by simu-
lating a wide range of physical situations
tliat
are not easily observable. (Courtesy of L.
Smarr and J. Hawley)
We
have
mass of a neutron star cannot exceed three solar

we learn that stellar corpses more massive than 3
become compacted to a single point of infinite density, surrounded by gravity
not even light
so intense that nothing
can escape. We discuss the theory of
seeî that the
masses. In this chapter
of these objects called black holes, and we

learn how astronomers search for evidence of them. We fiyid that a black hole
has only three physical properties, and that the stellar corpse in the center of a
black hole has disappeared forever jrom our direct observation.
relativity that predicts the existence
Suppose
masses
mass of a dying star's burned-out core exceeds 3 solar

Mq). This mass is well above the Chandrasekhar limit, so de-
that the
(3
generate-electron pressure could not possibly support the stellar corpse.

Similarly, because 3
is also above the mass limit for neutron stars,
degenerate-neutron pressure is also incapable of supporting the crushing
weight of the burned-out matter pressing inexorably toward the dead
star's center. If it can become neither a white dwarf nor a neutron star,
what might this massive stellar corpse become?
As we saw in the previous chapter, a

A/^ of matter compressed by
roughly
typical
its
own
neutron star consists of

gravity to nuclear density
liliuk holes
in a
sphere roughly 30
km
in diameter. The star's gravity is so strong

from its surface is one-half the speed of light.
corpse, whose weight overpowers degenerate-neutron
that the escape velocity
massive
stellar
pressure, easily compresses
its matter to densities greater than nuclear

does not take much further compression to cause the escape
velocity to exceed the speed of light. For example, if 3 Mq of matter is
squeezed inside a sphere 18 km in diameter, the escape velocity from the
object becomes greater than the speed of light. Nothing can travel faster
than the speed of light, and thus nothing
not even light
can manage
density.
It
to escape from the dead star. The star has disappearecl from the observable universe, although some of its effects can be detected.
The general theory
of
To
appreciate fully the nature of massixe stellar corpses,
we must
use the
relativity describes gravity in
best theory of gravity at
terms of the geometry of

space and time
of these massive dead stars is so strong that

gives the wrong answers. We must turn instead to Albert Einstein's general theory of relativity.
According to the classical physics of Newton, space is perfectly uniform and fills the universe like a rigid framework. Time passes at a
our disposal. The gravitational
around one
the theorv of Isaac Newton
field
monotonous, unchangeable rate. It is always possible to know how fast

you are moving through this rigid fabric of space and time, and the results of your observations depend on your state of motion. Significant
differences between observations made b> a stationary person and bv a
moving person arise at speeds near the speed of light.
In 19().T. -Albert Einstein began a revolution in phvsics with his special
theory of relativity. His goal was to reformulate electromagnetic theorv
so that it did not depend on the motion of an observer. In other words,
using the special theory of relativity, both you on Earth and a friend in
a rocketship traveling near the speed ot light would ha\e the same logical, complete description of electricitv and magnetism, dexoid of anv pitfalls or paradoxes caused by your relative motions.
Neither our location in space and time nor motion through space and
time shall prejudice our description of physical realitv was tlie loftv principle that guided Einstein. To achieve this goal, Einstein found that he
had to abandon the old-fashioned, rigid notions of space and time. For
example, imagine your friend whizzing across the solar system in a rocketship while you retnain here at rest on the Earth. Einstein proved that,
in order for both of vou to agree on the same coherent description of
reality, vou must say that vour friend's clocks have slowed down and her
rulers have shrunk. SpccificalK an observer always finds that moving
clocks seem to be slowed ,uid moving rulers seem to be shortened in the
.
direction of motion.
After developing the special theory of relativity, Einstein turned his

He began by demonstrating that it is not necessary
to think of gravity as a force. According to Newton's theorv, an apple
falls to the ground because the force of gravity pulls the apple down.
Einstein pointed out that the apple would behave exactly the same in
attention to gravity.
from any gravity if the floor were accelerating upward to

meet the apple, as sketched in Figun- l.')-l.
This example of Einstein's principle of equivalence explains that, in a
small volume of space, the downwaid pull of gravity can be accurately
and completely duplicated bv an upwaid .iiceleration of the observer.
The two men in their closed compartments in Figure 1.5-1 in fact have
no way of telling who is at rest on the F.arth and who is in the elevator
free space far
moving upward
at a constantly increasing speed.
Black holes
Figure 15-1
The equivalence principle
The equivalence principle
asserts that
you can-
not distinguish between being at rest in a
and being accelerated upward in a gravity-free environment. This idea

was an important step in Einstein's develop-
gravitational field
ment of
the general theory of relativity.
This approach allowed Einstein to focus entirely on motion (rather

than force) in discussing gravity. From his special theory of relativity, he
knew exactly how rulers and clocks are affected by motion, and he could
thus describe gravity entirely in terms of its effects on space and time.
Far from a source of gravity, the acceleration is small and the effect on
clocks and rulers is therefore small; nearer a source of gravity, the acceleration is greater and the distortion of clocks and rulers is greater. In
this way, Einstein "generalized" his special theory to arrive at his general
theory of relativity.
The general theory of relativity describes gravity entirely in terms of
the geometry of space and time. Far from a source of gravity, space is
"flat"
and clocks
tick at their
normal
rate.
However,
as
you move closer
down and
space becomes increasingly curved, as seen by someone far from the region that is affected by the mass.
In a weak gravitational field, Einstein's general relativity theory gives
the same results as Newton's classical theory. But in stronger gravity,
such as that near the Sun's surface, the two theories predict different
results. For example, recall the precession of Mercury's perihelion and
the deflection of a light ray grazing the Sun, which we discussed in
Chapter 3 (Figures 3-15 and 3-17). In these and other situations, general
relativity has withstood numerous tests. It is by far the most elegant and
accurate description of gravity ever devised.
to a source of gravitv (a mass) clocks slow
black hole is a very simple

object that has only a
"center" and a "surface"
Imagine a dying star too massive to become either a white dwarf or a

neutron star. The overpowering weight of the star's burned-out matter
pressing inward from all sides causes the star to contract rapidly. As the
star's matter is compressed to enormous densities, the strength of gravity
around the collapsing star increases dramatically. According to the general theory of relativity, distortions of space and time become increasingly pronounced around the dying star. Finally, the escape velocity
from the star's surface equals the speed of light, and the star seems to
The geometry of a black hole
Figure 15-2
diugmm ihuws how
This
distorted
Far from
hole.
space
is
the hole, gravity
"jlat."
and space
the ihafje
<tj
space
is
the gravitational field of a black
is
Sear
is
weak and
the hole, gravity
is
strong
highly ciitved. This diagtam haca
a two-dimensional surface as an analogy for

the three-dimensional fabric of space.
disappear from the universe. At this stage, space has become so severely
curved that a hole is punched in the fabric of the universe. The dvinji
star disappears into this hole in space, leaving behind onl\ a black hole.
The geometry of space around a black hole is sketched in Kiginc
15-2. Note that space is flat far from the hole because gravity is weak
there. Near the hole, however, gravity is strong, and the curvature of
space is severe.
The location in space where the escape velocitv from the black hole
equals the speed of light is called the event horizon. This sphere is also
sometimes thought of as the surface of thi' l)lack hole. Once a massive
dying star collapses inside its event horizon, ii permanently disappears
from the uni\'erse. This surface is literallv a horizon in the geometry of
space beyond which we cannot see any events.
In addition to having the effect of curving space, gravity causes time
to slow down. If you stood at a safe distance and watched your friend
fall toward a black hole, you would note that the friend's clocks tick
more and more slowly. In fact, you would conclude that at the event
horizon the clocks stop entirely.
Once a dying star has contracted inside its event horizon, no forces in
the universe can prevent the complete collapse of the star down to a single point at the center of the black hole. The star's entire mass is
crushed to infinite densitv at this point, called the singularity, often considered the center of the black hole.
The structure of a black hole is therefore verv simple. As sketched in
Figure 15-3, it has only two parts: a singularity (the center) surrounded
by an event horizon (the surface). The distance between the singularity
and the event horizon is the Schwarzschild radius, named after the CierHi l)ecame the first to
man astronomer Karl Schwarzschild who in
solve completely Einstein's equations of general relativ ity.
To understand why the complete collapse of a doomed star is inevitable, think aboiu the nature of space and time here on Earth, far from
anv black holes. On Earth we have the freedom to move as we wish
through the three dimensions of s])ace up and down, left and right, or
forward and back. But we do not have freedom to move at will through
the dimension of time. Whether we like it or not, we are all dragged inexorably from the cradle to the grave.
Inside a black hole, its powerful gravity distorts the shape of space
and time so severelv that the orientations of space and time become interchanged. In a limited sense, inside a black hole we can have freedom
to move through time. It does us no good, however, because we lose a
corresponding amoiuit of freedom to move through space. Whether we
like it or not. we are dragged inexorablv from the event horizon to the
singularitv. Just as no force in the uni\erse can prevent the forward
I '.I
march of time
ity)
Figure 15-3
The structure of a black hole
black hole has only two parts: a singiiliirily
surrounded by an event liunzun. Inside the

event lionzon, the escape velocity exceeds the
speed of
light, so the
event horizon
is
one-way
surface. Things can fall in, but nothing can

get out.
and
The distance between
the ei'enl horizon
scliild radiiLs,
Rsrh-
is
the singiilanly
called the .Schwarz-
no force in the unispace (event horizon lo singul.n -
(past to futine) outside a black hole,
verse can prevent the inward
march
of
inside a black hole.
At the singularity, the strength of gravity is infinite, so the curvature

of space and time is infinite. In other words, space and time at the singularity are all jiniibletl up. Space and time do not exist there as separate, identifiable entities.
This confusion of space and time has piofound implications for what
goes on inside a black hole. .All the laws of ])h\sics recjuire a clear, distinct backgroinid of space and time. Without this identifiable backgroimd. we could not s|)eak ralionallv about ihe ariangemeni of objects
in space or the ordering of e\eiits in lime. Because space and lime are
all jimibled u|) at the center of a black hole, the singularitv does not
obey the laws of physics. The singularity behaves in a random and capricious fashion, totally devoid of any rhyme or reason.
Fortunately, we are shielded from the singularity by the event horizon. Although irrational things happen at the singularity, none of the
effects manage to escape to the outside universe. Consequently, the outside universe remains understandable and predictable.
The irrational, random behavior of the singularity is so disturbing to
physicists that, in 1969, the British mathematician Roger Penrose and his
colleagues proposed the so-called law of cosmic censorship: "Thou shalt
not have naked singularities." In other words, every singularity must be
completely surrounded by an event horizon. If a naked singularity could
exist,
The structure
of a black hole
can be completely described

with only three numbers
it
would
affect the universe in unpredictable
and random ways.
In addition to shielding us from singularities, the event horizon prevents

us from ever
knowing
about the material that has fallen into a

is no way we can ever discover the chemical composition of the massive star whose collapse produced a black
hole. Even if someone went into a black hole for measurements or chemical tests, there is no way that observer could get any of this information
back to the outside world. Indeed, a black hole is an "information sink,"
because in-falling matter carries many properties about it (chemical comdetails
black hole. For example, there
position, texture, color, shape, size) that are forever
removed from
the
universe.
Because a black hole removes information from the uni\erse, there is

that this information can affect the structure or properties of the
hole. For example, consider two black holes, one made from the gravitational collapse of 10 Mq of iron, the other made from the gravitational
collapse of 10 Mq of peanut butter. Very different substances obviously
went into the creation of these two black holes. Once their event horizons formed, however, both the iron and the peanut butter permanently
disappeared from the universe. As seen from the outside, the two holes
are absolutely identical, so we cannot tell which ate the peanut butter
and which ate the iron. In this way, a black hole is unaffected by the information it destroys.
Because a black hole is an information sink, you might wonder
whether we can know or measure anything about a black hole. In other
words, what properties characterize a black hole?
First, we can measure the total mass of a black hole, by placing a satellite in orbit about the hole, for example. Kepler's third law relates a satellite's orbital period and semimajor axis to the mass around which the
satellite moves. Thus, after measuring the size and period of the satellite's orbit, we can use this law to determine the mass of the hole. This
mass is equal to the total mass of all the material that has gone into the
no way
hole.
Science fiction abounds with nasty rumors that black holes are evil
around gobbling everything in the universe. Not so! The
bizarre effects of highly warped space and time are limited to a volume
things that go
extending only a few million kilometers from the hole. Farther away,
gravity is sufficiently weak that Newtonian physics adequately describes
everything. For example, at a distance of only a few astronomical units
from a 10 Mq black hole, the behavior of gravity is identical to that
around any ordinary 10 Mq star.
In addition to the total mass, we can also measine the total electric
charge possessed by a black hole. Like gravitational force, electric force
a long-range interaction,
making
its
effects felt in the space
is
around the
Black holes
hole. Appropriate equipment on a space probe passing near the hole

could measure the intensit\ of the electric Tield around the hole, from
which the electric charge ot the hole can be determined.
In reality, we would not expect a black hole to possess ativ appreciable
electric charge. For example, if a hole did start off w ith a si/able positi\e
charge, it would vigorously attract vast numbers of negativeh charged
electrons
from the
hole's charge.
electric
Ergosphere
Figure 15-4
The ergosphere A rolalini^

u nmounded by a region where Ihe
black hole
d^''g^"g
0/ space
and
time around Ihe hole
is
For
medium, which would soon neutralize the

astronomers generally do not consider
interstellar
this reason,
charge when discussing real black holes.
Although a black hole might have a tiny electric charge, it cannot

have any magnetic field whatsoever. It can be mathematically proved
that Einstein's equations do not permit a north-pole/south-pole asymmetry in the geometry of space around a black hole. During the creation
of a realistic black hole, we would expect the collapsing star to possess an
appreciable magnetic field. This magnetic field must be radiated awav in
the form of electromagnetic and gravitational waves before the dead
star can settle down inside its exent horizon. C^ravitational waves are ripples in the overall geometry of space. Some physicists are exploring the
so sei'ere that
it is impossible for anything to

remain at a fixed location. Because the ergo-
possibility of obserx ing the creation of black holes by detecting the bursts
of gravitational radiation emitted by massive, dying stars as they col-
sphere (the shaded region in this cross-section
lapse.
diagram)
is
zare region
outside the event horizon, this bizis
accessible to us
and can
be tra-
versed by astronauts or asteroids without hav-
ing them disappear into the black hole.
its mass and electric charge, we can also measure a

tolal angular momentum. Because of the principle of the conservation of angular momentum, we expect a black hole to be spinning
very rapidly. Einstein's theory makes the startling prediction that this
rotation causes space and time to be dragged around the hole. .\ spin-
In addition to
black hole's
ning black hole
is therefore surrounded by space that rotates yvith the

the event horizon of every rotating black hole is a region
where this dragging of space and time is so severe that it is impossible to
stay fixed in the same place. No inatter yvhat you do, vou get pulled
Around
hole.
around the hole along with the rotating geometry of space and time.
This region yvhere it is impossible to be at rest is called the ergosphere
(see Figure 15-4).
To
ineasure a black hole's angular momentum. y\e could place tyvo

about the hole. Suppose that one satellite ciides the
hole in the same direction the hole rotates and the other satellite circles
in the opposite direction. One satellite is therefore carried along y\ith the
satellites in orbit
rotating geometry of space
and
time, but the other
is constantly fighting
thus yvill have different orbital periods. From a comparison of these periods, the total angular momentum
of the hole can be deduced.
And that is all there is to black holes. .\s the famous no-hair theorem
its
yvay "upstream."
formulated
The
ty\o satellites
in the early 197()s put
"Black holes have no hair.

lo
is completely specified by
only three numbers: its mass, its charge, and its angular momentum.
Any and all additional properties carried by the matter that fell into the
hole haye disappeared from the universe and thus can have no effect on
first
say
it
more
"
it:
formally, a black hole's structure
the structure of the hole.
Black holes have

been detected in
binary-star systems
Finding black holes
in
the skv
is
a difficult
and ollen
frustrating busi-
vou cannot directly observe a black hole in the same

sense vou can observe a star or a planet. The best you can hope lor is to
detect the effects of the hole's powerful gravity. Double-star systems can
offer the opportunity lor such detection. For example, a black hole
ness. Obviously,
might capture gas from its companion star, and the fate of this material
might reveal the existence of the hcjle.
Shortly after the laimch of the Uhurii satellite in the early 1970s, astronomers became intrigued with an X-rav soince called Cygniis X-1.
The source is highly variable and irregular. Its X-ray emission flickers
on time scales as short as a himdredth of a second. One of the fundamental concepts in physics is that nothing can travel faster than the
speed of light. As a consequence, an object cannot vary its brightness or
flicker faster than the travel time of light across the object. Because light
travels 3000 km in 0.01 sec, Cygnus X-1 must be smaller than the Earth.
Cygnus X-1 occasionallv einits radio radiation, and in 1971 radio astronomers succeeded in identifying it with the star HDE 226868 (see
Figure 15-5). Spectroscopic observations promptly revealed that HDE
226868 is a BO supergiant. Such stars do not emit significant amounts of
X rays, and thus HDE 226868 alone cannot be Cygnus X-1. Double stars
are very common, so astronoiners began to suspect that the visible star
and the X-ray source are in orbit about each other.
Further spectroscopic observations soon showed that the spectral lines
in the spectrum of HDE 226868 shift back and forth with a period of
5.6 days. This behavior is characteristic of a single-line spectroscopic binary. The companion of HDE 226868 is too dim to produce its own set
of spectral lines. The clear implication is that HDE 226868 and Cygnus
X-1 are the two components of a double-star system.
Using the mass-luminosity relation for a BO star, HDE 226868 is estimated to have a mass of roughly 30 A/q. This information implies that
Cygnus X-1 must have a mass greater than 6 A/o, otherwise it would not
Figure 15-5
HDE 226868
This star
the
opiuni companion of the X-ray source Cygnus
X-1. The ilar

light years
is
BO
from Earth.
agree that Cygnus X-1

hole. This
supergiant located
Many
is
8000
astronomers
probably a black
photograph was taken with the
200-in. telescope at Palomar. (Courtesy of

J. Kristian,
Mount
Wilson and Las
Campanas Obsematories)

Black holes
Figure 15-6
stellar
The Cygnus X-1 system A

HDE 226S68 pours mailer
wind from
onto an accretion disk

hole.
sunoundmg
The infalting gases are heated
a black
to
high
temperatures as they spiral in toward the hole.
At
the inner edge
of the
disk, just
black hole, the gases are so hot

vast quantities of
above the
tluil
they emit
rays.
enough gravitational
amount deduced from the
exert
Six solar masses
is
pull to
make
the
BO
star
wobble bv the
periodic Doppler shifting of
its
spectral lines.
too large for Clvgnus X-1 to be either a while
dwarf
or a neutron star, so the only remaining possibility is a black hole.

Of course, the X ravs from Cvgnus X-1 do not come from the black
hole itself. Gas captured fiom HDE 226iS6iS goes into orbit about the
forming an accretion disk about 4 million kilometers in diameter

As material in the disk spirals in toward the hole, friction heats the gas to temperatures approaching 2 million kelvins. In the
final 200 km above the hole, these extremely hot gases emit the X rays
that we detect with our satellites. Presumablv, the X-rav llickering is
caused by small hot spots on the rapidly rotating inner edge of the achole,
(see Figure 15-6).
cretion disk. In this way, the black hole's existence
doomed
gases just before thev
pkmge
is
announced bv
to oblivion.
In the early 1980s, a binarv system similar to Cvgiuis X-1 was identi-
nearbv galaxv called the Large Magellanic Ciloud. The X-rav

LMC X-3, exhibits rapid nuctualions just like those of
Cygnus X-1. LMC X-.'< circles a B;^ main-set|uence star everv 1.7 davs.
From its orbital data, astronomers conclude that the mass of the compact
Once again, a black hole is the
X-ray source is probably about 9
only reasonable candidate for such a massive, compact object.
There are several other rapidly llickering X-ray sources in binary svstems. \\\ are excellent black-hole candidates. They include Circinus X-1
and (iX 339-4 in Scorpius. With these and similar objects, a great deal of
observational effort is necessary to rule out all non-black-hole explanations of the data. ()nl\ then can we feel confident that additional black
holes have been discovered.
Lhe bhu k hole is one ot the most bizarre and fantastic objects ever to
emerge from modern physical science. Although the idea of black holes
was initiallv met with skepticism, it is now clear that manv of the stars we
see in the skv are doomed someday to disappear from the universe, leaving behind (jnlv black holes. Even more astounding is the idea that enormous black holes containing millions or even billions of solar masses
are located at the centers of many galaxies and quasars. Indeed, as we
shall see in the next chapter, one of these monstrosities mav be lurking
fied in the
source, called
at
the cenitr of the Milk\ \Va\, onlv 10 kil<)i)arsecs from the Earth.
Black holes
The
Summary
general theory of relativity asserts that gravity causes space to become

curved and time to slow down. These effects are significant only in the
vicinitv of large masses or very compact objects.
If a stellar corpse has a mass greater than about 3 A/: gravitational compression will make the object so dense that the escape velocity exceeds the
speed of light. The corpse then contracts rapidlv to a single point called a
.
singularity.
The
singularity is surrounded bv a surface called the event horizon

(where the escape velocitv equals the speed of light). Nothing not even
light
can escape from the region inside the event horizon.
black hole (a singularity surrounded by an event horizon) has only

three physical properties; mass, electric charge, and angular momentum.
In the case of a rotating black hole, a region called the ergosphere surrounds the event horizon. In the ergosphere, space and time themselves
are dragged along with the rotation of the black hole.
The
general theorv of relativitv predicts the existence of gravitational rawaves are ripples in the overall geometry of space
that are produced by moving masses.
diation. Gravitational
and time
.\ few binary-star systems contain a black hole. In such a svstem, gases

captured from the companion star bv the black hole emit detectable X
ravs.
Review questions
1 Under what circumstances are degenerate-electron pressure and degenerate-neutron pressure incapable of preventing the complete gravitational
collapse of a dead star?
2 In what wav
is
a black hole blacker than black ink or a black piece of
paper?
3 If the
affected?
Sun suddenh became
a black hole,
how would
the Earth's orbit be
4 According to general relativity, why can't some sort of yet-undiscovered

degenerate pressure prevent the matter inside a black hole from collapsing
all
the
way down
5 Why do
text are
Advanced questions
to the singularitv?
all the black hole candidates mentioned in the
verv short period binarv systems?
you suppose that
members of
6 As a binary system loses energy bv emitting gravitational waves,

speed up, and why does the period become shorter?
why do
its
members
7 Under what circumstances might a white dwarf or neutron star in a

double-star svstem become a black hole?
8 Describe the kinds of observations vou might
make
in
order to locate and
identify black holes.
9 Speculate on the effects vou might encounter on a trip to the center of a

black hole.
291
For further reading
Goldsmiih, D.
"When Time
Slows Down." Mercury, May/June 1975.
p. 2.
Oil special relativity.
Kauf'mann. W. Black Holes and Warped Spacetime. W. H. Freeman and
Company,
1979.
Cosmic Frontiers of General Relativity. Little, Brown, 1977.
Stokes. G.. and Michalskv. |. "Cvgnus X-1." Mercury. May/June 1979. p.
Kaiifmann,
VV.
60.
Thorne. K. "The Search for Black Holes."
Scientific
American. Dec. 1974.
16 Our galaxy
The galaxy NGC 253 A spiral galaxy, siuh

own or NGC 253, contains a signijicant amount of interstellar dust, primarily
concentrated in the galaxy's disk. In our own
galaxy, this dust blocks tight from remote stay^
and severely restricts our visual observations in
as our
the plane of the
Milky Way. In other galaxies,
the dust appears as dark blotches
as
shown
around
here.
this
The pinkish
galaxy are
line the spiral arms.
H II
NGC
and
mottling,
spots scattered
regions that out-
253
belongs to a
cluster of galaxies called the Sculptor group,
nearly
10
million tight years
from Earth.
(Anglo-Australian Observaloryj
The hazy band across

of the disk of our
learned about the
the night sky called the
own
galaxy. In this chapter
size,
shape,
and
Milky Way
we
see
is
our edge-on view
how astronomers have
rotation of this galaxy.
We
learn about the
arms of our galaxy, cmd we probe the mystegalactic center. These studies of uur own galaxy
processes that produce the spiral
rious source of energy at the

prepare us to explore other galaxies in the next chapter.
On
a clear, moonless night far from city lights, you can often see a hazy,
luminous band stretching across the sky. Ancient peoples devised fanciful myths to account for this "Milky Way" aiiKJug the constellations.
Today we realize that this hazy band is actually our view from inside a
vast disk-shaped assemblage of several hundred billion stars that includes
the Sun.
In studying the Milky Way, we explore the universe on a large scale

and ask comprehensive questions. Instead of focusing on the location
and life of an isolated star, we look at the overall arrangement and history of a huge stellar community of which the Sun is a member.
Uiir galaxy
The Sun
is
located in the
Claiilco
was the
He
first
person
lo look
.it
the Milk\ \\a\ tluough a tele-
disk of our galaxy, about
scope.
30,000 light years from the

galactic center
The Milk) Way stretches all the way around the sky in a continuous band that is almost perpendicular to the plane of the ecliptic. Figure
16-1 is a wide-angle photograjjh showing loughlv hallOf the Milky Way.
Because the Milky Way compleleK enciicles us. astronomeis in the
eighteenth century began to suspect that the Sun and all the stars in the
sky are part of an enormous disk-shaped assemblage called the Milky
Way Galaxy. In the 1780s, William Herschel attempted to deduce the
Sun's location in the galaxy by counting the number of stars in 683 regions of the sky. He reasoned that the greatest density of stars should be
seen toward the galaxy's center and a lesser density seen toward the
edge of the galaxy. Herschel found roughly the same density of stars all
along the Milky Way and concluded that w'c are at the center of our galaxy. We now kno\v that he \\as wrong.
The reason for Herschcl's mistake was revealed in the 1930s by R. ].
Trumpler. While studying star clusters, Trumplcr discovered that remote
clusters appear unusually dim, more so than would be expected from
their distance alone. Trumpler therefore concluded that interstellar space
must not be a perfect vacuum: it contains dust that absorbs light from
iniinediately discovered that
it
is
com|)oseci oi countless
dim
stars.
distant stars. Like the stars, this obscuring material
is concentrated in the
plane of the galaxy.
Great patches of this interstellar dust are cIcarK \ isible in wide-angle
photographs such as Figure 1()-1. At optical wavelengths, the center of
the galaxy is totally obscured from our view. The absorption of starlight
by interstellar dust misled Herschel. Because he was actually seeing only
the nearest stars in the galax\ he had no line idea either of the enormous size of the galaxy or of the \ast nimiber of stars concentrated
,
around the
Because
galactic center.
interstellar dust
the absorption of starlight

Figure 16-1
The Milky Way
angle jilwlograph
\li(nf\ ihe
Milky
tng jrum Sagillanti.s on the right
on the
left.
dtist that
Wat
to
Note the dark lanes and
This mottling
stars.
Tins unde-
is
extend-
Casswpem
blotches.
caused by interstellar gas and
obscure Ihe light from background
(Steward Observatory)
is
is
concentrated
in the
plane
of
our galaxy,
strongest in those parts of the sky covered
by the Milky Way. However, our view is relatively imobscured to either

Way. Knowledge of our position in the galaxy came
from observations of globular clusters in these imobscured portions of
the sky. Before we turn to those observations, we must review the nature
of variable stars.
In 1912, the American astronomer Henrietta Leavitt reported her
important discovery of the period-luminosity relation for classical (Type
side of the Milky
Our
I) Cepheid variables. As we saw

Cepheid variables are pulsating
acteristic
lanic
in
Chapter 13
(recall
galaxy
Figure 13-21),
stars that vary their brightness in a char-
way. Leavitt studied numerous Cepheids in the Small Magel(a small galaxy very near the Milky Way) and found that
Cloud
their periods are directly related to their average luminosities (see Figure
16-2).
Today, astronomers
two kinds of Cepheid

Cepheids that Leavitt studied, and the
Cepheids. As shown in Figure 13-22, the Type II
variables: the metal-rich
metal-poor Type
_l_l
4
Period
Figure 16-2
20
10
40 60
The peiod luminosity

This graph
relation for classical Cepheids
shows the relalwmhip between the periods and

luminosities of classical (Type I) Cephetd variables.
Each
dot represents a Cepheid in the
Small Magellanic Cloud whose brightness and

period have been measured. The line
"best fit" to the data.
is
the
(Adapted from
H. C. Arp)
slightly
tool in
astronomy be-
can be used to determine distances. For example, suppose you

find a Cepheid variable in the sky. By measuring its period and using a
graph such as Figure 16-2, you promptly discover the star's average luminosity. This measure of the star's true brightness can easily be expressed as an absolute magnitude. Meanwhile, you can observe the star's
apparent magnitude. Since you now know both the apparent and the
absolute magnitudes, you can calculate the star's distance.
Shortly after Leavitt's discovery, Harlow Shapley, a young astronomer
at the Mount Wilson Observatory in southern California, became interested in a family of pulsating stars called RR Lyrae variables that are
quite similar to Cepheid variables. RR Lyrae variables are commonly
found in globular clusters (see Figure 16-3). Shapley guessed that these
variables are simply short-period classical Cepheids like those Leavitt had
cause
(days)
II
dimmer than the Type I.

The period-luminosity law is a very important
Cepheids are
realize that there are
Type
it
studied.
By 1915, Shapley had noticed a peculiar property of globular clusters.

stars and open star clusters are rather uniformly spread along
the Milky Way. The majority of the globular clusters that Shapley studied were preferentially located in one half of the sky, however, widely
scattered around the portion of the Milky Way in Sagittarius. Figure
Ordinary
16-4 shows two globular clusters in this part of the sky.
Shapley used the period-luminosity relation to determine the

tances to the then-known 93 globular clusters in the sky.
rections
and
distances,
of these clusters
figure 16-3
Three
RR
anozus in
The globular
cluster
M55
Lyrae variables are identified with

this
globular cluster located in the
constellation of Sagittarius.
From
apparent brightness (as seen in
the average
this
photo-
graph) and the average true brightness (known

to be
roughly
100
suns), astronomers deduce
that the distance to this cluster

years.
(Haivard Obseivatory)
is
20,000
light
From
dis-
their di-
he mapped out the three-dimensional distribution

By 1917, Shapley had discovered that the glob-
in space.
Our
Figure 16-4
view toward the galactic
More than
center
295
galaxy
million stars are visible
looh toward a
"window" just 4 south of the
in this photogiaph. This view

relatively clear
galactic nucleus in Sagittarius. There
pnsingh
obscuring matter in
little
The two globular
tion of the sky.
XGC
6522 and
\GC
6528.
is stir-
this lin\ sec-
clusters are
(XOAO)
JU,UUU
o
ly
o
o
o
Galactic
nucleus
ular dusters
1
t
2000
""
Iv
o
o
Globular clusters
(edge-on view)
bulge,
to
o
o
_^
Thin laver
of dust
is
There are three mnjur camfm-
halo.
The
a central
disk,
disk contains gas
dust along with metal-nch (population
The halo
Earth.
our gala.x\: a thin
and a
The structure of the galaxy
Figure 16-5
nents
o/
ôô
o/\^C^^
o^^^
"
oo
and
/) stars.
composed almost exclusively of old,

stars. The central
metal-poor (population II)
bulge around the galactic nucleus
is
of population I and population II
slurs.
a mixture
Edge-on view of the galaxy

we could view our galaxy
from a great distance, il would proba-
Figure 16-6
NGC
4265
edge-iin
bly look like
Coma
is
If
tim galaxy in the constellation nf
Berenices.
clearly visible
thin layer of
dmt and
gas
the plane of the gala.xy.
Also note the reddish color of the bulge thai
surrounds the gaUm's nucleus.
Naval Observatory)
(L'.S.
The
form a huge spherical system
that
is
not centered on the
clusters are instead centered about a point in the Milkv
Way
toward the constellation of Sagittarius. Shaplev then made the bold conjecture (subsequently confirmed) that the globular clusters outline the
true size and extent of the galaxy.
Today we know that the disk of our galaxs is about 100. 000 light
\ears in diameter and about 2000 light years thick (see Figure 16-5). The
galactic nucleus is about 30.000 light vears from Earth and is surrounded h\ a spherical distribution of stars called the central bulge. The
spherical distribution of globular clusters outlines the halo of the galaxv.
If we could view our galaxy edge-on from a very great distance, il would
probably look much like the galaxy shown in Figure 16-6.
Oui
The
spiral structure of the

galaxy has been determined
from radio observations
Parallel spins
Antiparallel spins
Figure 16-7
Electron spin
hydrogen atom
hydiugrn
nliim.
In
Ilir
can be spinning
directions.
atom
llir
and
liiu'i-.!
tippii\ilr
the elativn Jlijn nx'n
either gains or loses a tiny
energy. This energy
llif
iif
rlnlinn anil llw pmtun
filhei in the \aiiir
When
the
mini
i.s
amount
l/ir
of
either absorbed or emit-
ted as photons with wai'elengths of
21 cm.
galax\
Because interslellar dust effccti\el\ obscures our visual \iews in the

plane of the Galaxy, a detailed understanding of the structure of the galactic disk had to await the development of radio astronomy. Because of
their long wavelengths, radio waves easily penetrate the interstellar medium without being scattered or absorbed. As we shall see in this section,
radio observations reveal that our galaxy has spiral-shaped concentrations of gas and dust called spiral arms unwinding from the center in
the shape of a pinwheel.
We have seen that hydrogen is by far the most abundant element in
the universe. By looking for concentrations of hvdrogen gas, we should
therefore detect important clues about the structure of the disk of the
galaxy. Unfortunately, the major electron transitions in the hydrogen
atom (see Figure 5-11) produce photons at ultraviolet and visible wavelengths that do not penetrate the interstellar medium. What hope do we
have of detecting all this hydrogen?
In addition to mass and charge, particles such as pnjtons and electrons possess a tiny amount of angular momentum cominonlv called
spin. .\n electron or a proton can be crudely \isualized as a tiny spinning sphere. According to the laws of quantum mechanics, the electron
and proton in a hydiogen atom can be spinning either in the same or
opposite directions (see Figure 16-7), but they can have no other spin
orientations. If the electron flips over from one configuration to the
other, the hydrogen atoin must gain or lose a tinv amount of energy.
For example, in going from parallel to antiparallel spins, the atom emits
a low-energy photon whose wavelength is 21 cm. In 1951, a team of astronomers succeeded in detecting the faint hiss of 21 -cm radio static
from
interstellar
hvdrogen.
The
detection of 21 -cm radio radiation was a major breakthrough

that permitted astronomers to map the structure of the galactic disk. To
see how this is done, suppose that vou aim your radio telescope along a
particular line of sight across the galaxy as sketched in Figure 16-8. Your
radio receiver picks up 21-cm emission from hvdrogen clouds at points
1, 2, 3, and 4 (point 5 is the location of the Sun). However, the radio
waves from these various clouds are Doppler shifted bv slightly different
amounts because they are moving away from vou at different speeds as
they travel with the rotating galaxv.
The final result is that the 21-cm radiation is smeared out over a
range of wavelengths. Because radio waves from gas in different parts of
the galaxy arrive at your telescope with slightly different wavelengths, it
is
possible to sort out the various gas clouds
Figure 16-8
galaxy
technique for
Hydrogen
along our
line
cloiiils at
mapping our
different liictitwns
moving
Radio waves from
of sight are
different speeds.
gas clouds are therefore subjected

different
Doppler
tronomers
the miUixy.
to sort
shifts,
at slightly
the various
to slightly
permitting radio as-
out the gas clouds
and map
map
and therebv map the galaxy.
of the galaxv produced in this fashion.

The arrangement of long, curved lanes of hydrogen gas in the map
clearly suggests that our galaxy has a spiral structure. Further clues
about these spiral arms are obtained from studying the locations of OB
associations and H II regions in the sky. Unfortunately, interstellar absorption limits visual observations in the plane of the galaxv to a range
of less than 10,000 light years from the Earth. Nevertheless, from looking at photographs of other galaxies we know that these bright stars and
nebulae outline spiral arms. From mapping the locations oi ncarbv O
and B stars and H II regions, astronomers have discovered that the Sun
is located between two major spiral arms. One spiral arm (the Sagittarius
arm) dominates our view of the Milky Wav during the summer, while
the other arm (the Perseus arm) is best seen during winter months. If
we could view our galaxy from afar, it would probably look somewhat
like the spiral galaxv shown in Figine l(i-l().
Figure 16-9
is
Our
A map
Figure 16-9
map
based
w(is
uii
of our galaxy
I'liis
radw-tehscope M/n'ns u]
21-cm
radiation. The distribution of h\drogen

gas across our galaxy clearly reveals spiral
The Sun's
structure.
the symbol
O.
location
is
indicated with
Details in the large, blank,
wedge-shaped region toward the bottom of the

map are unknown because gas in this part o)
the sky
sight
is
and
Doppler
moving perpendicular
If
our line
The spiral galaxy
we could
I'lni'
from a great distance,
il
NGC
our gala.\y jace-on
would
look
somewhat
like this spiral gala.xy in the constellation of
(Agnus. The
(U.S.
.spiral
arms are outlined by nu
and OB assocuilmns
Naval Obsen'alory)
merous
of
(Courtesy of G. Westerhout)
Figure 16-10
6946
to
thus does not exhibit a detectable
shift.
II regions
galaxy
Our
Moving
galaxy
The
existence of spiral arms suggests that galaxies rotate. If our galaxy

were not rotating all the stars would fall into the galactic center. However, detecting and measuring the rotation of our galaxy has been a dif-
at half a million
miles per hour, the Sun

takes 230 million years
to complete one orbit
of our galaxy
ficult business.
Radio observations of 21 -cm radiation from hydrogen gas give imporBy measuring Doppler shifts, astronomers can at least determine the speed of objects parallel to our line
of sight across the galaxy. These observations clearly indicate that our
galaxy does not rotate like a rigid body but rather exhibits differential
tant clues about the galaxy's rotation.
rotation: stars at different distances
from the
galactic center travel at dif-
ferent orbital speeds about the galaxy.
Galactic center
Figure 16-11
galaxy
Differential rotation of the
Different purls of ihe galaxy are ro-
tating at different speeds. Thus, stars traveling
more quickly than
the
Sun appear
mov-
to be
ing in one direction, whereas stars traveling
more slowly than
the
Sun appear
to
be
moving
in the opposite direction.
J
20,(KI0
4(1.000
60,000
Distance Irom galactic center
(ly)
Figure 16-12
The
orbital
served
The galaxy's rotation curve

velocity of stars and gas is ob-
to increase
60,000
out to a distance of at least
light years
Va.st quantities
therefore
from
Ihe galactic center.
of sublumitunis matter must
surround our galaxy.
Further clues to this rotation come from examination of the motions

of stars in the sky by the Dutch astronomer Jan H. Oort. Because of differential rotation, the Sun is like a car on a circular freeway with the fast
lane on one side and the slow lane on the other. As sketched in Figure
16-11, stars in the fast lane are passing the Sun and thus appear to be
moving in one direction, while stars in the slow lane are being overtaken
by the Sun and therefore appear to be moving in the opposite direction.
Unfortunately, like the 21 -cm observations, this study reveals only
how fast things are moving relative to the Sun. Of course, the Sun itself
is moving. To get a complete picture of our galaxy's rotation, we must
find out how fast the Sun is traveling.
A method of computing this speed was proposed by the Swedish astronomer Bertil Lindblad. Not all the stars in the sky move in the orderly pattern in Figure 16-11. Distant galaxies and some components of
our own, such as globular clusters, do not participate in the general rotation of the galaxy; instead they seem to have more or less random motions. Using the average of these random motions as a background,
Lindblad determined that the Sun moves along its orbit about the galactic center at a speed of 250 km/sec, or about i million miles per hour.
Knowing the Sun's speed and its distance from the galactic center, astronomers easily calculated the Sun's orbital period to be 2.3 x 10* years.
Traveling at i million miles per hour, our Sun takes 230 million years to
complete one trip around the galaxy. These results demonstrate how
vast our galaxy is.
Knowing the details of our orbit around the galaxy, we can use Kepler's third law to estimate the mass of the galaxy. Putting in the numbers, we obtain a mass of 1.3 x 10" Mq.
This estimate for the galaxy's mass must be too low because Kepler's
law gives us only the mass inside the Sun's orbit. The matter exterior to
the Sun's orbit does not affect the Sun's motion and thus cannot be calculated from Kepler's law. Obviously there is matter out there. In recent
years, astronomers have been astonished to discover just how much matter actually lies beyond the orbit of the Sun.
Because we know the true speed of the Sun, we can convert the
Doppler shifts measured by radio astronomers into actual speeds of the
spiral arms. This computation gives us a rotation curve, a graph of the
velocity of galactic rotation measured outward from the galactic center
(see Figure 16-12).
This graph is surprising because the rotation curve keeps rising, even
out to a distance of 60,000 light years from the galactic nucleus. Thus
we still have not detected the actual edge of our galaxy. According to
Kepler's third law, the orbital speeds of stars or gas clouds beyond the
confines of most of the galaxy's mass should decrease with their distance
away from the galaxy's center, just as the orbital speeds of the planets
decrease with the planets' distance away from the Sun. Instead, galactic
Our
galaxy
299
orbital speeds continue to climb well

lactic disk.
surprising
beyond the
visible
edge of the ga-
amount of matter must therefore be
scattered
around the edges of our galaxy. In fact, our galaxy's mass could easily
be at least 6 x 10" A/q. To make matters even more mysterious, this
outlying matter is dark and does not show up on photographs. .Astronomers suspect that this hidden matter is spherically distributed all aiound
the galaxy along with the globular clusters. Thus our galaxy's halo is
moie massive than previously expected. The nature of this hidden masswhcther black holes, gas. or dim stars is so far a complete mysterv.
Spiral arms are produced by

density waves that sweep
across the galaxy
That spiral arms should exist at all was another mystery that confoimded
astronomers for years. .Many galaxies exhibit these beautiful arching
arms outlined by brilliant H II regions and OB associations (recall Figure 16-10). As we think through the effects of a galaxy's rotation, a dilenmia arises. All spiral galaxies have rotation curves similar to our own.
As Figure 16-12 demonstrates, the velocity ol stars and gas in a galaxy is
fairly constant over a large portion of the galaxy's disk. Howe\er. the
farther stars are
from
a galaxy's center, the farther they
must
travel to
complete one orbit of the galaxy. Thus stars and gas in the outskirts of a
galaxy take much longer to complete an orbit than does material near its
center. Consequently, after a few galactic rotations, the spiral arms
should wind up and disappear altogether.
The enigma of spiral structure intrigued several brilliant and ingenious scientists. Among the first was Bertil Lindblad, who in the 1920s
argued that the spiral arms of a galaxy are merely a pattern that moves
among the stars, just as waves on the ocean are a moving pattern. As the
waves move across the surface of the water, the individual water molecules siniplv bob up and down in little circles as the wa\es ripple b\
The
waves are simply a pattern that moves across the water no water actually travels with the wave pattern. Lindblad in fact used the term density
.
waves
in discussing the possible cause of spiral structure.

This density-wave theory was greatly elaborated upon and mathematically embellished by the American astronomers C. C. Lin and Frank .Shu
in the mid-1960s. Lin and Shu argued that density waves passing
through the disk of a galaxy cause material to "pile up" temporarih A
spiral arm therefore is simply a temporary enhancement or compression
.
of the material in a galaxy.
This situation is analogous to a temporarv traffic jam created by

workers painting a line down a busv freeway where cars normally cruise
at 55 mph. .An individual car spends only a few moments in the traffic
jam before resuming the usual 55-mph speed. The traffic jam in the vicinity of the painters lasts all day long, however. The traffic jam
which
would be seen so clearly from an airplane is simply a temporar\ enhancement of the number of cars in a particular location.
To understand better how a density wave operates in a galaxv think
once again about the ocean. If the water molecules were left completely
undisturbed, the surface of the ocean would be perfectlv smooth. In reality howe\ei, these molecules arc constantly bufleted bv small disturbances, or perturbations, such as the wind. Because of a pcrtmbaiion,
one molecule pushes the next, wliiih pushes the next, and so on. The
result is a water wave. Individual molecules on the surface of the ocean
move in tiny elliptical paths as the wave pattern passes across the water
(see Figure I6-i;5).
In a galaxy, the stars are separated by vast distances. Collisions bestars virtually never happen. Nevertheless, stars do interact be-
tween
Our galaxy
Figure 16-13
waves
rotates
level in
Water waves and kinematic
about a point on the undisturbed water
a liny
ellipse,
(b) Similarly, a small
disturbance in the orbit of a star can cause the

star to oscillate in tiny ellipses about
nal
orbit.
precessing
a Water wave
In a water wave, each molecule
(a)
The actual path of

ellipse.
its
the star
(Adapted from
,4.
is
"^
origi-
Toomre)
cause they are affected by each others gravity. In water waves or sound
waves, molecular forces are responsible for orchestrating the motions of
molecules. In a galaxy, the force of gravity controls the interactions be-
tween stars.
Seen from above, the undisturbed orbit of a star about the center of a
galaxy would be a nearly perfect circle. However, the motions of other
matter in the galaxy produce small gravitational perturbations that cause
the star to deviate from its undisturbed orbit. Just as a water molecule
bobs up and down on the surface of the ocean, a star oscillates back and
forth about its undisturbed orbit. Lindblad demonstrated that these oscillations can be described by thinking of the star as attached to a tiny epi-
As sketched in Figure 16-13i. the star rotates counterclockwise

around the epicycle while the epicycle itself is moving clockwise along
cycle.
the undisturbed path.
The final path

Of course, the
of the star is a precessing ellipse,

gravity of this star affects the motions of its neighbors, so a wave disturbance, called a kinematic wave,
propagates from one stellar orbit to the next.
In 1973, Agris J. Kalnajs in Australia bridged the gap between
Lindblad's ideas and the density-wave theory of Lin and Shu. Kalnajs
argued that the precessing elliptical orbits of stars are not randomly oriented, as sketched in Figure 16-14. Instead, there is a strict correlation
between orbits: each precessing elliptical orbit is tilted with respect to its
neighbor through a specific angle. The result is shown in Figure 16-14/*,
where a beautiful spiral pattern has emerged.
This spiral pattern arises wherever the ellipses are bunched closest
together. Of course, stars in a galaxy are scattered randomly along their
orbits. With correlated orbits, however, some of the stars happen to get
close together along the huge, arching spiral arms. After all, these spiral
arms are where the stars' orbits are closest together for the longest
an
ellipse that rotates.
stretches.
Figure 16-14
The origin of spiral density
waves Both drawings hai'e exactly the same

number of ellipses, each representing the urbit
of a
star, (a)
Ellipses with
Randomly
oriented ellipses, (b)
a correlation between the orienta-
tions of adjacent ellipses.
This temporary enhancement of stars has a profound effect on the

and dust. Because of the presence of extra stars, there is
an increased gravitational attraction all along the spiral. This added
force has almost no effect on the massive stars, which simply continue to
lumber along their orbits. The lightweight atoms and molecules in the
interstellar medium are readily sucked into the gravitational well along
the spiral, however, forming the crest of the density wave.
interstellar gas
Our galaxy
301
A density wave moves through tlie material ol a gaiax\ at a speed

greater than the speed of sound in the interstellar medimn. lluis, the
density wave is supersonic. As in the ease of a supersonic airijlaiic. a
shock wave is created along the leading edge of the density wave. Shock
waves ate charactci ized by a sudden, abrupt compression of the medium
through which they mcjve. which is the "scjnic boom" we hear from a
supersonic airplane. The interstellar medium is violentiv compressed bv
the cosmic sonic boom of the density wave.
This densitv-wave theorv seems to explain man\ of the properties ol
As sjîtal density waves sweep through the plane ol a

galaxy, they recycle the interstellar medium. Old gas and dust left bespiial structure.
sMim
Spiral
luhiil
arm
of hydrogen
Figure 16-15
Shock fronts and density

A shmk iitiic iiccum/Mitiics a spiral ileiisily wave as it s-weeps
aroumi the galaxy. The resulting spiral arms
wai'es in our Galaxy
behind the galactic rotation (clockwise in
trail
this
drawing).
hind from dead stars are compressed into new nebulae to form new
The sprawling dust lanes alongside the string of emission nebulae
outlining a spiral arm attest to the recent passage of a compressional
shock wave. Because the material left over froin the deaths of ancient
stars is enriched in heavy elements, new generations of stars are more
metal-rich than were their ancestors. The overall structure of spiral
shocks and density waves in our own galaxv is shown schematicallv in
Figure 16-15.
We have by no means solved all the c|uestions about spiral structure in
galaxies. .Vfanv astronomers believe that the densitv-wave theorv is on the
right track, but many gaps remain in our undei standing. For example,
what keeps the density waves going why don't thev simpiv fade awav?
Density waves expend an enormous amount of energv to compress the
interstellar gas and dust. In order to keep these waves going, there must
be a constant replenishing of that eneigy. We do not understand where
this energy comes from, but the nuclei of galaxies seem to be the place
stars.
to look.
The
may
galactic nucleus
contain a supermassive
black hole
The
nucleus of our galaxy is an active, crowded place. The stars in Figure 16-4 give some hint of the stellar congestion. If you lived on a
planet near the galactic center, you would see a million stars as bright as
our own nighttime sky. The total intenthose nearby stars would be eciuivalent to 2(10 of
Sirius, the brightest single star in

sity of starlight
from
all
our full moons. Night would never really fall on a planet near the center
of our galaxy.
Because of seveie interstellar absorption at visual wavelengths, some
of Our most important information about the galactic nucleus comes
from radio and infiared observations rather than from visible light. Pioneering radio observations were made b\ Jan H. Oort and C>. W.
Rougoor in 1960. By observing I)op|)lei shifts in '21 -cm radiation. Oori
and Rougoor discovered two enormous expanding arms of hvclrogen
gas. One arm is located between us and the galactic center and is appioaching us at a speed of 3'^ km/sec. The other arm is on the other
side of the galactic nucleus and is receding from us at 135 km/sec. The
total amoimt of hvdiogcn in these expanding arms is at least several million solar masses. Something c|uite extraordinarv must have happened
about 10 million \ears ago to ex|)el such an enormous amount of gas
from the center of our galaxy.
In addition to the 21 -cm radiation coming from the expanding arms,
radio astronomers also detect a vast amount of radio noise emitted directly from the galactic nucleus. The radio noise does not come from
hydrogen, however. It is produced bv high-speed elections s|)iraling
around a magnetic Held. This kind of radio emission is called synchrotron radiation. Its powei ful source at the galactic center, named Sagiiiar-
Our
The galactic center at
Figure 16-16{a)
visual wavelengths
centered
oti
Thii wule-angle view
the nucleus of
"stinger" at the
end of the
tail
of Scorpnis
and
at the bottom of the photograph,
pot" of Sagillarius
beloic the trail
slightly left
of
of an Earlh-orbilmg
is
The white rectangle shows

by the infrared photograph
(U.S.
(\
our galaxy. Thv

""X^
i\
the "lin
center,
/i,
satellite.
the region covered
on Figure 16-16h.
Naval Obsen'alory)
'.
"
M^:.
'^'^^^BS^J%:âiaclicocntrr
Figure 16-16(b)
The galactic center
[left]
at infrared wavelengths
This "false-color"
infrared vine from IRA.S covers a field about
48" X 33. Black represents the dimmest
re-
gions of infrared emission, with blue the nexl
dimmest, followed by yellow
and
red, with
while for the strongest emission. The numerous

knots
and
blobs
along the plane of the galaxy
are interstellar clouds of gas

by nearby stars.
and
dust heated
(NASA)
Figure 16-16{c) [right]
The galactic
nucleus at infrared wavelengths

of duit
around
in blue in this "false-color" view

tliat
Streamers
the galactic center are visible
from IRAS
covers the region indicated by the white
rectangle in Figure 7 6-7 6b.
The
bright object
near the lower right
is a large
II region
about 7 southwest of the galaccenter. The prominent band across this view
(NGC 6357)
tic
and
the previous
in the
IRAS
view
plane of the galaxy.
is
a layer of dust
(NASA)
galaxy
Our
j;laxy
303
ius A. is one of the brightest radio sources in ihe entire skv. Recent
observations suggest that Sagittarius A is ver\ small. In spile ol its
enormous energy output,
it is only 40 light vears in diameter.

Figure 16-16 shows three dillereni views ol the center of our galaxy.
Figure I6-I60 is a wide-angle photograph of the Milkv Way around Sag-
ittarius at visible
wavelengths. Figure 16-16/; is an infrared view ol the

by the white rectangle in Figure 16-1 60. Ihe prominent
band across this view is a thin layer ol dust in the platie of the galaw
(recall Figure 16-5). Figure 16-16r is a detailed infrared view of the galactic center, covering the area outlined b) the white rectangle in Figure
16-16i, The strongest infrared emission (in white) conies from Sagittarius A. which is also a \er\ powerful soiuie of radio waves. Fhe galactic
center, believed to coincide with the western portion of this somce, is
called Sagittarius A West.
The center of our galaxv is a verv active, dynamic place. Numerous
infrared sources scattered along the galactic plane and around the galactic center are believed to be giant clouds of gas and dust heated b\
groups of young O and B stars. In addition, much of the gas aroimd the
galactic nucleus is extremelv rich in molecules. The giant molecular
cloud called Sagittarius B2 is a favorite hunting ground lor radio astronfield indicated
omers searching
for
new
interstellar chemicals.
Important clues about the dvnamics of the center of our galaxv have
come recently from high-resolution infrared and radio observations. Figure 16-17 shows a detailed view of the galactic center obtained in 1983
by radio astronomers at the Very Large .Arra\ .\ second team of astronomers used NASA's infrared telescope on Hawaii to measure the Doppler shifts of a spectral line of neon in the gas that lies within a few light
years of the galaxy's center. Measurements revealed that the gas along
the S-shaped feature seen in Figure 16-17 is orbiting the galactic center
at an exceptionallv high speed.
.
Figure 16-17
The galactic nucleus
coliii-rmlnl mdiii maji

III)'
cetilcr
(ij
our
f;ril<ixy iil
llie
slroujrrst rmiwioti
is
iimveletij^th of
11
6 rm. Strong radio ymissivn
is
shown
while.
siom are colored green through

bright object at Ihe middle of
coiiifxirl
This
iipJM'auiiKi- a/
'liini's Ihi-
in
ml.
Weaker emi\
blue.
lliis
The
a
I'ieii'
radio source jneci.seh at Ihe center nj
our galaxy. The area covered in /Am view

measures about
oj K. Y.
1(1 light years across.
hi and
.\I.
/.
Claus.sen)
(Courlis^
Our
galaxy
Something must be holding this high-speed gas in orbit about the gaUsing Kepler's third law, astronomers estimate that 5 x 10*'
Mq of material is needed to pre\ent this gas from flying off into interstellar space. These observations imply that an object with the mass of
5 million Suns is concentrated at Sagittarius A West. This object must be
extremely compact much smaller than the few light years over which
the observations extended. Many astronomers vigorously argue that an
object this massive and this compact could only be a black hole. Because
of its enormous mass, it is called a supermassive black hole.
Whether our galaxy's nucleus actuall\ liailjors a supermassive black
hole will be decided onh after man\ years of observation, analysis, and
debate. All non-black-hole explanations of the data must be ruled out, or
at least rendered most improbable. Nevertheless, as we shall see in Chaplactic center.
ter 18,
astronomers find \iolent activitv occurring at the nuclei of other

seeming to indicate the presence of supermassive black
galaxies, also
holes at those galactic centers.
Summary
Ihe disk of our galaxy is about 100,000 light years in diameter and about
2000 light years thick. There is a high concentration of interstellar dust
and gas in the galactic disk.
The
galactic nucleus
is
surrounded by a spherical distribution of stars

The entire galaxy is surrounded by a spherical
called the central bulge.
distribution of globular clusters called the halo of the galaxy.
The Sun
is
tween two
located about 30,000 light \ears
Interstellar dust obscures
sight that
from the
galactic nucleus, be-
spiral arms.
lie in
our view
at visual
wavelengths along
lines
of
the plane of the galactic disk.
The
galactic nucleus has been effectively studied at infrared and radio

wavelengths, which pass readily through intervening interstellar dust.
Hydrogen clouds outlining our galaxy's spiral arms are detected by anaIvzing the 21 -cm radio waves emitted in the spin-tlip transition in the hydrogen atom.
Spiral
arms are probably caused bv density waves
that
sweep around the
galaxv.
A spiral arm compresses interstellar clouds through which

compression triggers the formation of OB associations and
it
passes. This
II
regions
that illuminate the spiral arms.
supermassive black hole with a mass of about 5 x
10''
Mq
possibly
exists at the galactic center.
Review questions
Why
do you suppose
that the Milky
Way
Way appear
the
is
far
more prominent
in July
than in December?
2 How would the Milky
the galaxy?
if
Sun were
located at the edge of
3 Why don't astronomers detect 21 -cm radiation from the hvdrogen

molecular clouds?
in giant
4 Describe the Doppler shifts of the 21-cm line that vou would observe
the galaxy were rotating like a rigid body.
if
Our
fjdlfixy
why
5 Explain
halo,
spend
gloljular clusters
even though iheir eccentric
niosi of their lime in the galactic
(jrbiis
take
them verv
close to the galactic
center.
6 Approximaleiy
how nianv
Sun and
the galaxy since the
How
would you estimate the
Why
hasn't the supermassive black hole at the center of
in
Advanced questions
times has the solar system orbited the center ol

planets were formed?
and swallowed
all
total
number
of stars in the galaxy?
our galax\ pulled
the stars in the galaxv?
What can you surmise about
galactic evolution from the fact that the gadominated by population II stars while population I stars are
predominandy found in the galactic disk?
lactic
halo
is
*10 The galaxv is about 30, ()()() parsecs in diameter and 600 parsecs thick. If
supernovae occur randomly in the galaxy at the rate of about five each century, how often on the average would we expect to see a supcrno\a within
300 parsecs (1000 light years) of the Siui?
1 1
Speculate on the reasons for the rapid rise in the (Jalaxv's rotation
nuve
(see Figure 16-12) at distances close to the galactic center.
12
From what you know
medium, and
and structure of spiral arms
at)out stellar evolution, the interstellar
the densitv-wave theory, explain the appearance

in spiral galaxies.
13
What
observations would you propose in order to determine the nature of
the hidden mass in our galaxy's halo?
For further reading
Bok, B. "Our Bigger and Better Galaxv." Meirur\. Sept./Oct. 1981.
p.
130.
Bok,
B..
and Bok.
P.
Tlw Milk\
\V(i\.
,")th
ed.
Hai vaid L'niveisitv Press,
1981.
Chaisson, E. "lournev to the Center
ot tiic Clalaxv." Aslii)iiom\. .Aug.
1980,"p. 6.
(leballe, T.
"The Central Parscc of
the C.alaxy." Scientific Amt'iicaii. July
1979.
Weaver. H. "Steps Toward L'lKierstanding the l,arge-Scale Structure of

the Milkv Wav. Mciciin. Sept./Oct. 1975. p. 18; Nov./Dec. 1975.
"
p.
18: jan./Eeb.
197(i, p.
19.
17
Galaxies
The spiral galaxy M66 Tins galaxy is one

member of a cluster of galaxies in Leo at a
30 million light years from
distance of about
Earth. The galaxy's dislorlnl ajipeurana- mini
from
bors.
giavilaluiiiiil iiiliiiulnnis ivilh ih mn^li-
which nuliule the spiral galaxies .\/6i
and NGC 3628. Spiral galaxies are usually

found in sprawling clusters with few members.
Rich
clusters with
thousands of members are
usually dominated by elliptical galaxies. (U.S.
Naval Obsenialory)
Turning our
ies exist
in
attention to the universe beyond our galaxy,
many shapes and
sizes.
We
we find
that galax-
learn that galaxies are grouped in
clusters rather than being scattered randomly through space, and we discuss
some of the spectacular phenomena associated with collisions of galaxies.
We find that there is a simple relationship between the distance to a
galaxy and the speed with which it moves away from us as measured by
its redshift. In our discussion of this relationship, we learn something about
the measurement of intergalactic distances.
William Parsons was the third earl of Rosse in Ireland. He was rich, he
liked machines, and he was fascinated with astronomy. Accordingly, he
set about building gigantic telescopes. In Febriiarv 1845, his piece de resistance was finished. This telescope's massive mirror measured 6 ft in
diameter and was mounted at one end of a 60-ft tube controlled by cables, straps, pullevs, and cranes. For a brief period of time. Lord Rosse's
and
contraption enjoyed the dubious reputation of being the largest
most dangerous
telescope
in the
world.
During this period. Lord Rossc examined manv of the nebulae discovered and catalogued by William Herschel. Lord Rosse observed that
some of" these nebulae have a distinct spiral structure. Perhaps the best
example is M51 (also called NGC 5194. the 5194th object in the Xeiv
General Catalogue, which is a list of all the nebulae and star clusters observed by William Herschel, his son John Herschel. and others).
Because Lord Rosse did not have any photographic equipment, he
had to make drawings of what he saw. His drawing of M51 is shown in
Figure 17-1. and a modern photograph appears in Figure 17-2. Xiews
such as these inspired Lord Rosse to echo the famous (German philosopher Immanuel Kant, who in 1755 suggested that these objects might be
"island universes"
vast collections of stars far bevond the confines of
the Milky Way.
Many astronomers did not subscribe to this notion of island universes.
Many of the objects listed in the NGC were in fact nebulae and star clusters scattered about the .Milky Way, and it seemed just as likelv that these
intriguing spiral nebulae could also be members of our galaxv.
The astronomical community became increasingly divided over the
nature of spiral nebulae. .\ debate on the topic was scheduled in .\pril
1920 at the National .-Xcademv of Sciences in Washington. D.C. On one
side was Harlow Shaplev. the young, brilliant astronomer renowned for
his recent determination of the size of the Milks Way galaxy. Shaplev
believed the spiral nebulae to be relatively small, nearbv objects scattered
around our galaxy like the globular clusters he had studied. Opposing
Shaplev was Heber D. Curtis of the Lick Obser\atorv near San Jose, California. Curtis championed the island-universe theorv and argued that
each of these spiral nebulae is a rotating svstem of stars much like our
Figure 17-1
MSI
sign.
[left]
Using a large
Lord Rosse's sketch of

telescope oj lus oicu de-
Lord Rosse was able
to distinguish the
spiral structure of this "spiral nebula." (Courtesy
of
Lund Humphrie\l
Figure 17-2 [right]
M51
The spiral galaxy
This spiral galaxy in the constellation
of Canes Vantici is often called the Whirlpool

Galaxy because of its distinctive appearance.
Its
distance
light years.
spiral
arms
from Earth
The blob at
is
is
about 15 million
the
end of one of
a companion gala.xy. (U.S.
Naval Observatory)
the
own galaxy.
The Shapley-Curtis debate generated much
heat but little light. .Nothing could be decided because no one had anv firm evidence to demonstrate exactly how far away the spiral nebulae are. .Astronomv desper-
needed definitive measurements of the distances to the spiral

nebulae. .Making such measurements was the first great achievement of a
young lawyer who abandoned a Kentucky law practice and moved to
Chicago to study astronomy. His name was Edwin Hubble.
atelv
SOS
Edwin Hubble discovered the

distances to galaxies and
devised a system for
classifying
Figure 17-3
(called
MSI
them
The Andromeda Galaxy
224) This murhy gal-
[left]
or
NGC
axy covers an area oj the sky roughly five

times as large as the full moon.
Under good
observing conditions, the galaxy's bright central bulge
can be glimpsed in the constellation
of Andromeda with the naked

tance
to the
galaxy
The white rectangle
is
2i
The
Figure 17-4 [right]
dis-
million light years.
outlines the area
Figure 17-4. (Palomnr
the
eye.
shown
in
Ob^watmy)
Cepheid variables
Andromeda Galaxy
in
Tiuo Cepheid varia-
bles are identijied in litis view of the outskirts

of the Andromeda Galaxy. Because these stars
Edwin Hubble joined the staff of the Mount Wilson Observatory in Pasadena, California, and in 1923, took an historic photograph of the socalled Andromeda nebula, one of the spiral nebulae around which so
much controversy raged. A modern photograph appears in Figure 17-3.
Careful examination of Hubble's photographic plate revealed what was
at first thought to be a nova. Reference to previous plates of that region
soon showed, however, that the object is actually a Cepheid variable.
Scrutiny of additional plates over the next several months revealed inany
additional Cepheids, two of which are identified in Figure 17-4.
As we saw in the previous chapter, Cepheid variables help astronomers determine distances. From the period-luminosity relation (recall
Figure 13-22), astronomers can determine the average absolute magnitude of a Cepheid variable. From both the absolute magnitude and the
apparent magnitude seen in the sky, a star's distance can be deduced.
Cepheid variables are intrinsically ver)' bright. They typically have
luminosities of a few thousand Suns. Hubble realized that, in order for
these luminous stars to appear as dim as they do on his photographs of
the "Andromeda nebula," they must be extreinely far away. Straightforward calculations using modern data about Cepheids demonstrated that
M31
far away. (Mount Wilson and Las
is 2t million light years away, proving it to be not a traditional nebbut an enormous stellar system located far beyond the Milky Way.
Hubble's results were presented at a meeting of the American Astronomical Society on December 30, 1924, settling the Shapley-Curtis de-
Campunas
bate once and for
appear so faint (although they are
in fact veiy
luminous), Hubble successfully demonstrated

that the
"Andromeda nebula"
Ohseivalories)
is
extremely
ula,
all.
The
universe was recognized to be far larger and
Figure
1 7-5
Hubble
classified spiral galaxies
to the
size
poijulatcd witli tar bigger objects than ainone had thus far seriously
Various spiral galaxies
winding of
the spiral
acconimu
arms and
imagined. Hubble had discovered the realm of the galaxies.

There are millions of galaxies all across the sky. visible in every unobscured direction. A typical spiral galaxy contains 100 billion stars and
measuies lOO.OOO light years in diameter. Galaxies are the biggest indi-
the
of the central bulge. Three examples

Campanas Obseivatorus)
are shown here.
vidual objects in the universe.
Hubble found
ries: spirals,
broad categoThese categories
thai galaxies can be classified into four
barred
spirals, ellipticals,
and
irregulars.
form the basis for the Hubble classification scheme.

Both M51 and M3I (Figures 17-2 and 17-:^) aie examples of spiral
galaxies. While studying spiral galaxies. Hubble noted that this class
could be further subdivided acct)rding to the size of the central bulge
and the winding oi the spiral arms. Spirals with tighth wound spiral
arms and a prominent, fat central bulge he called Sa galaxies, fhose
with moderatelv wound spiral arms and a moderate-sized ceulial bulge
are Sb galaxies. Finally, loosely wound spirals with a tiny central bulge
are Sc galaxies. Both M3I and Mol are Sb spirals. I'hese tvpes of spirals are shown in Figure 17-5.
Fortunately, the size of the central bulge and the degree oi winding
of the spiral arms go hand in hand, so we can classify even those spiral
galaxies that are viewed nearl) edge-on. For example. Ml 04 (see Figure
17-6) must be an Sa galaxy because of its huge cential bulge. In contrast.
Figure 17-6
The Sombrero Galaxy (called

4594) Became uj the large
Ml 04
or
NGC
size
Its
central bulge, this galaxy
of
is
classijiid
an Sa. If we could see it face-on, we imuld

find that the spiral arms are tightly wound
as
around a vidumnious
bulge. This galo.\y
the constellation of Virgo

6.
from our
line
of sight.
and
is
tilted
(NOAO)
tiy
is
in
o/ih
Figure 17-7
Various barred spiral

As with spiral galaxies, Hubble
galaxies
classified
barred spirals according
winding of
their spiral
arms and
lo the
the size
of the central bulge. Three examples are
shown
here.
Campanas
Obseivatories)
tlie edge-on galaxy NGC 4265, which appears in Figure 16-6, must be an
Sb galaxv because of its smaller central bulge. The tiny central bulge of
an Sc would hardly be noticeable at all in an edge-on view.
In barred spiral galaxies the spiral arms originate at the ends of a
bar running through the galaxy's nucleus rather than froin the nucleus
itself (see Figure 17-7). As with ordinary spirals, Hubble subdivided
barred spirals according to the size of their central bulge and the winding of the spiral arms. An SBa galaxy has a large central bulge and
tightly wound spiral arms. A barred spiral with a moderate central bulge
and moderately wound spiral arms is an SBb galaxy, and an SBc galaxy
has loosely wound spiral arms and a tiny central bulge.
Elliptical galaxies, named for their distinctly elliptical shapes, have no
spiral arms. Hubble subdivided elliptical galaxies according to how round
or llattened they look.

axies, the flattest are
amounts of
Of
Figure 17-8
Hubble
to
how round
galaxy
Various elliptical galaxies
classified elliptical galaxies
is
or flattened they look.
round; the
according
An EO
flattest elliptical galaxies
are E7s. Three examples are shown here.

The roundest
E7
elliptical galaxies
are called
EO
gal-
galaxies. Elliptical galaxies with intermediate
flattening receive intermediate designations (see Figure 17-8).
E2 galaxy might actually be a very flattened disk

of stars that just happens to be viewed face-on, or a cigar-shaped E7 galaxy might look spherical when viewed end-on. The Hubble scheme classifies galaxies entirely by their appearance from our Earth-bound view.
Statistical studies ha\e shown, however, that elliptical galaxies of various
actual degrees of ellipticity do exist.
Elliptical galaxies look far less dramatic than their spiral and barred
spiral cousins because they are virtually devoid of interstellar gas and
dust. In addition, there is no evidence of young stars in most elliptical
galaxies because there is no material from which stars could have recently formed. Star formation in elliptical galaxies ended a long time
course, an El or
agoElliptical galaxies exist in an enormous range of sizes and masses.

Both the biggest and the smallest galaxies in the universe are ellipticals.
Galaxies
Figure 17-9
[left]
giant elliptical galaxy
Thii huge elliptical galaxy (called
NGC
4486)
sits
or
near the center of a rich clus-
ter oj galaxies in the constellation
Although
M87
of Virgo.
some
galaxy
is
classified by
tronomers as an El),
it
has unusual properties.
this
For example,
and
it
is
a strong source of
radio waves. (Palomar Observatory)
A dwarf elliptical

galaxy
147)
is
as-
rays
is
I'his lu-urbt elliptical (called
NGC
of Cassiopeia and
galaxy (d stands for
classified
cls
dE5
dwarf). Millions of individual stars are visible

in this
remarkably clear photograph taken
through the Palomar 200-in.
Many dwarj
telescope.
galaxies are even smaller
M87
(see Figure 17-9)
huge galaxy
is
is
an example of a giant elliptical galaxy. This
located near the middle of a large cluster of galaxies in
the constellation of Virgo.
We
shall
have more to say about the fascinat-
ing case of M87 in the next chapter.

Giant ellipticals are rather rare, but dwarf elliptical galaxies are quite
common. Dwarf
ellipticals
are only a fraction the size of their normal
contain only a few million stars. Some

nearby dwarf ellipticals contain so few stars that these galaxies are transparent
you can actually see straight through the galaxy's nucleus and
out the other side. Figure 17-10 shows a typical dwarf elliptical.
The relationships that might exist between spiral, barred spiial, and
elliptical galaxies are current topics of research and debate. Hubble connected these three types of galaxies in his now-famous timing-fork diagram (see Figure 17-11). Note that according to this scheme the SO and
SBO galaxies are transition tvpes between ellipticals and the two kinds of
counterparts, and each
mav
and
spirals.
contain far Jeu'cr stars than
this one.
Hubble found
(I'alomar Obseivaton)
spirals,
barred
galaxies.
Some
that
spirals,
some
and
galaxies that
ellipticals.
exain])les include the
He
do not
fit
into his
scheme
of
called these "oddballs" irregular
Large Magellanic C^loud (l.MC^) and
Normal
spirals
Sc
l-;ilil)iicals
Figure 17-11
diagram
tion
Nubble's tuning-fork
Hubble summanied
scheme jor regidar galaxies with
tuning-fork diagram. The
"""'^
his classificathis
SO and SBO
Barred spirals
galaxies are transitional types between

ellipticals
and
spirals.
SBc
Figure 17-12
Cloud
The Large Magellanic
Al a distance of
years, this galaxy
07ily 1 50.1)1)0 light
companion
the nearest
is
our Milky Wa-t Galaxy. Note the huge

region (called the Tarantula Nebula ur
Doradus) just above
the middle uj the photo-
graph. With a diameter oj
a mass of 300,000 Suns,
known
H II
iij
II
BO
800
it
light years aiu
the largest
is
region. (Courtesy of R. J. Talbot
R. J. Dufonr. and E. B. Jensen)
Figure 17-13
The Small Magellanic Cloud

slightly jarther away from ii\
SMC only
than the LMC
The
IS
is.
Because of
asymmetrical shape, the
SMC
young blue
stars.
Talbot. R. J. Dufour.
is
sprawling,
classified as
SMC
is
(Courtesy of R.
an irregular galaxy. Note

rich in
its
and
that the
E. B. Jensen)
the Small Magellanic Cloud (SMC:). both of which can be seen with the
naked eye from southern
latitudes.
Telescopic views of the Magellanic clouds easily distinguish individual

stars (see Figures 17-12 and 17-13). Note that the SMC does not exhibit
any geometric symmetrv characteristic of spirals or ellipticals and is
therefore a true irregular. However, the LMC does apparently have a
It is thus classified as a transition type between a
barred spiral and an irregular. The Magellanic clouds are the nearest
members of a cluster of galaxies to which the Milky Way and Andromeda galaxies belong.
barlike structure.
Galaxies are grouped
Galaxies are not scattered
in clusters
in clusters.
randomh
grouped
depending on how
many galaxies they have. For example, the Milky Way Galaxy, the Andromeda Galaxy, and the Large and Small Magellanic Clouds belong to
a poor cluster familiarlv called the Local Group. The Local Group contains nearly two dozen galaxies, most ol which are dwarf ellipticals. A
map of the Local Group is shown in Hgure 17-14.
The
tion of
12.
far
across the unixerse but are
Clusters are said to be poor or rich,
is the Virgo cluster, a collecgalaxies sprawling over an area of the sky 10 x
nearest fairly rich cluster of galaxies
more than 1000
The
is about 50 million light years, too

be seen from our galaxy. Instead, the
distance to the Virgo cluster
away for Cepheid variables
to
Figure 17-14
The Local Group
Our
gal-
a pour tlmter comtsting of

about two dozen galaxies, half of which are
axy belongs
dwarf
tlian
tu
Nine galaxies
ellipticals.
thai are less
a million light years from us are
of the Milky Way. Similarly, the
satellites
Andromeda
Gala.x\ has eight satellites.
distance to the Virgo cluster has been determined bv the apparent faintness of
and B supergiant stars, by the brightness of globular clusters
surrounding some of the galaxies, and by the angular sizes of H II regions in some of the cluster's spiral galaxies. The overall diameter of the
Virgo cluster is about 7 million light vears.
The center of the Virgo cluster is dominated bv three giant elliptical
galaxies. Two of them (M84 and M86) appear in Figure 17-15, the third
(M87) in Figure 17-9. These enormous galaxies may be as large as 2 million light years in diameter. 20 times as large as an ordinary elliptical or
spiral. In other words, one giant elliptical is roughly the same size as the
entire Local (iroup.
In addition to the rich-versus-poor classification scheme, astronomers

often classify clusters of galaxies as regular or irregular, depending on
Figure 17-15
cluster
l'hi\
The center of the Virgo

Jairh
million light \ears

//i
huge
rich cluster lies
from
us.
about
elliptical galaxies
The giant
elliptical
a member of
MS7
5(1
the center oj
cluster appears in this view.
two giant
also
Only
Note
tin
(MS-f and MSfyi
(see
Figure 17-9} n
this rhisln. (.\'().\()l
the overall shape of the cluster.
The Virgo
cluster
is
called irregular be-
cause of the unsymmetrical way its galaxies are scattered about the sky.
In contrast, a regular cluster has a distinctly spherical appearance, with a
marked concentration of galaxies at its center.
The nearest example of a regular cluster is the Coma cluster, located
350 million light years from us toward the constellation of Coma Berenices. It is also a very rich cluster, with more than 1000 bright galaxies
that are easily visible in photographs (see Figure 17-16). And many thousands of dwarf ellipticals in the cluster are certainly too faint to be detected from our distance. The total membership in the Coma cluster
may therefore be as many as 10,000 galaxies.
A cluster of galaxies must be a gravitationally bound system. In other
words, there must be matter enough in the cluster to produce gravity
sufficient to prevent the galaxies from wandering away. Analyses demonstrate that the total mass needed to bind a typical rich cluster like the
Coma cluster is ten times greater than the mass of visible material that
shows up on visual photographs such as Figure 17-16. The observed
line-of-sight speeds (measured by Doppler shifts) of the cluster galaxies
are so great that more mass than has been observed is needed to keep
them boimd in orbits about the center of the cluster. This dilemma is
called the missing-mass problem. A lot of nonluminous matter in some
form must be scattered about each of the clusters or else the galaxies
would long ago have wandered away in random directions and the cluster would not exist today.
Some of this mystery has been solved recently by X-ray astronomers.
Satellite observations of rich clusters reveal that X rays are pouring from
the space between galaxies in rich clusters. These observations are evidence for substantial amounts of hot intergalactic gas at temperatures
between 10 million and 100 million degrees. Computations show that the
mass of this gas is typically as great as the combined mass of the visible
galaxies in the cluster.
Unfortunately, the discovery of hot intergalactic gas in rich clusters
solves only part of the missing-mass problem. Most astronomers agree
that a great deal of matter still remains to be discovered in rich clusters.
One popular speculation is that these clusters may contain a lot of undetected dim stars. These faint stars could be located in extended halos
Figure 17-16
cluster
sky.
is
The Coma
cluster
This
the nearest rich regular chister in the
Hundreds of galaxies are
visible in this
wide-angle view. Regular clusters are pre-
dominantly composed of
and are commonly
elliptical galaxies
sources of
rays.
315
UGC
:iiii)
211(1
2885
2 c
-
1(11)
I)l'
Figure 17-17
30
50
40
ancL' from ccnit-r

orgalaxv(kpc)
The rotation curve of a
large spiral galaxy
Thi.',
gral)li s/iouj llw
orbital speed of material in the disk of the gal-
axy
VGC 2SS5
listed in the
Many
{that
is.
the 2,S85lh galaxy
Uppsala General Calalogue).
galaxies have flat rotation cun'es, indi-
cating the presence of extended

luminosity material. (Adapted
K. Ford, and N. Tlwnnard)
lialos
from
of tow-
V.
Rubin.
surrounding individual galaxies or be scattered throughout the space

between galaxies of a cluster.
The evidence for extended halos comes from the rotation curves of
galaxies. Many galaxies have rotation curves similar to that of our Milky
Way Galaxy (recall Figure 16-12). These rotation curves remain remarkably flat to surprising distances from the galaxy's center. For example,
Figure 17-17 shows the rotation curve for the Sc spiral galaxy UGC
2885. Note that the orbital speed is fairly constant out to 60 kpc
(200.000 Iv) from the galactic center. Bevond this distance, the galaxy's
stars and H II regions are so dim and sparsely scattered that reliable
measurements are not possible. Nevertheless, we still have not detected
the true edge of this and manv similar galaxies. In the outer portions of
a galaxy, we should see a decline in orbital speed, in accordance with
Kepler's third law. Because this decline has not been observed, astronomers conclude that there must be a considerable amount of dark material extending well bevond the visible portion of a galaxy's disk. Although many proposals have been made (including cold gas, black holes,
and massive particles) to explain the nature of this outlying subluminous
matter, an extended halo of dim stars seems to be one reasonable candidate.
Galaxies sometimes
collide
and merge
loosely dispersed sea of stars scattered across intergalactic space
seems
of galactic collisions within a cluster. .411 the galaxies

in a cluster are in orbit about their common center of mass, and occasionally two galaxies might pass close enough to each other to collide.
When two galaxies collide, their stars pass b\ each other. There is so
much space between the stars that the probabilit) of two stars crashing
into each other is extremely small. However, huge clouds of interstellar
gas and dust cannot interpenetrate. Thus, when two galaxies collide,
their interstellar clouds slam into each other, producing strong shock
waves. The stars keep right on going, but the colliding interstellar clouds
are stopped in their tracks. Two colliding spiral galaxies are thus
stripped of their interstellar gas and dust. Because of the violence of the
collision between these interstellar clouds, the gas stripped from their
galaxies reaches extremely high temperatures. This ma\ be a major
source of the hot intergalactic gas often observed in rich regular clusters.
As stars pass each other during the collision of two galaxies, their
gravitational interaction can produce spectacular results. Ciravitational
forces can hurl thousands of stars out into intergalactic space along huge
arching streams. Computer simulations by .Alar Toomre at MIT and Juri
Toomre at the [oint Institute for Laboratory .Astrophysics in Colorado
dramatically illustrate these results. One of these simulations is shown in
Figure 17-18; note the striking simihiiiiv with the photograph in Figure
to
be a
likely result
17-19.
As you might expect, the most dramatic results occur when two galaxsuffer a head-on collision. Howeser, in a rich cluster
there must be many near misses. If, as we suspect, galaxies are surrounded by extended halos of dim stars, these near misses should effiies in a cluster
A looseh dispersed sea

of dim stars might come to populate the void between galaxies in a cluster in this way. Searching for these dim stars in extended halos and in
intergalactic space is one of the projects assigned the Hubble Space Teleciently strip the galaxies of their outlying stars.
scope.
Figure 17-18
simulated collision
Two
between two galaxies

ulated with
350 mass
galaxies are sim-
points, each point repre-
senting a star. As the two galaxies pass
through each other, the gravitational forces

between these masses spew out the stars along
long streamers. The drawings represent the

structure at intenmls of
200.000
\ears.
(Adapted from Alar and Jun Toomre)
vCJ:'-:"^
2U0.n(){) yr
some
600, 000 vr
800,000 yr
flung far and wide,

simply suffer a
loss of energy and momentum. Because these stars are slowed clown,
their galaxies may merge. Several dramatic examples of galaxy mergers
have recently been discovered by Fran(;ois Schweizer at the Ccrro Tololo
Inter-American Observatory in Chile (see Figure 17-20).
When two galaxies merge, the result is a bigger galaxy. If this galaxy
is located in a rich cluster, it may capture and devour additional galaxies
and grow to enormous dimensions. This phenomenon is called galactic
cannibalism. Cannibalism differs from mergers in that the dining galaxy
During a
collision
between
galaxies,
scattering material into intergalactic space.
is
bigger than
size.
Figure 17-19
with "antennae"
colliding pair of galaxies
Man\ pans
of colliding
galaxies exhibit long "antennae" of stars

ejected by the collision. This particular system,
called
tion
NGC
2623,
of Cancer.
is
located in the constella-
Many
of these colliding pairs
are strong sources of radio radiation.
(Palomar Obserimtory)
its
stars are
Other
stars
dinner, whereas merging galaxies are about the same
Figure
NGC
7-20
1316
The giant radio galaxy

huge galaxy
Tilts
the (imslellatwn of
lOU
million light
tem seems
to be
Foniax
is
laralnl in
at a dislancc oj
yean from Earth. This
an
excellent
example
galaxies merging. Note the loops
and
of
sys-
two
.szcirls
of
material churned up by the collision. This ob-
a strong source of radio radiation.

(Courtesy of Francois Schweizer)
ject
is
Many astronomers
suspect that galactic cannibalism is the reason whv

As we have seen, giant ellipticals tvpicallv
occupy the centers of rich clusters. (See the giant ellipticals M84 and
M86 near the center of the Virgo cluster in Figure 17-15, and look carefully in Figure 17-16 for the two giant elliptical galaxies
4874 and
giant ellipticals are so huge.
NGC
NGC
center of the Coma cluster.) In manv cases, smaller

galaxies are located aroimd giant ellipticals. .As these smaller galaxies
pass through the extended halo of a giant elliptical, they slow down and
4889
at the
are eventually devoured by the larger galaxy.
The Hubble law is a simple

relationship between the
redshifts of galaxies and
their distances
Whenever an astronomer fmds an
object in the sky that can be seen or

photographed, the natmal inclination is to attach a spectrograph to the
tclestope. As long ago as 1914, \'. M. Slipher, working at the Lowell
Observatory in Arizona, began taking spectra of the "spiral nebidae." He
was surprised to discover that
oi liie
spiral nebulae he studied had
their spectral lines shifted towaid red. indicating substantial recessional
velocities. This marked dominance of redshifts was presented b\ Curtis
at the 1920 Shapley-Curiis debate as evidence that these spiial nebidae
i
could not be ordinary nebul.ie
Dining the
192()s.
many
in
.'i
oui Milk\ VVav (ialaxv.
Edwin Hubble and
.Milton
Humason photographed
on Mount Wilson.
shown in Figure 17-21. As indicated in this illustration, there seemed to be a direct
correlation between (he distance to a galaxy and the size of its redshift.
In other words, nearby galaxies are moving away from us slowlv and
more distant galaxies are rushing away much more rapidU. This universal recessional movement is sometimes called Hubble flow.
the spectra of
galaxies with the
Five representative ciliptiial galaxies
l()()-in.
and
telescope
iheir spectra are
Figure 17-21
Five galaxies
and
their
Cluster
nebula
The plwtugraplu of theie jive elliptical galaxies all have the same magnification.
The spectrum of each galaxy is the hazy band
spectra
in
between the comparison spectra. In all five

cases, the so-called
and
lines of
calcium
are seen. The recessional velocity (calculated
from
is
the
Doppler
shifts
of the
H and K
lines)
given below each spectrum. Note that the
more distant a galaxy
is,
the greater
is
its

Campanas Obsemalories)
redshift.
III
:i
III
Hvdra
III
ill
III
III
61,000 km/sec
Using various techniques, Hubble estimated the distances to a number

of galaxies. From the formula for the Doppler effect, he calculated the
speed with which each galaxy is receding from us. When he plotted the
data on a graph of distance versus speed, he found that the points lie
near a straight line. Figure 17-22 is a version of Hubble's graph based
on modern
is
data.
This relationship between the distances to galaxies and their redshifts

one of the most important astronomical discoveries of the twentieth
we shall see in Chapter 19, this relationship tells us that

are living in an expanding universe. Hubble published his discovery,
now known as the Hubble law, in 1929.
The Hubble law can be stated as a formula:
century. As
we
H.,r
where
i'
is
commonly
is the distance, and H,) is a constant

Hubble constant. This formula is equivalent to the
displayed in Figure 17-22. The Hubble constant Hq tells us
the recessional velocity.

called the
straight line
Figure 17-22
The Hubble law
lames and recessional
velocities
of
The
ral galaxies are plutled on this graph.

straight line
called the
spi-
The
the "best fit" for the data. This
linear relationship beteween distance
dis-
60 Sc
Hubble
laic.
and speed
(Adapted from
Sandage and Tammanii)
100
Distance (10"
the slope of the Une in Figure 17-22.
graph we find
Hn =
From
ly)
the data plotted on thai
that
kni/sec/Ml\
where iMly stands for 10" light years. (Say: Fifteen kilometers per second per million light years.") In other words, for each million light vears
to a galaxv, the galaxy's speed away from us increases bv 15 km/sec.
Thus, for example, a galaxy 100 million light vears from Earth should
be rushing away from us at 1500 km/sec.
Incidentally, many astronomers prefer to speak of megaparsecs (Mpc)
rather than millions of light years. Using that unit.
f/ii
= 50 km/sec/Mpc
(Say: "fifty kilometers
per second per megaparsec")
The exact value of the Hubble constant is a

among astronomers today. The data plotted in
the values of
topic of heated debate
Figure 17-22, as well as
from the recent meticulous work of

.Allan Sandage and Gustav fammaiin. who did most of their observations with the 20()-in. Paloniar telescope. Howc\er, other prominent astronomers feel stronglv that the Sandage- Fammann \alue for // is too
low. They argue that the true value is about 'M) km sec/Ml\ (= 100 km/
sec/Mpc). Many astronomers simplv use a number between the two ex//,,
given above, are
all
tremes, such as 75 km/secVMpc.
The Hubble
is one of the most important numbers in all

As we shall see in Chapter 19, it expresses the rate at
which the universe is expanding and tells us the age of the universe.
constant
physical science.
Naturallv, astronomers are verv interested in an accurate determination

of //.
Determining the value of //,, is an extremelv difficult and trickv busiFo measure H. vou would have to observe main galaxies and determine the recessional velocity of each from the redshift of its spectral
lines, ôu would also have to determine the distance to each galax\. Finally, you would plot all the data on a graph like Figure
7-22. I'he
ness.
fits through the data would give you Hq. Unyou would encounter numerous pitfalls along the way.
There are pitfalls with the redshifts. Measuring the recessional velocity of a galaxy sounds easy enough: just take a spectrum, see how far the
spectral lines are shifted, then use the Doppler formula to get the speed.
Galaxies come in clusters, however, and all the galaxies in a cluster are
in orbit about their common center of mass. Because of these orbital
motions, some galaxies are coming toward you, others moving away.
Thus, tacked onto the true recessional velocity that you are trying to determine is an additional velocity toward or away from you because of the
galaxy's motion around its cluster.
One way of coping with this problem is to turn your attention to extremely distant galaxies. For very remote galaxies, the recessional velocity
caused by the expansion of the imiverse far outweighs the orbital motions in clusters. Redshift measurements of remote galaxies will thus
more accurately reflect the Hubble flow. Astronomers trying to determine Ha therefore commonly restrict their observations to galaxies farther than 100 million light years from Earth. Unfortunately, a new pitfall arises: you are no longer exactly sure about the distances to these
remote galaxies.
The distances to nearby galaxies are determined by fairly reliable
methods. For example, Cepheid variables can be seen out to 20 million
light years from Earth. The distances to galaxies in this nearby volume
of space can therefore be determined from the period-luminosity law.
Beyond 20 million light years, however, even the brightest Cepheid variables fade from view. Astronomers then turn to more luminous stars
the brightest supergiants. These stars can be seen out to distances of up
to 80 million light years. By comparing the known absolute magnitude
of supergiants with the apparent magnitude of a supergiant that you
observe in a galaxy, you can determine the distance to that galaxy.
Beyond 80 million light years, individual stars are no longer discernible. Astronomers must then turn to entire clusters and nebulae whose
absolute magnitudes are known. The brightest globular clusters can be
seen out to 130 million light years, and the brightest H II regions can be
detected out to 300 million light years from Earth. From the observed
faintness of these clusters and nebulae in a remote galaxy, the distance
to that galaxy can be estiinated.
Finally, to get beyond 300 million light years, astronomers must wait
for supernova explosions (see Figure 17-23). These brilliant outbursts
can be seen out to distances of 8 billion light years from Earth.
These various objects Cepheid variables and the most luminous supergiants, globular clusters, H II regions, and supernovae
are commonly called standard candles. As you might suspect, astronomers go to
great lengths to check the reliabilit\ of their standard candles. After all,
a tiny mistake in the absolute magnitude of a supergiant star or globular
cluster can lead to an error of many millions of light years in calculating
the distance to a remote galaxy.
The major obstacle in determining the Hubble constant is that the
farther we look into space, the fewer standard candles we have. The distance to a nearby galaxy can be cross-checked in many ways. For example, the distance computed from the period luminosity relation can be
compared to the distance determined from the magnitudes of the most
luminous supergiants. These results can be further compared with the
magnitudes of the galaxy's globular clusters and the angular sizes of its
H II regions. After all this work, the results from these various methods
slope of the line that best

fortimately,
Figure 17-23
In 1961
supernova in
NGC
4303
a supernova erupted in the spiral
4303
M61l. which u
a member of the Virgo cluster. Supemovae can
be seen in exiremeh remote galaxies and are
galfLxy .\'GC
(also called
important "standard candles" used to deter-
mine
the distances to these
faraway
galaxies.
(Lick Obsen'atory)
can be averaged to obtain a distance to the galaxv in which astronomers

can have confidence.
As we turn to more distant galaxies, however, w-e can see fewer and
fewer standard candles. For example, the Coma cluster in Figure 17-16
are so far away that it is impossible to see individual stars in any of its
galaxies. Since there are fewer standard candles, fewer cross-checks are
and the distance to these remote galaxies becomes less certain.

L'nfortunatelv, remote galaxies are precisely the ones whose distances we
possible
must determine in order to find the value of the Hubble constant. The
uncertaintv in determining their distances is the cause of our uncertainty
in knowing the exact \aluc of f/,,.
Astronomers hope that the Hubble Space Telescope will solve many
of these problems. This superb telescope, named after the father of
extragalactic astronomy, will produce views of galaxies far sharper than
anything that is possible from the ground and give us important information ahoui the fundamental nature of our universe.
Summary
Galaxies can be grouped into four major categories in the Hubble

cation: ellipticals, spirals, barred spirals,
Spiral galaxies like

ral galaxies
appear
and
classifi-
irregulars.
our own are sites of active star formation. Barred spito be temporary stages in the normal development of
spiral galaxies.
are virtually devoid of interstellar gas and dust.

occurring in these galaxies.
Elliptical galaxies
formation
is
No
star
Galax
A rich cluster contains hundreds or

even thousands of galaxies whereas a poor cluster may contain only a few
dozen galaxies.
Galaxies are grouped into clusters.
Our
galaxy
is
member
of a poor irregular cluster called the Local
Group.
elliptical galaxies
Giant
are often found near the centers of rich clusters.
matter observed in a cluster of galaxies is not large enough

to account for the gravity needed to hold the cluster together. A large
amount of unobserved mass must therefore be present between the galaxies ("the missing-mass problem").
The luminous
When
two galaxies
media
collide violently, stripping the gas
collide, their stars pass
each other, but their interstellar
and dust from the
galaxies
and
heating the intergalactic gas.

Gravitational effects during a galactic collision can throw stars out of the
galaxies into intergalactic space.
Galactic mergers
grow
steadily
may
occur.
large galaxy in a rich cluster
may tend
by galactic cannibalism, perhaps producing a giant
to
elliptical
galaxy.
a simple linear relationship between the distance from the Earth
and the redshift of that galaxy (which is a measure of the
speed with which it is receding from us). This relationship is the Hubble
There
is
to a galaxy
law,
I'
H(,r.
Because of
difficulties in
Hubble constant
f/,, is
Standard candles such
measuring distances
not
as
giants, globular clusters,
known
to galaxies, the value of the
with certainty.
Cepheid variables and the most luminous superII regions, and supernovae are used in esti-
mating intergalactic distances.
1 Describe the Hubble classification scheme. Illustrate your answer with a

tuning-fork diagram.
Review questions
2 What types of galaxies are most likely to have new stars forming? Describe
the observational evidence that supports your answer.
3 Are there any galaxies besides our
eye? If so, which one(s)?
own
that can be seen with the
naked
4 How is it possible that galaxies in our Local Group remain to be discovered? In what part of the sky would these galaxies be located? What sorts of
observations might reveal these galaxies?
5 Explain why the "missing mass"
hydrogen.
Why do some
galaxies in the Local
lines? Is this a violation
Whv do
What
in galaxv clusters
Group
could not be neutral
exhibit blueshifted spectral
of the Hubble law? Explain.
you suppose there are discordant determinations of Ho?
kinds of stars would you expect to populate space between galaxies
in a cluster?
Advanced questions
9 Suppose that you take a spectrum of a distant galaxy and find that its
redshift corresponds to a speed of 22.000 km/sec. How far away is the galaxy?
"10
Pavo is located
from us?
cluster of galaxies in the southern constellation
Mpc from
Earth.
How
fast
is
this cluster receciing
at
100
11 How might you determine what part of a galaxy's redshift results from
the galaxy's orbital motion about the center of mass of its cluster?
12 Discuss the advantages and disadvantages of using the various distance

indicators to obtain extragalactic distances.
13 Discuss the idea that the various Hubble types of galaxies represent
some
sort of evolutionary sequence.
For further reading
Btrendzen, R., et al. Man Discovers the Galaxies. Neale Watson, 1976.
DeVaucouleurs, G. "The Distance Scale of the Universe." Sky
Telescope.
&
Dec. 1983,
p.
511.
"The Hidden Mass in Galaxies." Mercury, May/June 1982, p.

P., and Tucker, W. "Rich Clusters of Galaxies." Scientific
Field, G.
74.
Gorenstein,
American, Nov. 1978.
Hausman, M.
"Galactic Cannibalism." A/^m;n', Nov. /Dec. 1979, p. 119.
Hodge, P. Galaxies, Harvard Universitv Press, 1986.

Kaufmann, W. Galaxies and Quasars, VV. H. Freeman and Company, 1979.
Rubin, V. "Dark Matter in Spiral Galaxies." Scientific American, June
1983.
Smith, R. The Expanding Universe: Astronomy's Great Debate. Cambridge
University Press, 1982.
"Unscrambling the Local Superckister." Sky
Tullv, R.
1982, p. 550.
iff
Telescope.
June
18
The inner edge of an accretion disk around

a black hole
shmos
disk
is
llie
Tim
compuler-generaled piiiurv
jlow pattern vf gas
mrrounding a
an
in
black hole.
accretion
The black
hole
located at the center of this cross-sectional
diagram, where the broad-angU inner

the accretion
dhk
pressure, ranging
point. Color ili\jihi\s
from red ]oi
higli
irlge\
iif
:;(
pu^sim
across the spectrum to bhie for low piessuii
Many
astronomers explain the etiormous en-
ergy output of a quasar b\ theorizing that gas

is
falling onto a .supermassive (10''
hole.
The flow pattern (indicated
arrows) shows how infalling gas

into two oppositeh directed jets.
Mi
black
by the white
is
channeled
"centrifugal
barrier" created by the rapid rotation of the
inner edge of the accretion disk prevents thr

gas from falling directh into the black hole.
(Courtesy of John F.
Hawley and
Lan-y L. Smarr)
Aslronomers have discovered a large number of
objecls in the aky
whose
extreme redshifts indicate that they are at great distances from the Earth,
according to the Hubble law. To be obseiiiable at all at such enormous dis-
and active galaxies must clearly be

we see evidence indicating that a typical
quasar emits the energy output of 100 galaxies from a volume roughly the
size of the solar system. We explore a variety of active galaxies and find that
tances, these quasi-stellar objects (quasars)
extremely luminous. In this chapter,
the centers of these objects also are pou'erful sources of energy. Finally, ive
learn about recent theories that .supermassive black holes are involved in the
intense energy production of these luminous objects.
astronomy in tlie late 1940s ranks among the

accomplishments of the twentieth century. Prior
to that time, everything known about the distant universe had to be
gleaned from visual observations. Radio telescopes provided a view of
the universe in a wavelength range far beyond that of visible light. Many
surprising discoveries emerged from this new ability to examine the pre-
The development
most important
of radio
scientific
viously invisible imiverse.
Figure 18-1
3C 405)
Cygnus
(also called
Thii sliaiii;f-lijuknig
gulaw uns
discovered at the location of the radio source
Cygnus A. This galaxy has a

sponding
to a recessional
the speed of light.
redshijt
cone-
speed of 6 percent
According
to the
oj
Hubble
law, this redshift corresponds to a distance of
about
Cyg A
I billion light years

is
from Earth. Becaitsr
one of the brightest radio sources
the sky, the energy output
of
this
remote galaxs
must be enormous. (Palomar Obseii'atory)
The
radio telescope was built in 1937 bv an amateur astronomer.

in his backyard in Illinois. Bv 1944. Reber had detected
strong radio emission from Sagittarius. Cassiopeia, and Cvgnus. Two of
these sources (nicknamed Sgr A and Cas A) happen to be in our own
galaxy as the galactic nucleus and a supernova remnant. In 1951, Walter
first
Grote Reber,
Baade and Rudolph Minkowski used the 200-in.
omar Mountain
optical telescope
on
to discover a strange-looking galaxv at that position
the third source called Cygnus A (Cvg A). Figure 18-1

of the optical counterpart of Cyg A.
is
Pal-
of
photograph
peculiar galax\ associated with Cvgnus A is verv dim. .NevertheBaade and Minkowski managed to photograph its spectrum, which
shows a number of bright spectral lines, all shifted bv 5.7 percent toward
the red end of the spectrum. Astronomers were surprised bv this large
redshift because, according to the Hubble law. the distance to Cvgmis A
The
less.
is
roughly
billion light years!
The enormous
distance to Cygnus .A means that it has to be one of

the bightest radio sources in the sky. Although Cyg .\ is barelv visible
through the 2()0-in. telescope at Palomar. its radio waves can be picked
up by amateiu' astronomers with backvard ecjuipment. The energv output in radio wa\es from Cvg .\ must therefore be enormous. Indeed.
Cyg A shines with a radio luminosiis 10' limes as bright as that of an
ordinary galaxv such as M31 in .Andromeda. The object corresponding
to Cyg A nuist obviously be something c|uite extraorditiarv.
Quasars look like stars but

have huge redshifts
Duiing the late 195()s and early 196()s. radio astronomers were busy
making long lists ot all the radio sources ihcv were finding across the
sky. One of the most famous lists, published in 1959, is called the Third
Cambndge Catalogue (the first two catalogues produced by the British
Quasaii and uclive galaxies
The quasar 3C 48
Figure 18-2
Foe
sei'-
eral \ears. astronomers believed erroneouily

object
that
litis
that
happens
redshift
of
according
is
to
simph
a nearby peculiar star
emit radio waves. Actually, the
this starlike object
to the
Hubble law,
is
it
so gteat thai,
must be
roughh 5 billion light years away.

(Palomar Obsemulon)
team were filled with inaccuracies). It lists 471 radio sources. E\en today,
astronomers often refer to these sources by their "SC numbers." With
the discovery of the extraordinary luminosity of Cyg A (also called 3C
405, because it is the 405th source on the Cambridge list), astronomers
became eager to learn whether any other sources in the 3C catalog had
similarly extraordinary properties.
One interesting case is 3C 48. In 1960. Allan Sandage used the 200in. telescope to discover a "star" at the location of this radio source (see
Figure 18-2). Since ordinary stars are not strong sources of radio emis-
unusual. Its spectrum does in fact show a

of bright spectral lines that no one could identify. Although 3C 48
was clearly an oddball, many astronomers thought it was just another
strange star in our galaxy.
Another such "star," called 3C 273, was discovered in 1962. It was
found to have a luminous "jet" protruding from one side, as shown in
Figure 18-3. And. as was the case with 3C 48, this "star" contains a series
of bright spectral lines that no one could identify.
.Astronomers had difficulty identifying the emission lines in the spectra of 3C 48 and 3C 273 because the\ assumed that these starlike objects
were just peculiar stars nearby in our galaxy. .\ breakthrough finally
came in 1963, when Maarten Schmidt at the California Institute of
Technology found that four of the brightest spectral lines of 3C 273 are
positioned relative to one other just like four familiar spectral lines of
hydrogen. However, the emission lines of 3C 273 are at much longer
wavelengths than the usual positions of the Balmer lines. In other words,
3C 48 must be something
sion,
series
3C 273 has
a substantial redshift.
Doppler shifts, because a

our galaxy cannot be moving extremely fast relative to the Sun,
would soon escape from the galaxy. Schmidt therefore conjectured
Stellar spectra exhibit comparatively small
star in
or
it
The quasar 3C 273
Figure 18-3
327
This
greatly enlarged view shows the starlike object
associated with the radio source

the
luminous
jet to
one side of
3C
273. Note
this "star."
1963. astronomers discovered that the

of
this "star"
cording
to the
light years
so great that
Hubble law.
is
its
By
redshifl
distance, ac-
nearly 3 billion
from Earth. (XO.AO)
perhaps 3C 273 is not after all a nearby star. Pursuing this hunch,
he promptly identified all four spectral lines as h\clrogen lines that have
suffered an enormous redshift corresponding to a speed of 15 percent
of the speed of light. According to the Hubble law, this huge redshift
implies the incredible distance to 3C 273 of roughly 3 billion light years.
Because of their starlike appearance, 3C 48 and 3C 273 were dubbed
quasi-stellar radio sources. This term was soon shortened to quasars.
Instead of using photography to record a spectrum, many observatories use an electronic de\ ice at the focus of the spectrograph. The output of the machine is a graph of intensity versus wavelength, on which
emission lines appear as peaks. The graph showing the spectrum of 3C
273 in Figure 18-4 was obtained in this way. Note the four hydrogen
that
lines.
Figure 18-4
Four
The spectrum of 3C 273
bright nniy^Kin linrs caused
dominate the spectrum of

indicate
shifted
how far
from
their
lr\-
hydrogen
3C 273. The arrows
these spectral lines are red-
muat
wavelengths.
4801)
5600
Wavclcnglh
A)
Qiiawn ami
Figure 18-5
The quasar
OH
quasar has one of the largest
471
iirlivr galaxie.
This
redshifts ever dis-
covered. This redshift corresponds to a speed

slightly greater
light.
than 90 percent of the speed
According
must he 18
to the
Hubble law.
Inllion light years
OH
47
away.
(Palomar Obseniatory)
The spectral lines of 3C 273 are brighter than the intensity of the
background radiation at other wavelengths. The background is called the
continuum, and the bright lines are emission lines caused by excited
atoms that are emitting radiation at specific wavelengths. Spectra of ordinary galaxies are dominated by dark absorption lines (recall Figure 1721). Most quasars and many peculiar galaxies exhibit strong emission
lines in their spectra, a sign that something unusual is going on.
Inspired by Schmidt's success, astronomers next identified the spectral
of 3C 48 as having suffered a redshift corresponding to a velocity
of nearly one-third the speed of light. According to the Hubble law, 3C
lines
48 must therefore be twice as

billion light years from Earth.
far
away
as
3C
273, or approximately 6
that the Hubble conprevious chapter, some

astronomers believe that this value is too low and would prefer to use 30
km/sec/Mly for //. Doubling the Hubble constant would then halve these
quasar distances.
Hundreds of quasars have been discovered since the pioneering days
of the early 1960s. All quasars look like stars, and all have enormous
redshifts. There are even a few quasars, such as
47 1 (shown in Figure 18-5), with redshifts corresponding to speeds greater than 90 percent of the speed of light. From the Hubble law, it follows that the distance to
471 is 18 billion light years. Its light has taken 18 billion
years to reach us. When we look at
471, we are seeing an object as
it existed when the universe was very young.
Incidentally, these distances to cjuasars
stant (Hq)
is
15 km/sec/Mly.
As mentioned
assume
in the
OH
OH
OH
Qha<hs anil nrlive galaxies
quasar emits a huge

of energy from a
1 lie luminosities of
quasars can liest be appreciated in comparison with
the luminosities of galaxies. Galaxies are big and bright. A typical large
galaxy like our own or M3I in Andromeda contains several hundred billion stars and shines with a luminosity of 10 billion Suns. The most gigantic and most luminous galaxies (such as the giant elliptical M87 in
amount
very small volume
Figuie 17-9) are only ten

10(1 billion Suns. Beyond
brigiitcst galaxies are too
too dim to be detected at
Although
it
is
quasars are quite
nous
times brighter, shining with the brilliance of

10 billion light vears from Earth, even the
faint to be detected. Most ordinary galaxies are
half that distance.
difficult to find high-redshift galaxies, high-redshift
common. Quasars must
far more so than galaxies.
In fact,
therefore be incredibly lumi-
some quasars are 100 times
brighter than the brightest galaxies we have ever seen.

In the mid-1960s, several astronomers discovered that
some of
the
newly identified quasars had been inadvertently photographed in the

past. For example, 3C 273 was found on numerous photographs, including one taken in 1887. By carefully examining the images of quasars on
these old photographs, astronomers found that quasars fluctuate in
brightness, occasionally flaring up. For example, see the data from old
photographs of 3C 279 plotted in Figure 18-6. Note particularly the
prominent outbursts that occuried around 1937 and 1943. During these
outbursts, the luininosity of 3C 279 increased by a factor of at least 25.
Because of the enormous distance to this quasar, it must have been shining with a brilliance 10,000 times as great as that of the entire Andromat the peak of these outbursts.
Fluctuations in the brightness of quasars allow astronomers to place
strict limits on the maximum sizes of quasars, because an object cannot
vary in brightness faster than the time it takes light to travel across that
eda galaxy
For example, an object that is I ly in diameter cannot vary in

brightness with a period of less than I year.
The brightness of many quasars may vary over periods of only a few
weeks or months and some quasars even fluctuate from night to night.
Recent X-ray data from the Einstein Observators re\eal large \ariations
object.
three hours.
in as little as
This rapid flickering means that quasars are small. The energy-emitting region of a typical quasar
the "powerhouse" that blazes with the
luminosity of 100 galaxies
is less than
light-dav in diameter. If qua-
-HI
-
12
\3 -
14 -
l:i
The brightness of 3C 279
This g)aj)h shows variations in magnitude of

the
quasar
1937. The
3C
279. Note
(lata
the large outburst
were oblaineil jrom careful
examination of old pholographii plates in the

files of the
Harraril College Obsen'atory.
(Adapted jrom
I..
Eacluis
and W.
l.iller)
.V
Figure 18-6
^*
.
11)
17 -
IH
'V
1^
',
>
#
*
'
,*
Qiiasan and active galaxies
indeed at the huge distances indicated by their redshifts, they

must be producing the luminosity of 100 galaxies in a volume not much
farther across than the diameter of our solar system.
sars are
Active galaxies bridge the
gap in energy output

between ordinary galaxies
and quasars
In the 1960s, the gap in energy output between ordinary galaxies and
quasars seemed so huge that some astronomers questioned the Hubble
law instead of accepting the existence of such highly kmiinous objects. In
recent years, however, astronomers have discovered various kinds of pe-
whose luminosities fall between those of ordinary galaxies

and quasars. Some of these strange galaxies have unusually bright, star-
culiar galaxies
Others have strong emission lines in their spectra. Still others

And some have jets or beams of radiation emanating
from their cores. Most of these objects are more luminous than ordinary
like nuclei.
are highly variable.
galaxies. All of these types are called active galaxies.
The
first
active galaxies
were discovered
by Carl Seyfert
in
1943, during a survey of
Mount Wilson Observatory.
Called
Seyfert galaxies, these luminous objects have bright, starlike nuclei and
strong emission lines in their spectra. For example, NGC 4151 (see Figspiral galaxies,
at the
ure 18-7) has an extremely rich emission spectrum with 28 percent of

the galaxy's light concentrated in emission lines. These emission lines
include Fe X and Fe XIV (iron atoms with 9 or 13 electrons, respectively, stripped away), indicating that NGC 4151 contains some extremely
hot gas. Seyfert galaxies also
magnitude of
NGC
commonly
exhibit a variability in brightness
4151 changes every few months).

.'Approximately 10 percent of the most luminous galaxies in the sky
are Seyfert galaxies. Some of the brightest Seyfert galaxies shine as
brightly as faint quasars, leading many astronomers to suspect that re(the
The
Figure 18-7
4151
This
galaxies.
and
is
Seyfert galaxy
Because of
its
bright, starlike nucleus
emission-line spectrum,
be mistaken for a quasar if
away. According
shift
of
this
NGC
one of the best-studied Seyfert
to the
NGC
it
4151 might
were very far
Hubble law,
galaxy gives
it
the red-
a distance of 70
million light years from Earth.

iliiiiuiis
(iii/l
ijclive
galaxies
331
The exploding galaxy NGC

3C 84) I his Sa/nl ffo/
Peneus cluster, is a strong
rays and radio radiation. Accord-
Figure 18-8
1275
(also called
axy, located in the
source of
ing
to the
Hubble law,
the galaxy's redshift
gives a distance of nearly
400
million light
years from Earth. Note the streamers of gas.

Spectroscopic obsemations confirm sigtiificant
mass ejection from the galaxy's center.
(NOAOl
mote Seyfert galaxies could easily be interpreted as quasars. Many more

Seyfert galaxies have been discovered in recent years, largely through
the efforts of Wallace Sargent at Caltech.
Many
Seyfert galaxies exhibit vestiges of violent, explosive phenomena

Examine the twisted, tormented galaxv
1275 in Figure 18-8. Filaments of gas from tens to thousancls of light years long
in their nuclei.
NGC
protrude from the galaxy in all directions. Spectroscopic studies indicate

is being blasted away from the galaxy's nucleus at 3000 km/
sec. The nucleus of this galaxy is a strong source of radio waves and X
rays (see Figure 18-9).
Peculiar galaxies other than Seyferts also seem to be ejecting enormous amounts of matter and energy. Note the extraordinary dust lane
across the face of the disrupted galaxy NGC 5128 in Figure 18-10. Recent X-ray pictures reveal an X-ray jet (see Figure 18-11) sticking out of
that this gas
Figure 18-9
1275
An X-ray image
of
NGC
This picture from the liinslrin (Jhserr
atory shows the X-ray appearance of the

Seyfert galaxy
eviLssion
NGC
galaxy's nucleus.
ler
1275. Most
oj the
Xuiy
comes from a point source at /he
(Harvard-Smithsonian Cni
for Astrophysics)
Qjimari and active galaxies
Figure 18-10
NGC
5128
The peculiar galaxy
[left]
(also called
exiraordinan galaxy
tion
is
Centaurus A)
This
located in the constella-
of Centaurus, roughly IB million light
years
from Earth. Note
the
broad dust lane
across the face of the galaxy. Vast quantities
of radio radiation pour from extended

regions of the sky on either side of the dust
lane.
(NOAO)
NGC
5128
stein Obseniatory
shows that
bright X-ray nucleus.
from
An
the galaxy's nucleus
perpendicular
in
An
X-ray image of
This X-ray picture from the
to the
NGC
X-ray
5128
Emhas a
jet protrudes
along a direction
galaxy's dust lane seen
Figure 18-10. Diffuse X-ray emission
comes from
regio7is
surrounding the
galaxy's center. (Harvard-Smitlisonian
Center for Astrophysics)
Figure 18-12
radio image of Cygnus
This color-coded radio picture was produced at
Large Airay. Most of the radio emission from Cygnus A comes from the radio lobes
located on either side of the peculiar galaxy
the Very
seen in Figure 18-1. These two radio lobes
cal
an
160,000 light years from the optigalaxy, and each contains a brilliant,
each about
condensed region of radio emission. The

rectangle indicates the area covered by
Figure 18-1.
(NRAO)
the galaxy's nucleus.

18-11.
The
jet
is
perpendicular to the dust lane in Figure
NGC 5128 is one of the brightest sources of radio waves in the sky. It
was one of the first sources discovered when radio telescopes were built
in Australia. As a radio source, it is called Centaurus A, because of its
location among the constellations. In part, the brightness of Centaurus A
at radio wavelengths come from its proximity to Earth, only 13 million
light years away. Radio waves pour from two large regions on either side
of the galaxy's dust lane. These regions cover many square degrees of
the sky, corresponding to a volume spanning 2 million light years.
By 1970, radio astronomers realized that Centaurus A ancl Cygnus A
(discussed at the beginning of this chapter; recall Figure 18-1) belong to
a class of objects called double radio sources in which radio emission is
concentrated in two "radio clouds" or radio lobes. In many cases, a peculiar galaxy resembling a giant elliptical galaxy is located between the two
radio lobes (see Figures 18-1 and 18-12). Further evidence for a high
Qtiunin ami
ticttve
galaxies
of activity in these peculiar galaxies comes from their spectra,

which often exhibit many emission lines and resembles an excited Seyfert
level
galaxy.
Some douljle radio sonrces are enormous. The scale drawing in Figure 18-13 shows that the largest (such as 3('. 236) are as big as an entire
cluster of galaxies.
Every double radio source seems to have some sort of central "engine" that squirts particles (probably electrons) and magnetic field outward along two oppositely directed jets at speeds very near the speed of
light. .-Xfter traveling manv thousands or e\en millions of light \eais, liiis
ejected material slows
down, and interaciion between the electrons and
the magnetic field produces the radio
iadi,iii()n that
we
detect.
specific
type of radio emission called synchrotron radiation occurs whenever

electrons move in a spiral within a magnetic field. The radio waves that
come from the lobes of a double radio source have all the characteristics
of synchrotron radiation.
The idea that a double radio source inxolves powerful jets of relativistic (in the sense of "traveling near the speed of light ") particles is supported by the existence of head-tail sources, named because each such
source appears to have a region (its "head") of concentrated radio emission followed by a weaker tail of emission. \ good example is the active
elliptical galaxv NGC 1265 in the Perseus cluster of galaxies. NG(^ 1265
is known to be moving at a high speed (2500 km/sec) relative to the cluster as a whole. Figure 18-14 is a radio map of NGC 1265. Note that the
radio emission has a distinctly windswept appearance. Just as smoke
pouring from a steam locomotive trails a rapidly moving train, particles
ejected along two jets from this galaxy are deflected by the galaxy's passage through the sparse intergalactic medium.
At radio wavelengths, the double radio sources are among the brightest objects in the universe. Using some basic physics, astronomer Geoffrey Burbidge has demonstrated that the energy contained in the radio
lobes of a typical source roughly equals the energy released by 10 billion
supernova explosions.
\ final class of active galaxies is the N galaxies, so-named because
they have bright nuclei. Extreme examples are the BL Lacertae objects,
-oo.cm
Giant double radio sources
Figure 18-13
Some double
scale
radio sources are enormous. This
drawing shows our galaxy and Cygtius
A, along with several very large sources,
3C 236, whose
to thf size
(Adapted
of cm entire cluster
fmm
like
dimensions are comparable
R.
and J. H. Oort)
(..
Strum.
iif
(..
Milkv W.iv
i;.il,i\v
galaxies.
K. Milr\.
Wulll. (Iir
lii>lil
yens
Qiiamrs and active galaxies
Figure 18-14
1265
The
NGC
XGC 7265
The head-tail source
active elliptical galaxy
would probably be an ordinary double radio

source except that the galaxy
is
moving
at a
high speed through the intergalactic medium.
Because of this motion, the two jets trail the

galaxy, giving this radio source a distmcth
windswept appearance.
Figure 18-15
BL
(MRAO)
Lacertae
This superb
photograph shows "fuzz" aroung BL Lacertae

BL Lacertae objects appear to be giant elliptical galaxies with bright, starlike nuclei,
much
as Seyfert galaxies are spiral galaxies with
quasarlike nuclei.
much
ies.
less
BL
Lacertae objects contain
gas and dust than do Seyfert galax-
(Courtesy of T. D.
Kinman;
NOAO)
Quasars and at live galaxies
named
after their proiotvpe. Bl, Lac. in the constellation of Lacerla (the
Lizard).
BL Latertae (see Figure 18-15) was first discovered in 1929. when it

was mistaken for a variable star: its brightness varies by a factor of 5 in
only a few months. BL Lac's most intriguing characteristic is a totallv
featureless spectrum that exhibits neither absorption nor emission lines.
Careful examination of BL Lac revealed some "fuzz" around its
bright, starlike core. In the earh' 1970s, Joseph .Miller at Lick Observatory blocked out the light from the bright center of BL Lac and managed to obtain a spectrum of the "fuzz." The spectrum contains many
spectral lines and stronglv resembles the spectrum of an elliptical galaxv
(see Figure 18-16). In other words, a BL Lacertae object is an elliptical
1
galaxy with a bright starlike center, much as a Sevfert galaxv is a spiral

galaxy with a quasarlike center.
These discoveries ha\e prompted main astronomers to suspect that
Wavelength
The spech-um
Lacertae
rouiulnig
BL
spectrum of
the Local
tween
Lac
M32,
is
oj the "fuzz" \u>-
shown, along with the
a small
Group. The
the.se
assuming
The spectrum of BL
Figure 18-16
elliptical
galax\ in
slight differences be-
two spectra can he explained by
that
BL
Imc
is
a giant elliptical gal-
The spectrum of BL Lac is redshifted

an amount conespondmg to a distance of
axy.
I billion light years
from J.
S.
S. Miller.
from Earth. (Adapted
H. B. French, and
A. Hawley)
Supermassive black holes

may power quasars and
active galaxies
b\
quasars are the superluminous centers of very distant, verv active galaxies. Painstaking observations have in fact revealed faint galaxylike "fuzz"
around se\eral quasars. Of course, quasars are extremelv far awav and
are thus difficult to observe. Astronomers therefore look forward to the
launch of the Hubble Space Telescope, which should be able to resolve
features associated with quasars that are undetectable with Earth-based
instrimients. Some astronoiners suspect that the faint "fuzz" seen aroimd
some quasars might actualh turn out to be spiral arms. ihereb\ pro\ing
that quasars are embedded in galaxies. Meanwhile, the best clues about
the "engine" that powers a quasar come from studies of nearbv active
galaxies such as the giant elliptical galaxv MS7 in the \'irgo cluster (recall Figures 17-9 and 17-15), a scant 50 million light \ears from Earth.
How do quasars and acti\e galaxies produce such enormous amounts of

energv from such small \()lumes? .As long ago as 1968, the British astronomer Donald Lynden-Bell suggested that an extremeK massive black
hole could be the "engine" that powers a quasar or active galaxv. .\t the
center of a quasar or active galaxy, nature may be drawing upon the tremendous amounts of energv tied up in a black hole's powerful gravitational field.
As we saw
in
Chapter
business. At best,
we can
15,
finding black holes
is
a trickv
and
difficult
see only the effects of the hole's gravitv
and
trv
non-black-hole explanations of the data. This general

approach has been applied to the giant clli|)tical galaxv M87.
The active galaxv M87 sits near the center of the \irgo cluster onlv
50 million light years from Earth. Figure 17-9 shows the full extent of
this galaxv. Brief photographic exposures, like those in Figure 18-17,
reveal a bright, starlike nucleus from which a jet protrudes.
In the mid-1970s, several Caltcch astronomers headed bv Wallace Sargent made detailed spectroscopic observations of M87. carefullv examining the widths of spectral lines across its face. As sketched in Figure ISIS, a spectral line is broadened because of the motions of the individual
stars along the line of sight. 1 he spread in stellar velocities, called the
velocity dispersion, is directiv related to the overall width of a speiti.il
to rule out
all
line.
Sargent and
M87
his colle.igues (oiuid that spectral lines in the center of

are excejjtionallv broad, indii.iting a large range of siellai velocities.
Quasars ami tictwe galaxies
Figure 18-17
The core and jet of M87

luminmn jet
short lime exposure rexvah a
surging out of the bright,
MS7.
is
starlike nucleus
(a) This computer-processed
of
photograph
colored to shuic the galaxy's stars in red
Willi
light emitted b\ relalivi.stir electrons in blue.
(b) This
from
computer-enhanced exposure, made
several photographic plates, shoii'S details
in the jet.
The jet
is
about
6500
long. (Courtesy of T. Stephenson.
and J.
J.
light years
H. C. Arp.
Lone)
In other words, some stars near the center of M87 are coming toward us
and others are rushing away from us at speeds far greater than those we
would expect to find in an ordinary galaxy. This observation suggests
M87
along compact,
extremely massive object is located at the galaxy's center. Without the gnu ity of such
an object to keep the stars in their high-speed orbits, they would long
ago have escaped from the galaxy's core. From his observations, Sargent
estimated that the mass of this central object is about 5 billion solar
masses. That much matter confined to such a small volume strongly suggests the existence of a supermassive black hole.
that these stars are rapidly circling the center of
high-speed orbits, which
is
possible only
if
some
sort of
^
aiul iirlive galaxies
Figure 18-18
The broadening of galaxy
absorption lines
viduiil ilars are
how fast
The
sfierlKil lines / iiidi
Duppler shifted aecurdiug
moving parallel
the stars are
lu
our
In
of sight. From Earth, however, we see the

combined light of all the stars along our line
line
of
sight.
The
to tlie
we
spectral line that
fore broadened by an
amount
spread in stellar
see
is
there-
ilirecth related
velocities.
While Sargent was making his spectroscopic observations, a second

Caltech team headed b\ Peter Voiing carcfiilh measured ilie brigiitness
across the face of M87. Using Newtonian mechanics, it is possible to calculate how stars should be distributed on axerage in an elliptical galaxy.
Calculations of this type had been
University of California
at
done
in
1966 by Ivan King
Berkeley. King's theoretical models
what the brightness should be across the face of an ordinary

10
Distance
rriiiii
Figure 18-19
M87
measured
center.
Each
hrightiu-.', (),ns
iiiilward jrimi the galaxy's
dot represents
an
is
model
hole at
fits the
rei'ised to
tlie
Young and
data)
is
the
solid
standard
galaxy's center. (Ailiiptrd
jrom
I'.
us
loinid that bis observations did not agree with King's calcula-
As plotted
in
figure 18-19. M87's core
would be expec ted from
is
much
brighter than
a sini])le distribution of billions of stars alone.
account for the bright core, \'oung had to add a powerful, centrally
located source of giaxily to King's cak illations. With this source
include a supermassive black
colleagues)
Yoimg
tions.
the stan-
dard model calculated by Ivan King. The
tell
elliptical
galaxv.
I'o
iruliindual
measurement. The dcLshed curve
curve (which
The luminosity profile of
This graph diow, the
A/<V7.
10(1
galaxy's (tnlcr (arc sit)
at the
ity,
stars naturally
t)f
grav-
crowd aroimd the (enter of M87, thereb\ jjroduc-
ing a bright nucleus. Once again, the data suggest the pre.sence of a
5 X 10-' M<:y black hole.
Various realistic scenarios about how to tap the gravitational energy of
an extremely massive black hole have been worked out by Richard
Qiiasiin
Lovelace
Relativistic jet
Relativistic jet
Figure 18-20
supermassive black hole
The energy output of active galaxies and quasars may involve extremely massive black holes
that accrete matter
from
their surroundings. In
the scenarin drpirtcd here, the inflow of material Ihiinio/i
,111
iiiininin disk pnidiues two
powerjni. h'IhIivisIk
jets.
at
and
active galaxies
Cornell University and Roger Blandford at Caltech. In
es-
sence, these schemes are simply scaled-up versions of the explanation of
Cygnus X-1 discussed in Chapter 15. A large black hole is needed to

ensure a large energy output.
Imagine a billion-solar-mass black hole sitting at the center of a galaxy. The center of a galaxy is a very congested place, and we would expect this supermassive black hole to be surrounded by a huge rotating
accretion disk of matter captured by the hole's gravity. Because of the
constant inward crowding of matter in the accretion disk, the gas pressure surrounding the hole is tremendous. To relieve the pressure, matter
is ejected perpendicular to the accretion disk, the direction along which
the gases experience the least resistance. The result is two oppositely
directed beams of particles. The overall scenario is sketched in
Figure 18-20.
Details of an acceretion disk orbiting a supermassive black hole were
elucidated in 1984 by astrophysicists Larry L. Smarr and John F. Hawley
using a Cray-1 supercomputer. Their results show that matter in the accretion disk is accelerated to supersonic speeds as it plunges toward the
hole. Near the hole, however, this inward rush of gas is abruptly stopped
by a "centrifugal barrier" caused by the rapid orbiting of the gas around
the hole. This abrupt halt produces a standing shock wave that marks
the inner edge of the accretion disk, as shown in Figure 18-21. A
funnel-shaped cavity surrounding the axis of rotation of the accretion
disk is kept empty by centrifugal forces. The continuous deluge of matter crashing onto the standing shock wave propels gas along the walls of
the evacuated funnel, channeling the gas into two oppositely directed
beams that emerge from either side of the accretion disk.
Mechanisms of this type may explain double radio sources and some
of the jets and beams we see protruding from active galaxies and from
Figure 18-21
Standing shock waves in an

Gas plunges at supersonic
accretion disk
speed toward the black hole at the center of an

accretion disk,
this
shown here
in cross-section. In
diagram, color represents gas density: red
for the most dense
trum
to
funnel
blue fui
is
rei;iiin\
llir
through the spec-
linsl druse.
The evacuated
parallel In the axis of rotation of the
accretion disk.
The
black arrow-shaped fea-
tures in the deruest part of the accretion disk

lire
sliinding shock
iiilnlliHi^ i;in
waves created where the
slows abruptly
when
it
encounters
the lentrijugal barrier at the walls of the evac-
uated funnel. (Courtesy of J.

L. L. Smair)
F.
Hawley and
(imt iirlivf galuxies
(liiiisais
quasars. To account for the luminosity of a quasar, a 10^

hole would have to accrete matter at the rate of 10 A/ per year.
some
The
possible existence of supermassive black holes in quasars
tive galaxies
is
one of the most exciting
topics in
black
and
ac-
modern astronomy.
The Hubble Space Telescope will be used to probe the nuclei of galaxies
with much higher resolution than is possible from the ground. These
observations will undoubtedly reveal many details about the powei ful
"engines" at the cores oi active galaxies and quasars.
Summary
.\
quasar (or quasi-stellar object) is an object in the sk\ tliat looks like a
huge redshift corresponding to a great distance from the
star but has a
Earth.
To
be seen at all from Earth, a quasar must be very luminous

about 100 times brighter than an ordinary bright galaxv.
1\
picallv
Rapid fluctuations
be
much
in the brightnesses of quasars indicate that thes cannot

larger than the diameter of our solar system.
An
active galaxy is an extremely luminous galaxy that has one or more

unusual features: an unusually bright, starlike nucleus: strong emission
lines in its spectrum; extreme variations in luminosity, or jets or beams of
radiation emanating
from
its
core. Seyfert galaxies,
galaxies,
and BL
Lacertae objects are examples of active galaxies.
Most double radio sources seem to have an acti\e galaxv located between
the two radio lobes that distinguish this type of radio source.
The
strong energy emission from quasars, active galaxies, and double

radio sources may be produced as matter falls toward a supermassive
black hole at the center of the object.
Review questions
Suppose you saw an object
sar.
What
sort
in the sky that you suspected might be a quaof observations might you make to find out if it was indeed a
quasar?
2 Explain why astronomers do not use any of the standard candles described in Chapter 17 to determine the distances to quasars.
3 In the 1960s, it was suggested that quasars might be compact objects

ejected at high speeds from the centers of nearby ordinarv galaxies. Whv
does the absence of blueshifted quasars disprove
4
is
Why do
this
hvpothesis?
astronomers believe that the energv -producing region of
a (jua-sar
very small?
5 In
wii.il
wa\s arc active galaxies similar to quasars?
Why do
Compare and
you suppose there are no quasars relativeh ncai our
s;ala\\?
contrast SS 433 discussed in Chapter 11 with a tvpical dou-
ble radio source.
Whv do many
(|uas,ii
Advanced questions
IS
astronomers believe that the "engine" at the center of a

supermassive black hole surrounded by an accretion disk?
9 Some c|uasais show several sets of absorption lines whose redshifts are less
than the redshift of the quasars' emission lines. For example, the quasar PKS
0237-23 has five sets of absorption lines, all with redshifts somewhat less
Quasars anil active galaxies
than the redshift of the quasar's emission

these sets of absorption lines.
lines.
Propose an explanation for
When
quasars were first discovered, many astronomers were optimistic

extremely luminous objects could be used to probe distant regions
of the luiiverse. For example, it was hoped that quasars would provide highredshift data from which the Hubble constant could be accinately determined. Whv do you suppose these hopes have not been realized?
10
that these
Discussion question
11 Speculate on the possibility that quasars, double radio sources, giant ellipand so on form some sort of evolutionary sequence.
tical galaxies,
For further reading
"Quasars with Fuzz." Mercury, May/June 1983, p. 81.

Blandford, R., et al. "Cosmic Jets." Scientific American. May 1982.
Burns, J., and Price, R. "Ceiitaurus A: The Nearest Active Galaxy." Scientific American, Nov. 1983.
Ferris, T. "The Spectral Messenger: The Redshift Controversy." Science
Balick, B.
81, Oct. 1981, p. 66.
Kaufmann, W. "Exploding Galaxies and Supermassive Black Holes." Mercury, Sep./Oct.
Tananbaum,
1978, p. 97.
and Lightman, A. "Cosmic Powerhouses: Quasars and

Black Holes." In Cronell, J., and Lightman, A., eds. Revealing the
H.,
Universe.
Wyckoff,
S.,
ies?"
MIT
Press, 1982.
and Wehinger,
%&
Telescope,
"Are Quasars Luminous Nuclei of GalaxMar. 1981, p. 200.
P.
19
Cosmology: The creation and fate

of the universe
The creation of matter from energy

phiAuiipaph shows the coyiverswn oj a
I'liis
gamma
ray inln mailer inside a bubble chamber, a
dnnce
filled with liijiiid
hydrogen thai
is
de-
signed
to
make
visible
a long row of tiny bubbles. The path
of
tlie
gamma
the path of a
ray
is
charged particle
not visible because pho-
tons are electrically neutral.
Near
the bottom
of the photograph, the energy carried by the
gamma
an
ray
is
converted into an electron
antieteclron.
Because
oj a
and
magnetic field
surrounding the bubble chamber, the electron n

deflected to Ihe right while the antieteclron
veers toward the
tron
left.
The path of a
stray elec-
also seen al the right. (Courtesy of
iMwrence Berkeley Laboratory)
We
live
(in
expanding
in this chapter, we leani that the expanan explosion of space at the beginning of
unii'erse.
sion of our universe began with

time.
from
We find
that the universe
is
filled with microxvave photons left over
the primordial fireball that filled all space shortly after the
Big Bang.
expand forever or collapse in a Big
(Crunch, depending on the density of matter throughout space. We examine
the idea that all the forces oJ nature had the same strength immediately
after the Big Bang. We then see how the variom forces and familiar
particles in the world around us "froze out" of the new-born universe as
it expanded and cooled.
We
discover that the universe
As
loolisli as
ask
is,
"Why
it
is
may
the
sccni.
may
one
either
of the
most pi-ofbund questions yon can
sivv claik at niirht"-"
This question aijp.neiuK haunic'd
Johannes Kepler as long ago as UJIO. and it was popiilaii/ed in liie earlv
IHOOs by the (ierman aniatein- astionomer Heiniiih Olheis.
To appreciate the pioblem, you must ix-gin In assiuiiing thai the
universe is infinite and that stars are scattered inore-or-lcss randomly
across this infinite expanse of space. Isaac Newton argued that no other
Cosmology: The creation and fate of the universe
assumption makes sense. If the universe were not infinite or if stars were
grouped in onlv one part of the universe, then the gravitational forces
between the stars would soon cause all this matter to fall together into a
compact blob. Obviously, this has not happened. Thus, as classical Newtonian mechanics would have it, we must be living in a universe that is
infinite and static. Only then does each star feel a uniform gravitational
pull from every part of the sky, from all the other stars in the universe.
According to this model, the universe can exist forever without major
changes in its structure.
Imagine looking out into space in this static, infinite universe. Because
space goes on forever with stars scattered throughout space, your line of
sight must eventually hit a star. No matter where you look in the sky,
you should see a star. The entire sky should therefore be as bright as an
average star. Even at night, the entire sky should be blazing like the surface of the Sun. That this is not so is the dilemma called Olbers's paradox.
We
live in
an expanding universe
tells us that there is something wrong with Newton's

idea of an infinite, static universe. According to the classical, Newtonian
Olbers's paradox
picture of reality, space
is
laid
out in
all
directions like a great, infiexible,
flat space stretches on

independent of stars or galaxies or anything else. Similarly, a Newtonian clock ticks steadily and monotonously forever, never
slowing down or speeding up.
Albert Einstein demonstrated that this view of space and time is
wrong. In his special theory of relativity, he proved that measurements
with clocks and rulers depend on the motion of the observer. As explained in Chapter 15, Einstein's general theory of relativity tells us that
the shape of space is profoundly influenced by the masses occupying
that space. In other words, matter tells space how to curve, and curved
space tells matter how to move.
flat
sheet of rectangular graph paper. This rigid,
and on,
totallv
Shortly after formulating his general theory of relativitv in 1915, Albert Einstein applied his ideas to the structure of the universe.
To
his
dismay, his calculations could not produce a truly static universe, indicating instead that the universe must be either expanding or contracting.
The prevailing Newtonian opinion was so strong that Einstein doubted
the validity of his equations and missed the opportunity to postulate that
we
live in an expanding universe.

Edwin Hubble is usually credited with discovering that we live in an
expanding universe. As we saw in Chapter 17 (recall Figure 17-21),
Hubble discovered the simple linear relationship (now called the Hubble
law) between the distances to remote galaxies and the redshifts of those
galaxies' spectral lines.
The
greater the distance to a galaxy, the greater
Thus, remote galaxies appear to be moving away from us

with speeds proportional to their distances (recall Figure 17-22). Since
is its
redshift.
the galaxies are getting farther and farther apart as time goes on, astronomers sav that the universe is expanding.
What does it actually mean to say that "the universe is expanding"?
According to relativity theory, space is not rigid. Instead, the amount of

space between widely separated locations can change over time. A good
analogy is that of a person blowing up a balloon, as sketched in Figure
19-1. Small coins, each representing a galaxy, are glued onto the surface
of the balloon. As the balloon expands, the amount of space between the
Cosmology: The creation and /ale of
Figure 19-1
analogy
pared
lite
universe
The expanding balloon
The exfianding universe ion be roniexpanding surface of an inflating
to the
balloon. All the coins on the balloon recede
from one another
as the balloon expands,
as all the galaxies recede
(Adapted from
Thome, and J. Wheeler}
the universe expands.

ner, K.
from each other

('..
ju.'.t
as
Mis-
and
larger, in the same wav, as the universe expands,

between wideh separated galaxies grows larger and
larger. The expansion of the universe is the expansion of space.
The expanding-balloon analogy reveals several important characteristics of the expanding universe. For example, imagine sitting on one of
the coins in Figure 19-1. As the balloon expands, vou see all the other
coins moving away from you. Specifically, )ou obser\e that nearby coins
move away from you slowly, whereas more distant coins move awav
more rapidly. You find this Hubble-law-like relationship no matter which
coin you call home. Your coin is not at the center of the balloon, of
course. Indeed, the surface of the balloon does not have a center: you
coins gets larger
the
amount
of space
move around the surface forever without finding a center. And

the surface of the balloon has no edge; you could explore everv inch of
the surface and never find an edge.
could
Just as the surface of the balloon has neither center nor edge, our
universe has no center or edge. No matter what galaxv vou call home,
all the other galaxies are receding from vou. No one is ever at the "center" of the universe. Questions like "what is beyond the edge of the
universe?'"or "what
is the universe expanding into?" are as meaningless

"what is north of the Earth's north pole?"
The ongoing expansion of space explains how photons from remote
galaxies are redshifted. Imagine a photon coming toward us from a distant galaxy. As the photon travels through space, space itself is expanding, so the photon's wavelength becomes stretched. When the photon
reaches our eyes, we see a longer wavelength than usual, which is the
redshift. The longer the photon's journev. the more its wavelength will
have been stretched. Thus, photons from distant galaxies have larger
redshifts than those of photons from nearb\ galaxies, which is expressed
as asking
as the
Hubble
law.
l)v the expansion of the universe is propeilv called

cosmological redshift to distinguish it from a Doppler shift. Doppler
shifts are caused bv an object's motion through space whereas a cosmological redshift is caused by the exfrnnsion of space.
Finally, it is important to realize that the expansion of space occurs
redshift caused
only in the voids that separate clusters of galaxies, [ust as the coins in
Figure 19-1 do not expand as the balloon inflates, galaxies theinselves do
not expand. Finstein and others have established that a galaxv is always
contained within a |)atch of nonexpanding space. Fhe galaxv's gravitational field
from the
produces
flat,
this
nonexpanding patch, which is indistinguishable

Newton. Thus, the F.arth and \()ur bodv are
rigid space of
not getting bigger

in intergalactic
and bigger; the expansion
space.
of the universe occurs
onh
C.iismoUigy: Tlir
The Big Bang was an
rrmhiDi mid
fiili-
nj llic
universe
Ihe universe has been expandinj^ lor Isillions of years, so there niusl
ha\e been a time in the ancient past wlien all the matter in the universe
was concentrated in a state of infmite density. Presumablv some sort of
colossal explosion occurred to start the expansion of the universe. Ihis
explosion, commonly called the Big Bang, marks the creation ol ihe luii-
explosion of space at the

beginning of time
verse.
The Hubble
verse, can be
constant, which
used
us the rate of expansion of the unitime that has elapsed since the Big
tells
to estimate the
Bang, as follows:
120 billion
vears
15 km/sec/Ml\
is probablv between 15 and hS bilmutual graxitational attracti(jn. galaxies ha\e
not been flying away from each other with a constant velocity. Gra\it}
has caused the speed of separaticjn between galaxies to decrease gradually since the Big Bang. Thus, the expansion rate of the uni\erse has
been decreasing. An age of 20 billion years assumes no deceleration, so
this value would be valid only for an empty universe that contains no
matter and thus no gravity to slow the expansion.
The finite age of the universe offers a resolution of Olbers's paradox.
The entire sky is not as bright as the surface of the Sun because we cannot see any stars that are more than 20 billion light years away. The universe may indeed be infmite, with galaxies scattered throughout its limitless expanse. However, because the uni\ersc is less than 20 billion years
old. the light from stars more than 20 liillion light years awav has simplv
not had enough time to get here.
You can think of the Earth as being at the center of an enormous
sphere with a radius of roughlv 20 billion light vears (see Figure 19-2).
The surface of this sphere is called the cosmic particle horizon. The
entire observable universe is located inside this s])here. We cannot see
anvthing beyond the cosmic particle horizon because the travel time for
light from these incredibly remote objects is greater than the age of the
universe. Throughout the observable universe, the distribution of galaxies is sufficiently sparse that most of our lines of sight do not hit any
stars, which explains whv the night skv is dark.
The idea of a Big Bang origin of the universe is a straightforward,
logical consequence cjf ha\ ing an expanding universe. If vou just imagine far enough into the past, you arrixc at a time nearly 20 billion years
ago when the density of matter throughout the universe was infinite.
The entire universe was like the center of a black hole. For this reason,
perhaps a better name for the Big Bang is the cosmic singularity.
Comparing the Big Bang to the center of a black hole helps us appreciate certain aspects of the creation of the universe. As we saw in Chapter 15, matter is crushed to infinite densitv at the center of a black hole.
This location, called the singularit\-. is characterized b\ infinite curxature
where space and time are all tangled up. \\'ithout a clear background of
space and time, such concepts as jj(i.sl. future, here, and now cease to have
any meaning.
At the moment of the Big Bang, a state of infinite densit\ filled the
universe. Space and time throughout the universe were completelv jumbled up in a state of infinite curvature like that at the center of a black
hole. Thus we cannot use the laws of physics to tell us exactly what happened at the moment of the Big Bang. We certainlv cannot use science
to tell us what existed before the Big Bang. These things are fundamen-
Actualh
the true age of the universe
lion vears.
Cosmic
particle
horizon
Figure 19-2
ladius of
tile
The observable universe

cosmic pdihcte horizon
is
The
equal
to the (tislance that light hcti traveled since the
Big Bang. The Big Bang occuned about 20

billion years ago.
o about 20
.so
the cosmic particle horizon
billion light years
away.
We
can-
not see stars beyond the cosmic particle horizon
because then tight has

teach
us.
iint
hail
enough lime
In
Because
of their
(',o\iftuluf:n;:
The
(irdfioii
and
/alt-
tij
ihr iiniirrse
unknowable. I'lie terms "hi'fdtT the Big Bang" or "at the wniiinil of
Big Bang" are meaningless, because time itself did not realls exist.
In a very short time after the Big Bang, space and time came to exist
in the way we think of them today. Ihis interval, called the Planck time,
etiuals 10^^* sec. From the moment of the Big Bang (at time I = 0) to
the Planck time 10"''^ sec later, all known science fails us. We do not
know how space, time, and matter behaved under these extreme circumstances. Nevertheless, some physicists speculate that space and time as
we know them today burst forth from a seething, foamlike, space-time
mishmash during the Planck lime. We can think of the Big Bang as
an explosion of space at the beginning of time.
tall\
tlie
Microwave
radiation that
space
evidence for a
all
is
hot Big Bang
fills
One
of the major successes in modern astronomy involves discoveries

about the origin of the heavy elements. We know today that all the
heavy elements are created in the flerv infernos at the centers of stars.
As astronomers in the 196()s began understanding details of thermonuclear synthesis, a new problem arose: there is too much helium around.
For example, the Sun consists of about 7'^ percent hydrogen, 24 percent
helium, and 3 percent for all the remaining heavier elements combined.
Thi.s 3 percent can be understood as material produced at the centers
of ancient stars that long ago cast these heavy elements into space. Certainly, some freshly made helium accompanied these heavy elements, but
not nearly enough helium to account for one-quarter of the Sun's mass.
Shortly after World
War
II,
George
Gamow
at
George Washington
University proposed that, immediately following the Big Bang, the universe must ha\e been so incrediblv hot that thermonuclear reactions
could have occurred evcr\where thioughout s]3ace. Following up this
idea in 1960, Piinceton ph\sicists Robert Dicke and P.J. F. Peebles disc(3vered that they could indeed account for today's high abundance of
helium by assuming that the early universe was at least as hot as the
Sun's center (where helium is currently being produced). The early universe must therefore have been filled with manv high-energy, shortwavelength photons.
The universe has expanded considerablv since those ancient times,
and all those short-wavelength photons ha\e become so stretched that
they are now low-energy, long-wavelength photons. Thus, the temperature of this cosmic radiation field should now be quite low, perhaps onlv
a few kelvins above absolute zero. According to Wien's law, radiation at
this low temperature should ha\e its peak intensiiv at microwa\e wavelengths of a few centimeters, fn the earlv-19(i()s, Dicke, Peebles, and
their colleagues began designing an antenna to detect this radiation.
Meanwhile, a few miles away from Princeton University, Arno Penzias
and Robert Wilson of Bell Telephone Laboratories were working on a
new microwave horn anteima (see Figure 19-3). This antenna was designed to relay telephone calls to Farth-orbiting communication satellites.
Penzias and Wilson were deeply puzzled because no matter where they
pointed their antenna in the sky, thev detected a faint background noise.
I'hanks to a friend. the\ soon learned of the work o( Dicke and Peebles.
Pen/ias and Wilson realized that the\ had detected the (ooled-down cosmic backgiound ratliation left o\er from the hot liig Bang.
Since those pioneering da\s, scientists have made man\ measurements
of the intensity of this background radiation at a vai iety of wavelengths
(see Figure 19-4). I'he dashed curve that fits the data in Figure 19-4 is a
blackbody spectrum with a temperature of nearly 3 K. Because of its
Cosmology: The creation and j ale of
llie
universe
The Bell Labs horn antenna
Figure 19-3
This horn antenna at Holmdel.
New Jersey,
was used by Arno Penzias and Robert Wikon

to detect the cosmic microwave background.
(Bell Labs)
w""
10*
%'
Blackbody
at 2.7
0,1
0.01
Waseiength {cm)
-*
The spectrum of the cosmic

microwave background Measurements of
Figure 19-4
intensity at various
and
the cosmic
the
wavelengths (indicated by
shaded area) demonstrate that
microwave background
is
blackbody
radiation with a temperature of nearly 3 K.
\
KAAM^^-
Figure 19-5
^'^^^
Our motion through
Cool
the
Doppler
effect,
Because of the
the microwave background
is
warmer in that part of the shy toward

which we are moving. Recent measurements
slightly
indicate that our galaxy, along with the rest of

the Local
rection
Group
space
is
commonly
warmer than average in the direction of the constellation of Leo,

and in the opposite direction (toward Aquarius) the microwave background is slightlv cooler than a\erage. Between the warm spot in Leo
and the cool spot in Aquarius, the background temperature across the
sky declines in a smooth fashion.
slightly
This temperature variation can be explained as a result of the Earth's

motion through the cosmos. If we were at rest with respect to the
microwave background, the radiation would be truly isotropic. Because
we are moving through this radiation field, however, we see a Doppler
shift. As sketched in Figure 19-5, we see shorter than average wavelengths in the direction toward which we are moving. A decrease in
wavelength corresponds to an increase in photon energy and thus to an
increase in temperature. The observed temperature excess corresponds
to a speed of 390 km/sec.
Conversely, we see longer than average wavelengths in that part of
the sky from which we are receding. An increase in wavelength corresponds to a decline in photon energy and hence to a decline in temperature. We are thus traveling from Aquarius toward Leo at a speed of 390
km/sec. Taking into account the known velocity of the Sun around the
center of our galaxy, we find that the entire Milky Way Galaxy is moving
at 520 km/sec in the general direction of the Virgo cluster, probably because of the gravitational attraction of this massive, rich cluster.
To examine this motion and many other properties of the microwave
background, such as possible deviations from a perfect blackbody spectrum, astronomers are looking forward to the launching of the Cosmic
Background Explorer satellite (COBE) in the late 1980s (see Figure
overall
Hot
fills all
3-K cosmic microwave background.

An important feature of the microwave background is that its intensity is almost perfectly isotropic (meaning "the same in all directions"). In
other words, we detect nearly the same intensitv from all parts of the
sky. However, extremely accurate measurements made in high-flying airplanes reveal a very slight anisotropy. The microwave background is
called the
the dots
various properties, this radiation field that
is
of the Virgo
moving
cluster.
in the general di-
(û.Mno/iig\:
(mil
fall' /
llw universe
The Cosmic Background
Figure 19-6
Explorer (COBE)
be launched
The nealion
I'his sali-lli/e. sclieiluU-d In
1989,
will study the spectrum
and angular distribution of the cosmic background radiation over a wavelength range u/
1
II.
to 1
cm.
COBE
(pronounced co-bee)
look for dei'ialioiis from
<i
spectrum and from perfect

of J. Mather.
will
perfect blackboch
iîitropf.
I(:iiurti'^\
NASA)
19-6). Unhampered by the Earth's atmosphere, COBE's instruments will

enable astronomers to explore many aspects of this extremely ancient
radiation.
The future
of the universe
determined by the average

density of matter in it
is
During the 1920s, general relativity theory was applied to cosmology by

Alexandre Friedmann in Russia. Georges Lemaitre in France, Willem de
Netherlands, and, of course, by Einstein himself. The resulting picture of the structure and evolution of the universe, called relativSitter in the
istic
cosmology,
is
in surprisingly
good agreement with our
intuitive no-
tion that gra\ity should be slowing the cosmological expansion.
Imagine a rocket blasting off from the surface of the Earth.

rocket's speed
If the
than the escape velocitv (about 1 1 km/sec), the

spacecraft will fall back to Earth. If the rocket's speed equals the escape
velocity, the spacecraft will just barely manage not to fall back. If the
rocket's speed exceeds the escape velocity, the spacecraft will easilv leave
the Earth and never fall back, despite the relentless pull of gravitv.
A similar set of circumstances surrounds the ultimate fate of the universe. Instead of talking about some sort of "mutual esca])c \el()cil\
l)etween galaxies, however, asironomcis speak of the average density of
matter in the universe.
If the average density of matter thiougliout space is small, the graviis
associated with this matter is weak, and the expansion of the universe
will continue forever. Even infinitely far into the future, galaxies w'ill
continue to rush awa\ from eai h othi'r. In such i ircumslances, we say
that the universe is unbounded.
(Conversely, if the density of matter across spate is large, the resulting
gravity is strong enough to halt the expansion of the universe eventuallv.
The universe will rea( h a maximum si/e, then begin to tontract as gravity starts to pull the g.il.ixics back tow.ud each otiiei. In this case, we say
that the universe is bounded.
is
less
"
Cosmology: The creation and jate
nj the universe
Separating these two scenarios is the situation in which the universe is

marginally bounded. In this case, the density of matter across space exactly ecjuals the critical density (P(), so that the galaxies just barely manage to keep moving away from each other. This case is analogous to the
rocket leaving the Earth at a speed exactly equal to the escape velocity.
The size of the critical density depends on the Hubble constant. Using
a Hubble constant of 15 km/sec/Mlv. we find that
4.5 X
10'-'" g/cnr'
which is equivalent to a densitv of about three h\clrogen atoms per cubic

meter of space.
The average density of luminous matter that we see in the sky seems
to be about 3 x 10"^' g/cm^, clearly less than the critical density. However, as discussed in Chapter 17, the missing-mass problem for clusters
of galaxies strongly suggests that we may be seeing only a tenth of the
matter in the universe. Most of the mass of the universe seems to be
dark and remains to be discovered. In short, present-day observations
and estimates of density are not accurate enough to tell us whether the
luiiverse
The deceleration
diagram
Deceleration
and
the
Hubble
These two graphs compare the evo-
lution of the universe
and
the
appearance of
Hubble diagram. The case qo =
the
bounded or unbounded.
Another way of determining the future of the imiverse is to measure the

rate at which cosmological expansion is slowing down. The deceleration
of the imiverse shows up as a deviation from the straight-line relationship predicted by the Hubble law. Suppose that you measure the redshifts of galaxies several billion light years from Earth. Light from these
galaxies has taken billions of vears to get to your telescope, so your
measurements reveal how fast the universe was expanding billions of
years ago. Because the universe was expanding faster in the past than it
is today, your data will deviate slightly from the straight-line Hubble law.
The graphs in Figure 19-7 display the relationship between the deceleration of the universe and the Hubble law, extended to great distances.
Astronomers denote the amount of deceleration by the deceleration parameter ^(). Appropriately, q(^ =
corresponds to no deceleration at all.
This state is possible if the universe is completelv einpty and has no
gravity to slow the expansion. As sketched in Figure 19-7a, the q^ =
universe expands forever at a constant rate.
of the
universe can be determined
from observations of
extremely distant galaxies
Figure 19-7
is
is
an
empty universe (p = 0). The case qo = i is a

marginally bounded universe (p = pj. If
<
qo
<
i.
then the universe
is
unbounded and
expand forever. If qg > i, the universe

bounded and wilt someday collapse.
will
is
Tht' size ol ihe univ
Cosmology: The rmilion
.>00,0()0
^
y^
1
the critical density
r /
Vn =
50,000
20,000 :
.^
10.000
If
will
-^
tude of the brightest galaxy
is
This
to
The magni-
a cluster
is
di-
with the distance to the clus-
If the data points Jail between the cuti'es

q,,
= U and
unbounded. If
cutve marked qn
q,,
i.
then the universe
the data points [all
i.
above the
then the universe
is
bounded. (Adapted from A. Sandage)
The shape
and
i.
continue to expand forever because
the iniiverse
it
is
unbounded and
contains matter
at less
than
fill
me.
should be possible to determine f/,, b\ measuring the

many remote galaxies, then plotting the data
on a Hubble diagram like the one in Figure 19-76, If the data points fall
above the ^n = i line, the universe is bounded. If the data points fall
between the ô =
and î, = i lines, the universe is unbounded.
Unfortunately, such observations are extremely difficult to make. Galaxies nearer than 10''' ly are of no help in determining r/,,. Beyond 10' Iv.
galaxies are so faint and indistinct that uncertainties cloud the distance
and redshift determinations. For example. Figure 19-8 shows data obtained bv .-Mian Sandage of the Mount Wilson and Las Campanas Observatories. Note how the data points are scattered about the 90 = 5 line.
This scattering, a result of observational uncertainties, prevents us from
determining conclusively whether the universe is bounded or unbounded. Nevertheless, because the data lie close to the r/,, = i line,
we may conclude that the universe is not far from being marginally
bounded. Many astronomers hope that the Hubble Space Telescope will
enable them to do a better job of determining the vakic of q^.
it
redshifts ancf distances of
The Hubble diagram
include exinnely remote galaxies.
marked
p^.
the range between
In principle,
luslir galaxy
graph shows Ihe Hubble iliagiam extended
rectly correlated
in
9() is greater than i, the universe is bounded and is filled with matof a density greater than p, This universe is doomed to collapse in
upon itself and ultimatelv end in a Big Crunch in the extremch distant
ter.
is
ter
M.iMllU(U- ol bn!,'hlcsl
Figure 19-8
^1,
If
yi
2000 -
uuwerse
the critical density.
/^'
5000
10(10
100,000 -
J
>
of Ihe
The case ô = i corresponds lo a marginally bounded universe. Such

a universe just barely manages to expand forever. It contains matter at
?(i
iiiirl fiilc
of the universe
is
parameter and to the

average density of matter
the universe
There
another way
of the universe.
based on Einstein's general theory
of relativity, which explains that gravity cmves the fabric of space. The
gravity that is slowing the expansion of the universe must also give space
an overall "shape." This shape, or geometry, is directly related to the
average density of matter throughout space which, in turn, is related to
the deceleration parameter. By measuring the shape of the universe, we
might discover its ultimate fate.
To see what astronomers mean by the geometry of the universe,
imagine that we shine two powerful laser beams out into space. Suppose
thai these two beams are aligned so that they are perfectlv parallel as
thc\ leave the Earth. Finally, su|)pose that notiiing gets in the way of
these two beams, so that we can follow them for billions of light years
across the universe, across the s])ace whose curvature we wish to detect.
There are only three possibilities for what we might di.scover. First, we
might find that our two beams of light remain perfectly parallel, even
after traversing billions of light vears. In this case, we would conclude
thai sjjate is not cui\ed: the uni\erse has zero curvature and space is
is
of investigating the future fate
Our understanding of
related to the deceleration

in
the universe
is
flat.
we might find that our two beams of light graduallv get

and closer together as they move across the universe, so that thev
eveniually intersect al some enormous distaiue from Earth. In this case,
space is not flat. Recall that lines of longitude on the Earth's surface are
parallel at the etjuatoi bin intersect at the |)oles. The shape of the universe must be analogous lo ihe surface of a spluie. We iherefore say
that space is spherical and the universe has positive curvature.
Alternatively,
closer
Cosmology: The creation and fale of the universe
The
third and final possibility is that the two parallel beams of light
gradually diverge, becoming farther and farther apart as they move
across the universe. In this case, we say that the universe has negative
curvature. A sphere is a positively curved surface, and a saddle is a good
will
example of
a negatively curved surface. Just as parallel lines drawn on a

sphere always converge, parallel lines drawn on a saddle always diverge.
Mathematicians say that saddle-shaped stnfaces are hyperbolic. In a negatively curved universe, space is hyperbolic.
These three cases describing the geometry of the imiverse are
sketched in Figure 19-9. To illustrate them, three surfaces are drawn: a
plane, a sphere, and a saddle. Real space is of course three dimensional,
but it is much easier to visualize a two-dimensional surface. Thus, as you
examine the drawings in Figure 19-9, remember that the real universe
has one more dimension. For example, if the universe is hyperbolic, the
geometry of space is the three-dimensional analogue of the two-dimensional surface of a saddle.
Each of the three possible geometries corresponds to a different behavior and fate for the universe. Flat space corresponds to the margin-
Spherical space
>
Flat spac
Figure 19-9
p^
and
^,j
>
and
^(,
1/2
1/2
The geometry of the universe
The "shape" of space

ter
Pc
is
determined by the mat-
contained in the universe. The curvature
either positive, zero, or negative,
is
depending on
whether the average density throughout space

is
greater than, equal
cal density.
to,
or
less
than the cnlip
<
p,
and
<
^(1
<
1/2
Cosniolvgy: The crention
TABLE
1 9- 1
The geometry and fate
ami fate
the universe
i>j
oj the universe
Ultimate
Geometry
of space
Curvature
of space
Spherical
Positive
Deceleration
Type of
type of
space
parameter
universe
universe
Greater than the
Greater
critical
Flat
Zero
future of this
Average density
throughout
than
density
(qo)
Eventual
Closed
collapse
Exactly equal to
Exactly
the critical density
equal to
Less than the
Between
and i
Perpetual
Flat
expansion
(just barely)
Hyperbolic
Negative
critical
ally
density
case (q =
bound
i)
in
Open
Perpetual
expansion
which the galaxies just barely manage to keep
receding from each other, fhis flat-space scenario divides the positivecurvature cases from the negative-curvature cases. If the density across
space is greater than the critical density, then î, > |. and space is positively curved. Conversely, if the density across space is less than the criti< ô < i, and space is negatively curved. These relacal density, then
tionships are sumtnarized in Table 19-1.
Note that both the flat and the hyperbolic universes are infinite. They
extend forever in all directions, so that they have neither an "edge" nor
a "center." In contrast, the spherical universe is finite. Nevertheless, it
also lacks a center or an edge, just as the surface of a sphere has neither
an edge nor a center.

bility
The flatness of the universe

and the isotropy of the
suggest that a period of
vigorous inflation followed
the Big Bang
Relativistic
cosmology
strictly rules
out an\ possi-
of a center or edge to the universe.
Ever since Edwin Hubble discoveied that the universe is expanding, astronomers have struggled to determine the deceleration parameter. During the 1960s and 1970s, various teams of astronomers determined a
number of values for the deceleration parameter, some slightly larger

than (/ = 5 and some smaller. Because (/ = i is the special case of a flat
universe separating a bounded cosmological model from an unbounded
model, the piedicted fate of the universe swung back and forth. .According to some data, the universe seems to be just barely open and infinite,
whereas other data indicate
doomed
that the uni\erse
is
just
barcK closed and
to collapse.
Motivated h\ the fad tiiai ijo nia\ be nearU etiual to I. physicists

began looking lor special conditions associated with a flat universe in
which the aveiage (iensitv of mattei (p) is exactly equal to the critical
densitv. p, .S|)ecifically, suppose the average density of matter during the
.
Big Bang to lie slightly larger or smaller than the critical density. How
would this deviation grow or decrease as the universe evolves?
We saw that the earliest undeistandable moment in the universe was
and
the Planck time, about 10" '* sec after the Big Bang. Between I =
universe was so dense and particles were interacting so

no known theory can pro]}erl\ describe what happened.
Immediately after the Planck time, however, the imiverse became very
seiisili\e to the density of matter. Calculations demonstrate that the
slighiesi deviation Irom the precise critical density would have mush-
10
''
sec, the
\iolently that
roomed
very rapidly, multiplying
itself
e\er\
10
'
sec.
If
the density
Cosmology: The creation and jute
nj the
muvers
than p^. the universe would soon have become wide

emptv. It. on the other hand, the density were
slightly greater than p, the imiverse would soon have become tightly
closed and so packed with matter that the entire cosmos would have rapidly collapsed into a black hole. In other words, immediately after the
Big Bang, the fate of the universe hung in the balance so that the tiniest
deviation from the precise equality p = p^ would have rapidly propelled
the universe away from the special case of ^n = i-
were
slightly less
open and
virtuallv
Observations reveal that
q,) is
today approximately
i.
the density of the universe immediately after the Big
Consequently,
Bang must have
been equal to the critical density to an incredibly precise degree. Calculations demonstrate that, in order for c/n to be roughly 5 today, p must
have been equal to p, to more than 50 decimal places!
What could have happened immediatelv after the Planck time to ensure that p = Pj to such an astounding degree of accinacv? Because p =
pc means that space is flat, this enigma is called the flatness problem.
A second enigma closely related to the flatness problem is the isotropy
of the 3-K cosmic microwave background. We saw that the microwave
background is so incredibly uniform across the sky that sensitive temperature measurements can reveal our motion through this radiation field.
Subtracting the effects of our motion, we find that the temperature of
the microwave backgroimd is the same in all parts of the sky to an accuracy of 1 part in 10,000.
To appreciate this isotropy dilemma, think about microwave radiation
coming at us from two opposite parts of the sky. This radiation left over
from the primordial fireball has been traveling toward us for nearly 20
billion years. The total distance between opposite sides of the observable
imiverse is roughly 40 billion light years, so these widely separated regions have absolutely no connection with each other. Why then do these
unrelated parts of the imiverse have the same temperature?
In the early 1980s, Alan Guth, working at Stanford University, offered a remarkable solution to the problems of the fiatness of the universe and the isotropy of the microwave background. Guth analyzed the
suggestion that the universe had experienced a brief period of extremely
rapid expansion shortly after the Planck time. During this inflationary
epoch, as the period is called, the universe ballooned outward in all directions. In a tiny fraction of a second, the universe
expanded by
a fac-
perhaps 10'^" times. This infiation placed much of the material

that was originally near our location far beyond the edge of the observable universe today. Now the observable universe is expanding iiilo space
containing matter and radiation that was once in close contact with our
tor of
location.
Infiation accounts for the fiatness of the universe. To see why, think
about a small portion of the Earth's surface, such as your backyard. For
all practical purposes, your backyard looks very fiat and it is impossible
to detect the Earth's curvature over such a small area. Similarly, the observable universe is such a tiny fraction of the inflated universe that any
overall curvature is virtually undetectable. Like your backyard, our observable segment of space looks very fiat.
The inflationary epoch also accounts for the isotropy of the microwave background. When we look at microwaves that are from opposite
parts of the sky, we are seeing radiation from regions of the universe
that were originally in intimate contact with each other. That is why they
have the same temperature.
Finally, it is important to note that the concept of infiadon does not
violate Einstein's dictum that nothing can travel faster than the speed of
Cosmology: The oealion and fate of the univers
light. Remember that the expansion of the universe is the expansion of

space and does not involve the motion of objects through space. Inflation was a brief moineni during which the distances between particles
suddenly increased by an enormous amount. This expansion was accomplished entirely by a sudden vigorous expansion of space. I'articles did
not move: space inflated.
Grand unified theories

explain that
four forces
strength
all
had the same

immediately after
the Big Bang
of the behavior and interactions of everything in the universe can be

imderstood as the result of onh loin- phvsical forces: gravitv, electromagnetism, and the strong and weak nuclear forces. Toda\ these four
forces differ sharplv from cacii otlui. Howexer. certain ideas at the
forefront of physics called grand unified theories predict that all four
forces had the same strength immediatelv after the Big Bang.
We are all intimatelv familiar with the force of gravitv as a long-range
force that dominates the universe over astronomical distances. Electromagnetism is also a long-range force (in principle, its influence extends
to infinity, as gravity's does), but it is much stronger than the gravitational force. Just as the force of gravity holds the Moon in orbit about
the Earth, the electromagnetic force holds electrons in orbit about the
nuclei in atoms. We do not generally observe the long-distance effects of
the electromagnetic force, however, because in most cases there is a negative electric charge for every positive charge and a south magnetic pole
for every north magnetic pole. 0\er large volumes of space, the net effects of electromagnetism effectively cancel each other. .\ similar canceling does not occur with gravity because there is no "negative mass."
Both the strong and the weak nuclear forces are said to be short
range because their influence extends only over distances less than about
10^'' cm. The strong nuclear force holds protons and neutrons
together inside the nuclei of atoms. Without this force, nuclei would disintegrate because of the electromagnetic repulsion of the positiveK
charged protons. Thus, the strong nuclear force overpowers the electromagnetic force inside nuclei. The weak nuclear force is so weak that it
does not hold anvthing together. Instead, it is at work in certain kinds of
radioactive decay, such as the transfoi ination of a neutron into a proton.
Numerous experiments in nuclear physics stronglv suggest that protons and neutrons are composed of more basic particles called quarks,
the most common varieties being "up" c]uarks and "down" quarks. .A.
proton is composed of two up Cjuarks and one down c]uark, whereas a
neutron is made of two down quarks and one up quark.
In the 197()s, the concept of quarks gave rise to a more lundamenial
description of the strong and weak nuclear forces. In its most basic
.\11
form, the stiong nuclear force is the force that holds quai ks together.
weak nuclear force is at work whcnexer a quark clianges
from one \arict\ to another. For example, when a neutron decavs into a
proton, one of the neutron's down quarks changes into an up (juark.
.Similarly, the
/\ highly successful description of forces, called quantum field theory,

explains that two particles exert a force on each other h\ exdi.mging a
third panicle. The exchanged particle is the carrier of the force, lor instance, the gravitational force occins when particles exchange gravitons,
and the weak nuclear force
exists
when
carrier of the electromagnetic force
is
gether by exchanging gluons. Thus we

as indicated in
particles
exchange weakons. Ihe

and quarks stick to-
the photon,
mav sunnnari/e
die lour forces
Table 19-2.
To examine details of the Ioki-s. sdentisis use p.iriide accelerators

ihai luul high-speed elec lions and ])roi()iis ai laigeis. In such experi-
Cusmulugy: The creation and j ale
TABLE
iij
the universe
The four forces
19-2
Force
Strong
Relative
Particles
Particles
strength
exchanged
acted
Gluons
Quarks
upon
Range
10-'^
Example
cm
Holds nuclei
together
Electromagnetic
Charged
Photons
1/137
Infinite
Weak
Weakons
1/10,000
Holds atoms
together
particles
<10"'^ cm
Quarks,
Radioactive decay
electrons,
neutrinos
Gravity-
6 X 10"^^
Evervthing
Gravitons
Infinite
Holds the
solar system
together
merits, physicists find that the different forces begin to look the
same
as
the particles' speeds approach the speed of light. In fact, during experiat the CERN accelerator in Europe in the 1980s, particles were
slammed together with such violence that the electromagnetic force and
they were "unified."
the weak nuclear force had equal strength
ments
In recent years, physicists have labored to produce a comprehensive
theory that would completely describe these forces. Some of the most
promising attempts, such as the grand unified theory, predict that all
four forces would have the same strength if only we could slam particles
together with energies trillions of times greater than that in the CERN
accelerator. Physicists have no hope of building accelerators that powerful. However, the universe was so hot and particles were moving with
such high speeds immediately after the Big Bang that all four forces
were unified into a single "superforce." The earliest moments of the universe is thus a laboratory wherein scientists can explore some of the most
elegant ideas in physics.
Many of the ideas connecting particle physics with cosmology are very
new and still quite speculative. Nevertheless, it is possible to summarize
these ideas with the aid of Figure 19-10. Dining the Planck time (from
'
Planck time; gravity freezes out
-Strong force freezes out; inflation begins
-Weak and
electromagnetic forces freeze out
-Confinement {of quarks
Figure 19-10
universe
As
Universe transparent
to neutrinos
The early history of the

the universe cooled, the
Synthesis of
primordial helium
Jour
forces "froze out" of their initial unified slate.
The
to
Universe
from 10^^^ sec

Big Bang. Neutrons
inflationary epoch lasted
W~~''
sec after the
transparent
o photons
and protons
"froze out" of the hot "quark

soup" one millionth of a second after the Big
Bang. The universe became tran.iparent to
light (that h,
photons "froze out") when the
universe was a million years
from David Schramm)
old.
(Adapted
"sec
10
Time
sec
after
2 sec
Big Bang
C.Dsmuluf^: The nealiun anil
fall' nf Ihe u)in>ers,
lo I =
sec), particles collided with such high energies that
four forces were unified. By the end of the Planck time, however, the
energy of particles in the universe had declined to the extent that gravity
was no longer unified with the other three forces. We can therefore say
that, at I = 10"^* sec, gravity "froze out" of the otherwise unified hot
soup that filled all space. The temperature of the universe was 10*- K
when gravity emerged as a separate force.
Bv / = 10
sec, the temperature of the universe had fallen to
10-' K and the energy of particles had declined to the
extent that the
strong nuclear force was no longer unified with the electromagnetic and
weak nuclear forces. Thus, at / = 10^
sec, the strong nuclear force
made its appearance, freezing out of an otherwise unified hot soup. Calculations suggest that the inflationary epoch lasted from I = 10" '' sec
to about t = 10"-^ sec, during which time the universe increased its size
I
10"^-^
all
'
''*
by a factor of between
been.
At
10"'- sec,
to 10'' K, there
10-'"
when
was a
and
10*" over
what
it
would normally have
the temperature of the universe had dropped

out" as the electromagnetic force
nuclear force. From that moment on, all four
final "freeze
separated from the weak

forces interacted with particles essentially as thev do todav.
The next significant event occmred at / = 10"'' sec, when the
10" K. Prior to this moment, particles collided so violently that indixidual protons and neutrons could not exist
universe's temperature was
because they were constantlv being fragmented into quarks. .After this
moment, appropriately called confinement, quarks could finaliv stick
together to form individual protons and neutrons.
Among the remaining significant events are the fact that all the primordial helium was produced by / = 3 min. Finally, the universe became
transparent to photons at / = 1 million years. These photons are today
observed as microwaves of the 3-K background, mere ghostly relics of
the dazzling splendor of the fust few momcnls of the inii\erse.
Summary
The
universe began as an infinitely dense cosmic singularitv that exin the event called the Big Bang, an explosion of
space at the beginning of time.
panded explosively
The
observable universe extends about 20 billion light vears in everv difrom the Earth. We cannot sec objects bevond the cosmic particle
horizon at the distance of 20 billion light \ears because light from these
objects has not had enough time to reach us.
rection
During the Planck
titne (which lasted until about 10 '' sec aftei- the Big
Bang), the universe was so den.se that known laws of physics do not properly describe the behavior of space, time, and matter.
The 3-K cosmic microwave background

remnant of the very hot universe
radiation
that existed
is
about
the greatly redshifted

1
million years after
the Big Bang.

If the average deiisilv of matter in the universe is greater than the criticai
densitv p, then space is splierical (with positive curvature), the (feccleration parameter q has a value greater dian i. and the universe is closed
,
(bounded) and ultimately

.
will (oll.ipsr.
average densitv of matter in the universe is less than p, then space

is hyperbolic (with negative curvatine). (j has a value less than i. and the
universe is open (unbounded) and will continue to expand forever.
If the
(Cosmology:
The
creation
and
[alf of the
If the average density of matter in the universe is exactly equal to p^, then
space is flat (with zero curvature), qn is exactly equal to j, the universe is
marginally bounded, and expansion will just barely continue forever.
nearly Hat and that the 3-K microwave is almost peras the result of a brief period of very
rapid expansion (the inflationary epoch).
That the universe

fectly isotropic
is
may be explained
During the inflationary period, much of the material originally near our
moved far beyond the limits of our observable universe. The observable universe thus is today expanding into space, containing matter
and radiation that was in close contact with our matter and radiation durlocation
ing the
instant after the Big Bang.
first
Four basic forces gravity, electromagnetism. the strong nuclear force,

and the weak nuclear force explain all of the interactions observed in
the universe.
grand unified theory is an attempt to explain two or more forces in

terms of a single consistent set of physical laws. Such theories suggest that
all four forces were identical just after the Big Bang.
At the end of the Planck time, gravity "froze out"

force.
force.
weak
Review questions
what
Why
to
become
a distinctive
A
A
short time later, the strong nuclear force became a distinctive

final "freeze out" separated the electromagnetic force from the
nuclear force.
is
meant by the expansion
of the universe?
incorrect to think of the redshifts of remote galaxies

sars as being a result of the Doppler effect?
2
is it
3 How does the expansion of the universe offer a resolution

paradox?
and qua-
to Olbers's
4 In what ways are the fate of the universe, the geometry of the universe,
and the average density of the universe related?
5 How does modern cosmology preclude the possibility of a "center" or an
"edge" to the universe?
6 Suppose that the universe
come of
will
expand
forever.
What
will
eventually be-
the microwave background radiation?
7 Describe an
universe.
example of each of the four
basic interactions in the physical
8 What is the observational evidence for (a) the Big Bang, (b) the inflationary epoch, and (c) the confinement of quarks?
Advanced questions
9 Prior to the discovery of the 3-K cosmic microwave background, it seemed

we might be living in a "steady state" universe whose overall
properties do not change with time. The steady-state model, like the Big
Bang model, assiunes an expanding imiverse, but it does not assimie a "creation event." Instead, in the steady-state theory matter is assumed to be created continuously everywhere in space to ensure that the average density of
the universe remains constant. Explain why the cosmic microwave background is a major blow to the steady-state theory.
possible that
10 With a diagram show how you would expect the observed number of galDiscuss
depend on the distance from Earth for various values of
some of the problems of using such a diagram to determine ^h.
axies to
f/i,.
Cnswitlo^:
Tlif (iralioii
ami
fall' nj Ifif itnn<erse
11 Discuss iIr' tlieological im])lications ol the idea that

to
we cannot use
science
us what existed before the Big Bang.
tell
12 Do you think that there can be "other universes." regions of space and
lime that are not connected to our universe? Should astronomers be concerned with such possibilities? Why or why not?
For further reading
Barrow,
and
J.,
Silk, J.
the Universe.
Davies,
P.
The Left
Hand ojCredlum:
Oriiiin
and
F.volution of
Basic Books. 1983.
Simon & Schuster, 1984.

"The Future ol the L'niverse.
Superforee.
Dicus, D., et
al.
"
Scientific
American. Mar.
1983.
Guth.
.A.,
and Steinhardt.
can.
May
P.
"The
Inflationai v Universe.
'
Scienlijic
Ameri-
1984.
Harrison, E. "The Paradox of the Dark Night Sks." Mercun-, lulv/.Aug.

1980, p. 83.
Kaufmann, W. Black Holes and Warfjed Sfxtcetime. VV. H. Freeman and
Company,
1979.
Overbye, D. "The Universe According
to (iiith." Discover.
Jime 1983.
p. 92.
Shu,
F'.
"The Expanding Universe and the Large-Scale Geomelrv

Space-Time." Mercun. Nov. /Dec. 1983,
Silk, J., et al.
"The Large-Scale Structure of the
of
162.
p.
L'ni\erse." Scientific
American, Oct. 1983.
Wagoner,
R.,
n/JT),
and Goldsmith, D. "Quarks, Le]Dtons. and Bosons." MerJuly/Aug. 1983, p. 98.
Webster, A. "The Cosmic Background Radiation.
Aug. 1974.
'
Scientific
American,
Appendixes
/
The planets: Orbital data

Sidereal period
Semimajor axis
(tropical
km)
Mean
Synodic
period
orbital
Inclination
speed
of orbit to
(days)
(km/sec)
ececentricity
Mercury
0.3871
57.9
0.2408
87.97
115.88
47.9
0.206
7?00
Venus
0.7233
108.2
0.0615
224.70
583.96
35.0
0.007
3.39
Earth
1.0000
149.6
1.0000
365.26
29.8
0.017
0.0
Mars
1.5237
227.9
1.8809
686.98
779.87
24.1
0.093
(Ceres)
2.7671
414
4.603
466.6
17.9
0.077
Jupiter
5.2028
778
11.86
399
13.1
0.048
1.31
Saturn
9.588
1427
29.46
378
9.6
0.056
2.49
Uranus
19.191
2871
84.07
370
6.8
0.046
0.77
Neptune
30.061
4497
164.82
367
5.4
0.010
1.77
Pluto
39.529
5913
248.6
367
4.7
0.248
17.15
Planet
The
(AU)
(10*
terrestrial worlds
years)
This montage of pho-
tographs taken by various spacecraft shows the

terrestrial planets
same
scale.
and
six large
(Prepared for
Stephen P. Meszaros)
moons
NASA
by
at the
(days)
ecliptic
1.85
10.6
Appendixes
The pla nets: Physi cal data
Diameter
Planet
(km)
Mercury
(Earth
Mass
1)
(Earth
Mean
Rotation
Inclination
.Surface
Brightest
Escape
density
period
of equator
gravity
visual
velocity
to orbit
(Earth =
magnitude
-1.9
(km/sec)
(g/cm')
1)
(days)
Albedo
1)
4.878
0.38
0.055
5.43
58.6
0?0
0.38
0.106
N'enus
12,104
0.95
0.82
5.24
-243.0
177.4
0.91
0.65
Earth
12.756
1.00
1.00
5.52
0.997
23.4
1.00
0.37
Mars
6.794
0.53
0.107
3.9
1.026
25.2
0.38
0.15
-2.0
317.8
1.3
0.41
3.1
2.53
0.52
-2.7
60
-4.4
4.3
10.4
11.2
5.0
Jupiter
142.796
Saturn
120.000
9.41
94.3
0.7
0.43
26.7
1.07
0.47
+0.7
36
Uranus
50.800
3.98
14.6
1.3
-0.65
97.9
0.92
0.50
+ 5.5
21
Neptune
50,450
3.81
17.2
1.5
0.77
29
1.18
0.5
+ 7.8
24
3,400
0.27
6.387
90
0.03
0.5
Pluto
11.2
0.0023
0.5
(?)
of the planets
Satellites
Mean
distance
Sidereal
from planet
period
Orbiul
satellite*
Approximate
magnitude at
(km)
(days)
eccentricity
(km)
opposition
Diameter of
Planet
Satellite
Discovered by
Earth
Moon
Mars
Phobos
Diemos
.\lmalthea
Europa
Barnard (1892)
GaHleo (1610)
GaHleo (1610)
Ganymede
Galileo (1610)
Callisto
Galileo (1610)
Himalia
Perrine (1904)
11.470.000
250.57
0.158
(170)
14
18
Jupiter
lo
\'l>Tf.:
A. Hall (1877)
9.380
0.319
0.021
25
+ 12
A. Hall (1877)
23,500
1.262
0.003
13
13
181.300
0.498
0.003
13
421.600
670,900
1.769
0.000
3.551
0.000
1.070.000
7.155
0.002
1,880,000
16.689
0.008
240
3640
3130
5270
4840
6
J
11,8011,000
259.65
0.207
(40)
Nicholson (1938)
11.850.000
263.55
0.130
(10)
19
Leda
Aranke
Kowal (1974)
11.110.000
239.2
0.147
(8)
20
Nicholson (1951)
Nicholson (1938)
21.200.000
631.1
0.169
(10)
18
22.600.000
692.5
0.207
(15)
19
738.9
0.378
(25)
17
758
0.275
(15)
18
Melotte (1908)
23.500.000
Sinope
Nicholson (1914)
23,700,000
Mimas
Herschel (1789)
185,500
0.942
Enceladus
Tethvs
Herschel (1789)
237,900
294,700
1.370
0.020
0.004
1.888
0.000
377,400
526,700
2.737
0.002
4.518
0.001
1.222,000
15.945
0.029
1.481.000
2 .277
79.331
Cassini (1684)
Cassini (1684)
Cassini (1672)
Hvperioii
Huvgens (1655)
Bond (1848)
itaii
3.560.000
13
0.104
390
500
1050
1120
1530
5120
310
0.028
1410
11
0.163
40
16
17
I'.IO
l.->
12
10
10
10
8
14
lapetus
Gassini (1671)
Phoebe
Pickering (1898)
Miranda
Kuiper (1948)
129,900
1.414
Ariel
L.a.ssell
(1851)
2.520
0.003
480
1160
4.144
0.004
8.706
0.002
Kilo
14
13.463
0.001
1550
14
6000
13
12.930.000
Titania
Herschel (1787)
190.900
266.000
436,300
Oberon
Herschel (1787)
583,400
Umbriel
Pluto
3476
Perrine (1905)
Neptune
0.055
Elara
Dionc
Rhea
Lranus
27.322
Lvsithea
Carnie
Pasiphae
Saturn
-12.5
384.404
Lassell (1851)
550.45
Triton
Lassell (1846)
353,400
5.877
0.000
Nereid
Kuiper (1949)
5.560,000
359.881
0.749
Clharon
Christy (1978)
17.000
6.387
This table does not include several very small
*A diameter of a
salellile
given in parentheses
is
satellites
14
(500)
19
(1200)
17
aboul Jupiter, Saturn, and I'ranm that were disfovered by Voyagers
estimated from the
amount of sunlight
it
reflects.
and
2.
15.1
Appendixes
The nearest
stars
Parallax
Distance
Spectral
velocity
Proper
motion
visual
(arc sec)
(ly)
type
(km/sec)
(arc sec/yr)
magnitude
Radial
Name
G2 V
Sun
a Cen A
0.750
4.3
-26.7
-22
G2 V
3.68
KO V
B
0.772
4.2
M5e
0.552
5.9
M5 V
Wolf 359
0.431
7.6
M8e
Lalande 21185
0.402
8.1
Luyten 726-8A
0.1387
8.4
B(UV
Luminosity
(Sun = 1.0)
1.0
1.6
0.45
11.0
0.00006
10.30
9.5
0.00045
13
4.84
13.5
0.00002
M2 V
-84
4.78
7.5
M6e
+ 30
3.35
12.5
-108
M6e
Ceti)
-0.01
1.3
Barnard's star
Apparent
13.0
0.0055
0.00006
0.00004
-8
1.32
-1.5
M5e
-4
0.74
10.6
0.00048
10.3
M6e
-81
1.82
12.3
0.00011
0.303
10.7
K2 V
16
0.97
3.7
Luyten 789-6
0.302
10.8
M7e
-60
3.27
12.2
0.00014
Ross 128
0.301
10.8
M5
-13
1.40
11.1
0.00036
0.292
11.2
K5 V
-64
5.22
5.2
0.083
6.0
0.040
0.13
0.377
8.6
Ross 154
0.345
9.4
Ross 248
0.314
Eri
Sirius
wd
61
Cyg A
Ind
Procyon
0.291
11.2
K5 V
0.287
11.4
F5 IV-V
2 2398 A
0.284
11.5
Groombridge 34
0.282
11.6
9352
rCeti
1668
L725-32 (YZ
Lacaille
Ceti)
8760
Kapteyn's star
Kruger 60
4.7
-3
1.25
0.4
M3.5
10.7
2.29
A
B
Ml V
2.91
M6 V
4.67
M4 V
BD +
-40
wd
Lacaille
8.7
K7 V
B
e
Al V
0.279
11.7
M2 V
0.273
11.9
G8 V
0.266
12.2
M5
0.262
12.4
0.260
23.5
0.003
0.30
7.65
0.00055
8.9
0.0028
9.7
0.0013
8.1
0.0058
11.0
0.00040
6.87
7.4
0.013
-16
1.92
3.5
0.45
+ 26
3.73
9.8
0.0015
M5e
1.31
11.6
0.0002
12.5
Ml
3.46
6.7
0.028
0.256
12.7
MO V
+245
8.79
8.8
0.0040
0.254
12.8
M4
-26
0.87
9.8
0.0017
M5e
11.3
0.00044
361
Al>j)endixes
The
brightest stars
Apparent
Star
CMa A
a Car
a Boo
a Cen
visual
Spectral
Absolute
Distance
velocity
Proper
motion
Name
magnitude
type
magnitude
(ly)
(km/sec)
(arc sec/yr)
Sirius
-1.46
Al V
Canopus
-0.72
FO l-ll
-3.1
98
Arcturus
-0.06
-0.3
36
Rigil
Kentaurus
Radial
K2
III
1.42
0.01
G2 V
+4.39
8.7
-8
1.324
+21
0.025
-5
2.284
4.3
-25
3.676
a Lyr
Vega
0.04
AO V
+0.5
26.5
-14
0.345
a Aur
Capella
0.05
G8
III(?)
-0.6
45
+ 30
0.435
P Ori A
Rigel
0.14
B8
la
-7.1
900
+ 21
0.00
Proc\on
0.37
F5
IV-V
+ 2.7
a Ori
Betelgeuse
0.41
M2
lab
-5.6
Eri
Achernar
0.51
B3 V
13
Cen AB
Hadar
0.63
Bl
III
a Aql
Altair
0.77
A7 IV-V
+ 2.2
a Tau A
Aldebaran
0.86
K5
III
a Vir
Spica
0.91
Bl
Antares
0.92
Ml
Fomalhaul
1.15
A3 V
+ 2.0
Pollux
1.16
KO
111
aCyg
Deneb
1.26
A2
la
/SCru
Beta Crucis
1.28
B0.5
Regulus
1.36
B7 V
a CMi
o Sco
a PsA
/3
Gem
a Leo
lb
III
-3
1.250
520
+ 21
0.028
-2.3
118
19
0.098
-5.2
490
-12
0.035
16.5
-26
0.658
-0.7
68
+ 54
0.202
-3.3
220
0.054
-5.1
520
-3
0.029
22.6
+7
0.367
1.0
35
+3
0.625
-7.1
1600
-5
0.003
-4.6
490
+20
0.049
-0.7
87
+4
0.248
11.3
Glossary
absolute magnitude The apparent magnitude that a

star would ha\e at a distance of 10 parsecs.
apparent magnitude A measure of the brightness of

light from a star or other object at Earth.
absolute zero A temperature of 273C (or

K) where
all molecular motion stops; the lowest possible tempera-
apparent solar day

transits
The interval between two successive

of the Sun's center across the meridian.
ture.
Time reckoned by
apparent solar time

absorption line A dark
continuous spectrum.
line at a specific
absorption spectrum Dark

continuous spectrum.
acceleration
accretion
lines
superimposed on
in a
the
Sun
in velocity.
asteroid (minor planet) One of tens of thousands of

small, rocky planetlike objects in orbit about the Sun.
Mars and Jupiter
The gradual accumulation of

from the action of
matter in one
gravity-
the position of
in the sky.
asteroid belt
change
location, usually
wavelength
lA-AU-wide region between the orbits of

in which most asteroids are found.
A warm,
asthenosphere
plastic layer
of the mantle be-
neath the lithosphere of the Earth.
active galactic nucleus

The center of a galaxy that is
emitting exceptionally large amounts of energy; the center of a Seyfert galaxy or a quasar.
A very luminous galaxy, often containing

an active galactic nucleus.
astronomical unit (AU) The semimajor axis of the

Earth's orbit; the average distance between the Earth
and the Sun.
active galaxy
active
Sun The Sun during
such as sunspots,
ti\itv
flares,
albedo The fraction of

or satellite reflects.
Angstrom
(A)
times of frequent solar acand associated phenomena.
The smallest particle of an element that has the

properties that characterize that element.
atom
angular diameter (angular size)

b\ the diameter of an object.
angular
momentum A measure
The
angle subtended
of the
momentum
asso-
ciated with rotation.
annihilation
cle
and
The
antiparticle
astrophysics That part of astronomy dealing with the

physics of astronomical objects and phenomena.
sinilight that a planet, asteroid,
unit of length equal to \0~'^ cm.
.\
astronomy The branch of science dealing with objects

and phenomena that lie beyond the Earth's atmosphere.
process by which the mass of a partiis converted into energy.
An eclipse of the Sun in which the

too distant to completely cover the Sun so that a
ring of sunlight is seen around the Moon at mid-eclipse.
atomic number
of an atom.
aurora
Fhe number of protons
in the
nucleus
Light radiated by atoms and ions in the Earth's

in the polar regions.
upper atmosphere, mostly
autumnal equinox
The
the celestial equator
from north
intersection of the ecliptic and

where the Sun crosses the equator
to south.
annular eclipse
Moon
is
antimatter
Matter consisting of antiparticles such as an-
tiprotons, antielectrons (positrons),
and antineutrons.
aperture The diameter of an opening; the diameter of

the primary lens or mirror of a telescope.
aphelion
thest
The
point in
its
orbit
where a planet
is
far-
from the Sun.
apogee The point in its orbit where a

Moon is farthest from the Earth.
Apollo asteroid
to the
Sun than
An
asteroid
to the Earth.
whose
satellite
or the
orbit brings
it
closer
average density
volume.
Balmer
lines
The mass of an
object divided by
Emission or absorption
lines in the
its
hydro-
gen spectrum involving electron transitions between the

second and higher energy levels.
Balmer
bar
.\
series
All of the
Balmer
lines.
unit of pressure.
barred spiral galaxy A spiral galaxy in which the spiral

arms begin from the ends of a "bar" running through
the nucleus rather than from the nucleus itself.
belt asteroid
asteroid belt.
.\n asteroid
whose
orbit lies largely in the
363
Glossani
An
Big Bang
explosion ihroughoul all space roughly 20

ago from which the universe emerged.
billion \ears
Big Crunch The gravitational collapse of the universe;

the ultimate fate of a boiuid imiverse.
Two
binary asteroid
asteroids revolving about each
Two
binary star
double star.
stars revolving
Lacertae object
blackbody
and reeniits
.A
all
about each other: a
.\n optical defect wherebv different colors of light passing through a lens are focused at
chromatic aberration
different locations.
hvpothetical perfect radiator that absorbs

radiation falling upon it.
color index
is
so strong that the
shift
in
The difference in the magnitudes of a

two separate wavelength bands.
(of a comet) The diffuse gaseous

the head of a comet.
toward longer wavelengths.
noinical units.
star
model of the atom, described by
Niels
in
component
of
comet
.A small bod\ of ice and dust in orbit about the

Sim. While passing near the Sun, a comet's vaporized
ices gi\e rise to a
.-V
which electrons revolve about the nucleus
in
measured
light.
Bode's law A numerical sequence that gives the approximate distances of the planets from the Sim in astro-
Bohr atom
visible to the
cluster of galaxies .\ collection of galaxies containing a

few to se\eral thousand member galaxies.
coma
blueshift
layer in the solar atmosphere, between the photosphere and the corona.
Clouds of Magellan Two nearby galaxies

naked eye from southern latitudes.
tvpe of active galaxy.
.A
black hole An object whose gravity

escape \elocitv exceeds the speed of
Bohr,
sili-
chromosphere The middle
other.
BL
Charge-coupled device (CCD) A type of solid-state

con water designed for the detection of photons.
"coma
"
and
"tail."
condensation temperature The temperature at which a

substance turns from gas to solid at low pressure.
circular orbits.
breccia A rock formed by the sudden amalgamation of

various rock fragments under pressure.
bursts of
A nonperiodic
X ravs.
caldera
The
Callisto
One
burster
X-rav source that emits powerful
crater at the
summit
of a \olcano.
conduction The transfer of heat bv passing energ\

recth from atom to atom.
di-
configuration (of a planet) .\ particular geometric arrangement of the Earth, a planet, and the Sun.
conic section
The curve of
intersection
between a
circu-
cone and a plane. This curve can be a circle, ellipse,

parabola, or hyperbola.
lar
of the four Galilean
The
carbon burning
satellites.
thermotiuclear fusion of carbon
conservation of angular
momentum The
nuclei.
that sa\s that the total
Cassegrain focus .An optical arrangement in a reflecting

telescope in which light ravs are reflected by a secondary
mirror to a focus behind the primary mirror.
an isolated system remains constant.
equator A great circle on the celestial sphere

90 from the celestial poles.
celestial
mechanics The branch of asironomv dealing

with the motions and gravitational interactions of objects
in the solar system.
celestial
celestial {K>les
appears
Points about which the celestial sphere
to rotate.
sphere A sphere of verv large radius centered

on the observer; the apparent sphere of the sky.
celestial
Celsius temperature scale
See temperature (Celsius).
center of mass
That point
moves at a constant velocitv
in
first
in
an isolated system that

accordance with Newton's
Cepheid variable
One
of two tvpes of vellow. supergi-
The largest asteroid and the
first
to be discov-
ered.
Chandrasekhar
dwarf.
limit
The maximimi mass of
a white
named
in
after
continental drift
The gradual movement of the continents over the surface of the Earth due to plate tectonics.
continuous spectrum .\ spectrum of light over a range

of wavelengths without anv spectral lines.
convection The transfer of energv bv moving currents
of fluid or gas containing that energv
convective zone (convective envelope) A layer in a star
where energv is transported outward bv means of convection.
corona
he Sun's outer atmosphere.
cosmological model
zation
ant pulsating slais.
Ceres
constellation .A configuration of stars often

an object, person, or animal.
cosmic singularity
law.
law of phvsics
amount of angular momentum
The Big Bang.

.A
specific theorv
about the organi-
.md evolution of the universe.
cosmological redshift An increase

bv the expansion of the universe.
cosmology
The sludv
of the universe.
of the
in
wavelength caused
organization and evolution
coude focus
reflecting telescope in which a series of
mirrors direct light to a remote focus away from the
moving
parts of the telescope.
crater A circular depression on a planet or

caused by the impact of a meteoroid.
crescent moon One of the phases of the

it appears less than half full.
dust tail (of a comet) That part of a comet's

bv dust particles escaping from its nucleus.
satellite
Moon
in
which
The average density throughout the

imiverse at which space is flat and galaxies just barely
continue receeding from each other infinitely far into
critical density
the future.
cyclonic motion Circular wind motion (counterclockwise in the Earth's northern hemisphere) in a planets
atmosphere.
dark nebula
cloud of interstellar gas and dust that
oljsciucs the light of
more
double radio source An extragalactic radio source characterized by two large regions of radio emission, often
located on either side of an active galaxy.
distant stars.
effect The generation of a magnetic field by

circulating electric charges.
gas in which all the allowed states for

particles (electrons or neutrons) have been filled, therebv
causing the gas to behave different from ordinary gases.
The
density
ratio
of the mass of an object to
its
vol-
ume.
density-wave theory An explanation of spiral arms
galaxies proposed by C. C. Lin and his colleagues.
in
deuterium An isotope of hydrogen whose nuclei each

contain one proton and one neutron; heavy hydrogen.
differential rotation The rotation of a nonrigid object
in which parts adjacent to each other do not always stav
all
the light
from one
eclipse path The track of the tip of the Moon's shadowalong the Earth's surface during a total or annular solar
eclipse.
eclipse season
or lunar eclipse
period during the year
is
possible.
when
a solar
eclipsing binary star A double star system in which the

stiU s periodically pass in front of each other, as seen
from
FLarth.
celestial
degenerate gas
cutting off of part or

by another.
celestial object
months.
deferent A stationary circle in the Ptolemaic system

along which another circle (an epicycle) moves carrving a
planet, the Sun, or Moon.
The
eclipse
ecliptic
declination Angular distance of a celestial object north

or south of the celestial equator.
caused
dynamo
daylight savings time A time one hour more advanced

than standard time, usually adopted during the summer
deceleration parameter (qo) A quantity that characterizes the rate at which the expansion of the universe is
slowing down.
tail
The apparent annual
path of the Sun on the
sphere.
electromagnetic radiation Radiation consisting of oscillating electric and magnetic fields such as gamma rays, X
rays, visible light, ultraviolet
and infrared
radiation,
radio waxes, and microwaves.
electromagnetic spectrum
electromagnetic radiation.
A comprehensive
electromagnetic theory
electricity
The entire arrav or family of
and magnetism
first
description of
formulated by
J. C.
Max-
well.
electron
foimd
negatively charged subatomic particle usually

about the nuclei of atoms.
in orbit
electron volt (eV) The energy acquired by an electron

accelerated through an electric potential of one volt.
A substance that cannot be decomposed by

chemical means into simpler substances.
element
A conic section obtained by cutting completely

through a circular cone with a plane.
ellipse
galaxy A galaxy with an elliptical shape and

no conspicuous interstellar material.
elliptical
close together.
elongation
differentiation (geological) The separation of different
kinds of material in different layers inside a planet.
diffraction grating
system of closely spaced slits or

on a piece of glass) used to pro-
reflecting strips (usually
duce
spectrum.
direct motion The gradual, apparent westward motion

of a planet as seen from Earth.
diurnal
emission line
emission nebula
distance between a planet and
bright spectral line.
glowing gaseous nebula whose
comes from fiuorescence caused by a nearby

emission spectrum
light
star.
spectrum that contains emission
lines.
energy
The
abilitv to d(j
work.
Daily.
an atom) A particular amoimt of energy possessed by an atom above the atom's least enerenergy level
diurnal motion
Doppler
The angular
the .Sun.
effect
Motion
in
one day.
The apparent change
in
wavelength of
radiation due to relative motion between the source

the observer along the line of sight.
and
(in
getic state.
epicycle
In the Ptolemaic system, a
which planets revolve.
moving
circle
about
equinox
One
celestial
equator.
of
llie intersettioiis
of the
L-dipiic
and
llic
The region of space immediately outside

the e\ ent horizon of a rotating black hole where it is
impossible to remain at rest.
ergosphere
The speed needed
by one object to
achie\e a parabolic orbit a\vay irom a second object and
thereby permanently move away from the second object.
escape velocity
One
Europa
of the Galilean
satellites.
evolutionary track On the H-R diagram, the path

lowed by a point representing an eyolying star.
The
produced
at
Beyond the
galaxy.
is
visitjle
from Earth.
globular cluster A large spherical cluster of stars usually foimd in the outlying regions of a galaxy.
.A
gluon
small, dense,
particle that
grand unified theory
converging
The
granulation
photosphere.
from an
light
distance from a lens or mirror
Ihat which can change the
light rays
where
converged
momentum
of an ob-
theory that attempts to explain
The
rice-grainlike structure of the solar
Ihe
tendency of matter to attract
number of
particle that carries the gravitational force.
The trapping of infrared radiation

near a planet's surface by the planet's atmosphere.
greenhouse effect
Greenwich meridian
The meridian of
longitude that
passes through the old Royal Greenwich Observatory

near London; the longitude of 0.
I
region
region of neutral hydrogen
in interstellar
space.
Ihe number of waves that cross a given
point per unit time; the
dark nebula.
exchanged between quarks.
greatest elongation The largest possible angle between

the Sim and an inferior planet.
ject.
frequency
graviton
light rays meet.
focus (pi., foci) The point where

by a lens or mirror meet.
is
jects.
sudden, temporary outburst of

extended region of the solar surface.
vibrations per unit
time.
II
region
region of ionized hydrogen
in interstel-
lar space.
moon A phase of the Moon during which

daylight hemisphere can be seen from Earth.
full
galactic cannibalism
its full
unequal mass and size in

be absorbed into the larger galaxy.
galactic cluster
loose association of
young
stars in the
disk of the galaxy.
Ihe intersection of the principal plane
oi the .\Iilk\
with the celestial sphere.
\\',i\
large assemblage of stars; the galaxv of which
Sun and nearby
stars are
Any one
Galilean satellite
members.
of ihe four large
moons
rays
Ihe most energetic form of
elec Ironiag-
prefix referring to the Sun.
Gentered on the Sun.
The nearly explosive ignition of heliiun

bin ning in the dense core of a red giant star.
Herbig Haro object A compact

siu loiniding a Noung star.
face temperaturi-s.
A group of stars on the Hert/diagram of a typical globular cluster near

the main sequeiue and having loughly (onstant absolute
magnitude.
oi llu' O.ililean satellites.
uiig-Riissell
general theory of relativity .'\ des< riplion of gravity

foi nuil.tled b\ Kinslein explaining that gravils .ilfecis the
Hubble constant
geometry
their distances.
ol sjjace
and
ifie
flow of time.
variable nebula possibly
horizontal branch
s])!
radiation.
Ganymede One
heliocentric
Hertzsprung-Russell (H R) diagram .A plot of the absolute magnitude or hniiinosit\ ol stars against theii surof
Jupiter.
gamma
helio-
helium flash
galactic equator
galaxy
Heisenberg uncertainty principle .A principle of quanon the precision of simultaneous measurements.
timi mechanics that places limits
between two galaxies of

which the smaller galaxy seems
collision
to
nelit
phase of the Moon in which more

of the Moon's daylight hemisphere
th<in half. i)iu not all,
gravitational radiation (gravitational waves) Ripples in

the oveiall geometry of space produced by moving ob-
See temperature (Fahrenheit).
focal length
field.
large cloud of cool gas in a
the physical forces.
process of imparting energy to an atom
Fahrenheit scale
the
moon A
gibbous
.A
fol-
magnifying lens used to view the image

the focus of a telescope.
extragalactic
force
he Karths magnetic
gravitation (gravity)
matter.
eyepiece
flare
giant molecular cloud

galaxN
globule
event horizon The location around a black hole where

the escape \elocity equals the speed of light; the surface
of a black hole.
excitation
or ion.
geomagnetic
relation
between
Ihe (onstant
of pioportionalitv in the
llu- veloiiliis ol
remote galaxies and
Glossary
The
Hubble law
shifts
relationship which says that the red-
of remote galaxies are directly proportional to
hydrostatic equilibrium A balance between the weight

of a layer in a star and the pressure that supports it.
Three statements formulated by Gustav

Kirchhoff describing spectra and spectral analysis.
Kirkwood gaps Gaps in the spacing of asteroid orbits

discovered by Daniel Kirkwood.
light
the cone.
light
optical representation of an object
on
to
it
change
state of
its
is
epoch
Bang during which
curve
visible to
is
graph that displays variations
the eye.
in the
brightness of a star or other astronomical object.
The
distance light travels in a
vacuum
in
one
\ear.
that requires a force to
limb (of Sun or Moon)
motion.
or
configuration when Mercury

between the Sun and the Earth.
inflationary
calendar year with 366 days.
Electromagnetic radiation that
light year
The
inferior conjunction
or Venus
produced
by lenses or mirrors.
The property of matter
inertia
act
light rays
leap year
hyperbola A conic section formed by cutting a circular

cone with a plane at an angle steeper than the sides of
by the focusing of
light
KirchhofPs law
hydrogen burning The thermonuclear fusion of hydrogen nuclei to produce helium.
image The
3260
parsecs; about
\cars.
from Earth.
their distances
One thousand
kiloparsec
Moon
The apparent edge
of the Sun
as seen in the sky.
limb darkening The phenomenon whereby the Sun

darker near its limb than near the center of its disk.
is
brief period shortly after the Big
limiting magnitude
The faintest magnitude that can
be observed with a given telescope under certain condi-
the scale of the universe increased
yery rapidly.
tions.
infrared radiation Electromagnetic radiation of a wavelength longer than visible light yet shorter than radio
waves.
nodes
line of
line
connecting the nodes of an
hydrogen
interplanetary medium The sparse distribution of gas

and dusl particles in interplanetary space.
gen that
interstellar dust
tialh the Earth's crust.
compounds
Local
Sparse gas
galax\
medium
in interstellar space.
Interstellar gas
and
to the
lo
ion
One
of the Galilean
An atom
to the addition
that has
or
loss
is
orbit.
metal-like
upper layer of the Earth; essen-
cluster of galaxies of which
our own
member.
Large Magellanic Cloud,
companion galaxy
Milky Way.
satellites.
become
electrically
charged due
of one or more electrons.
The process by which an atom
ionization
solid,
Group The
LMC The
dust.
The
lithosphere
in interstellar space.
interstellar gas
interstellar
Microscopic solid grains of various
form of hydrocan be prcjduced under extreme pressure.
liquid metallic
luminosity The rate at which electromagnetic radiation

is emitted from a star or other object.
luminosity class The classification of a
spectral type according to its luminosity.
loses elec-
star
of a given
trons.
lunar
ionization potential The energy required to i"emo\e an
electron from an atom.
ionosphere A layer in the Earth's upper atmosphere

which many of the atoms are ionized.
ion tail (of a comet) The relatively straight
comet produced by the solar wind acting on
in
The same
of the Moon.
hydrogen proand from the lowest en-
spectral lines of
b} electron transitions to
magnetosphere
its
magnetic
The
region around a planet occupied
field.
magnification (magnifying power) The number of

times larger in angular diameter an object appears
through a telescope than with the naked eye.
in all directions.
Jovian planet Any of the four largest planets: Jupiter,

Saturn, Uranus, and Neptune.
kelvin
An eclipse
A series of
magnetic field A region of space near a magnetized

bodv within which magnetic forces can be detected.
by
several forms for the same chemical elenuclei all have the same number of protons,
but different numbers of neutrons.

isotropic
duced
lines
of a
ions.
Any of
ment whose
Lyman
ergy state of the hydrogen atom.

tail
irregular galaxy An unsymmetrical galaxy having neither spiral arms nor an elliptical shape.
isotope
Referring to the Moon.
lunar eclipse
See temperature (Kelvin).
Kepler's laws Three statements, discovered by Johannes

Kepler, that describe the motions of the planets.
magnitude A measure of the amount of

from a star or other luminous object.
light received
main sequence A grouping of stars on the Hertzsprung-Russell diagram extending diagonally across the
graph from the hottest, brightest stars to the dimmest,
coolest stars.
major axis
an
(of
ellipse)
longest diaiiieler of an
lie
(pi.,
free jilain
maria)
neutron
Dark
.A
measiue of the
total
amount of
material in an
object.
the
The interval between successive meridmean sun; the average length of a
day
solar
ian passages of the

solar day.
mean
mean
b\ the location of the
sun.
charge
many
General Catalogue (NGC)
ters,
nebulae, and galaxies
new moon A phase of
the
first
star
composed
al-
catalogue of star cluspublished in 1888.
Moon when
it
is
nearest the
in the skv.
Newtonian
An
reflector
optical
arrangement
in a re-
tlecting telescope in which a small mirror refiects converging light ravs to a focus on one side of the telescope
tube.
The laws of mechanics and graxitation

formulated by Isaac Newton.
node The
fictitious object that
moves eastward
at a
constant speed along the celestial equator, completing

one circuit of the sky with respect to the vernal equinox
one
in
Newton's laws
Time reckoned
solar time
mean sun A
in
electric
subatomic particle with no electric charge

mass nearly equal to that oi the proton.
New
Sun
mass luminosity relation The relationship between

masses and luminosities of main sequence'stars.
mean
.'\
neutron star A very compact, dense

most entirely of neutrons.
solidified la\a that covers the lunar
maria.
mass
no
important
Latin for sea; a large relatively crater-
on the Moon.
basalt
is
of a tciiesdial ]}lanet
and with
mare
particle with
little or no mass, yet which

nuclear reactions.
mantle (of a planet) That portion

lotaled between its crust and core.
mare
A subatomic
neutrino
and
ellipse.
tropical year.
mechanics The branch of physics dealing with the behavior and motions of objects acted upon by forces.
One
megaparsec (Mpc)
million parsecs.
meridian (celestial) The great circle on the celestial

sphere that passes through an observer's zenith and the
north and south celestial poles.
intersection of an orbit with a reference

plane such as the plane of the celestial equator or the
ecliptic.
nonthermal radiation Radiation emitted by charged

particles mo\ing through a magnetic field; synchrotron
radiation.
A star that experiences a sudden outburst of radiant energy, tempoiaiily increasing its luminosity roughlv
a thousandfold.
nova
nuclear
Referring to the nucleus of an atom.
nuclear bulge
The
central region of
our galaxv.
the stratosphere.
nuclear density The density of matter

an atom; about lO'^ g/cm'.
The luminous phenomenon seen when a meteoroid enters the Earth's atmosphere; a "shooting star."
composed of protons and neutrons, about which
The
mesosphere
layer in a planet's
atmosphere above
meteor
nucleus (of an atom)
The
in the
nucleus of
massive part of an atom,

elec-
trons revolve.
meteorite A fragment of a meteoroid that has survived

passage through the Earth's atmosphere.
meteoroid
small rock in interplanetarv space.
Main meteors
meteor shower
a
common
seem
nucleus (of a galaxy)
to radiate
from
point in the sk\
microwaves
Milky
that
.Slioit
(of
The smallest diameter
of
an
molecule
.\
oi
two or nioie atoms.
momentum
object's
.\ measure of the inertia of an

mass multijilied by its velocit\.
monochromatic
galaxy
nebula
dust.
(])1..
Of one wavelength
object: an
oi color.
.\
ioiul ol
inur
stellar y.is
concentration of stars and
An association of very young, massive

predominantly of spectral tvpes O and B.
association
The principal
lens or mirror of a telescope.
oblateness A measure of how much a flattened sphere

(or spheroid) differs from a perfect sphere.
Oort cloud .V region surrounding llie solar system

where comets spend most of their time.
occultation
lie ec li])sing of an aslronomiial objeit
the Moon or a planet.
f
galaxv with a bright slailike nudcus.
nebulae)
thai
Obler's paradox .Xn argimient, liased on Newtonian

cosmology, that the night sk\ should he as blight as the
Sun's surface.
Srr asteroid.
combination
The
and dust
the center of a galaxy.
objective
galaxy.
an ellipse)
at
stars
ellipse.
minor planet
dust
OB
wavelength radio waves.
Way Galaxy Our
minor axis
A collection of ices
tonstitutc the solid part of a comet.
nucleus (of a comet)
and
l)\
A loose association of Noting stars in the

disk of the galaxv; a galactic c luster.
open cluster
Glo.ssan
opposition The configuration of a planet when it is at

an elongation of 180 and thus appears opposite the Sun
in the sky.
ond
The
path of an object that
is
moving about
a sec-
object or point.
Volcanic processes by which gases escape
outgasing
from
a planet's crust into
The second
Pallas
asteroid to be discovered.
unit of distance; 3.26 light years.
lunar or solar eclipse in which the

eclipsed object does not appear completely covered.
partial eclipse
Pauli exclusion principle A principle of quantum mechanics that says that two identical particles cannot have
the same position and momentum.
penumbra The portion of

of the
source
light
penumbral eclipse
lunar eclipse in which the
Moon
is
Moon
penumbra.
The
point in its orbit where a satellite or the

nearest the Earth.
perihelion
est the Sun.
The
point in
its
population II star A star whose spectrum exhibits comparatively few spectral lines of elements heavier than helium; a metal-poor star.
positron
orbit
where
a planet
is
near-
period The interval of time between successive repetitions of a periodic phenomenon.

periodic table A listing of the chemical elements according to their properties invented by D. Mendeleev.
period-luminosity relation A relationship between the

period and a\erage density of a pulsating star.
precession (of the equinoxes) The slow westward motion of the equinoxes along the ecliptic because of precession.
prime focus
photon
,A
point in a telescope where the objec-
primeval fireball The extremely hot gas that

universe immediately following the Big Bang.
the
filled
principle of equivalence A principle of general relativity that states that, in a small volume, it is impossible to
distinguish between the effects of gravitation and acceleration.
A wedge-shaped piece of glass used to disperse

white light into a spectrum.
prism
prominences Flamelike protrusions seen near the limb

of the Siui and extending into the solar corona.
proper motion The annual change
star on the celestial sphere.
proton
is
proto-
The appearance of the
Moon
as
orbits the Earth.
photometry
The
in the location
of a
A heavy, positively charged subatomic particle

one of two principal constituents of atomic nuclei.
small, distinbing effect.
phases (of the Moon)

it
electron with a positive rather than negacharge; an antielectron.
precession (of the Earth) A slow, conical motion of the

Earth's axis of rotation caused by the gravitational pull
of the Moon and Sun on the Earth's equatorial bulge.
that
perturbation
An
tive electric
passes onlv through the Earth's
perigee
population I star A star whose spectrum exhibits spectral lines of many elements heavier than helium; a metal-
tive focuses light.
shadow in which only part

covered by an opaque body.
is
The motions of large segments (plates)

of the Earth's surface over the underlying mantle.
rich star.
parallax The apparent displacement of an object caused

by the motion of the observer.
hot ionized gas.
plate tectonics
atmosphere.
its
parabola A conic section formed by cutting a circular

cone at an angle parallel to the sides of the cone.
parsec
plasma
optics The branch of physics dealing with the behavior

and properties of light.
orbit
planetesimals Primordial asteroidlike objects from

which the planets accreted.
The measurement
of
light intensities.
prefix referring to the embryonic stage of a

(planet, star, etc.) that
is still
process of formation.
in the
discrete unit of electromagnetic energy.
young astronomical object
pulsar A pulsating radio source believed to be associated with a rapidly rotating neutron star.
photosphere The region in the solar atmosphere from

which most of the visible light escapes into space.
pulsating variable
Planck's radiation law A relationship between the intensity of radiation emitted by a blackbody, wavelength,
and the temperature of the black body.
quantum mechanics The branch of
n(jsit\
star that pulsates in size
and
limii-
the structiue
with light.
physics dealing with

their interaction
and behavior of atoms and
strength; about 10"'' sec.
One of several hypothetical particles presumed to

be the internal constituents of certain heavy subatomic
particles such as protons and neutrons.
planetary nebula A luminous shell of gas ejected from

an old, low-mass star.
quarter moon .\ phase of the

90 from the Sun in the sky.
Planck time
The
after the Big
Bang when
brief interval of time immediately

all four forces had the same
quark
Moon when
it
is
located
Glossais
quasar
resolving power A measure of the ability of an optical

system lo distinguish or resolve fine details in the image
it produces.
siarlike object with a very large rcdshili; a
quasi-stellar object or quasi-stellar radio source.
RR
ods
Lyrae variable One

less than one da\.
class
of pulsating star with periI'lie appaient westw.ird motion of a

planet with respect to background stars.
retrograde motion
That portion of an object's velocity parthe line of sight.
radial velocity
allel to
The motion of one body
revolution
about another.
coordinate for measuring the eastwest positions of objects on the celestial sphere.
radial velocity curve A plot showing the variation ol

radial velocitN with time for a binary star or variable star.
right ascension
radiant (of a meteor shower) The point in the sky from

which meteors of a particular shower seem to originate.
Roche lobe The sinaliest distance from a planet or

other object at which a second object can be held together by purely gravitational forces.
radiation
Electiomagnetic energy; photons.
The turning of a body about an

through through the boch
rotation
radiation pressure The transfer of momentum carried

bv radiation to an object on which the radiation falls.
radiative diffusion The movement of photons through
a gas liom a hot location to a cooler one.
region inside a star where energy

transported outward by radiative diffusion.
radiative zone
axis passing
rotation curve (of a galaxy) A graph showing how the

orbital speed of material in a galaxy depends on the dis-
tance from the galaxy's center.

satellite
is
.\
boch that revolves about a large one.
Schwarzschild radius
The
distance from the singularity
radio astronomy That branch of astronomy dealing

with observations at radio wavelengths.
to the event horizon in a nonrotating black hole.
radio galaxy A galaxy that emits an unusually large

amount of radio waves.
planet, usually associated with earthquakelike phe-
radio telescope
waves.
\'ibrations traveling
through a
terrestrial
nomena.
A device used to record and measure

mic waves, such as those produced by earthquakes.
seismograph
telescope designed to detect radio
Long wavelength electromagnetic
radio wave
seismic waves
radiation.
seismology
The
seis-
study of earthquakes and related phe-
nomena.
The
process whereby certain atomic nuclei

natinalh decompose by spontaneously emitting particles.
radioactivity
recurrent nova
semimajor axis
Seyfert galaxy
erupted more than
n()\a that has
.A
Half of the major axis
whose spectrum exhibits emission

red giant
.\
starlight as
it
The shifting to longer wavelengths of the light

from remote galaxies and quasars; the Doppler shilt ol
light from a receding source.
redshift
satellite
optical
component
reflection
is
ol light rays
com|)onenl
is
telescope in which the
sidereal month The period of the Moon's rcvoluiion

about the Karih with respect to the stars.
The orbital period of one object about

sidereal period
anolher with respect to the stars.
a lens.
sidereal time
refraction
The bending of light rays passing from one
trans|)arcnt medium to anothei
regolith The rocky material of the
planet like the Moon or Mars.
speed
particle
sm
moving
face of a lifeless
at
ncarlv the
The degree to which

image can be distinguished.
fine details in .m opli-
lime reckoned
In the location ol the ver-
nal equinox.
sidereal year
The orbital period of the Kailh .ibout the
.Sun with respect to the stars.
SMC
lo ihe
of light.
resolution
cal
meteor.
by a surface.
refracting telescope (reflector)
relativistic particle
.Sec
The interval between successive meridian

sidereal day
passages of the vernal equinox.
comparatively dense cloud of dust

intcrsicllaT space that is illuminated by a star.
princ ipal optical
lines.
.A
telescope in which the principal
a concave mirror.
Fhe return
reflection nebula
in
ellipse.
shock wave .-Xn abrupt, localized region of compressed

gas caused by an object traveling through the gas at a
speed greater than the speed of sound.
shooting star
reflecting telescope
an
small satellite whose gravity is responsible for maintaining a sharply defined ring of matter around a planet such as Saturn or Uranus.
shepherd
large, cool star of high luminosity.
reddening (interstellar) The reddening of

passes through (he interstellar medium.
of
spiral galaxy with a bright nucleus
I he Small Magellanic Cloud, a companion galaxy

Milky Way.
solar activity
jjliere.
Phenomena
that occur in the solar almos-
such as sunspots, plages,
flares,
and so
forth.
Glossai-f
solar cycle
magnetic
22-year cycle during
field reverses
its
wliicli ilic
Sun's
polarity.
The average amount
solar constant
of energy received
from the Sun per square centimeter per second, just
above the Earth's atmosphere.
Stefan-Boltzmann law
Stefan's law
solar nebula
Sun and
violent eruption of the Sun's surface.
perature,
fhe cloud of gas and dust from hich the

formed.
stellar
solar system
relationship between the tem-
Stefan-Boltzmann
Sir
The changes
stellar evolution
solar flare
perature of a blackhody and the rate at which

energy.
and
radiates
it
law.
in size, luminosity,
tem-
so forth, that occur as a star ages.
model The result of theoretical calculations that

give details of physical conditions inside a star.
solar seismology The study of the Sun's interior from

observations of vibrations of its surface.
stratosphere
solar system The Sun, planets, their satellites, asteroids,

comets and related objects that orbit the Sun.
strong nuclear force The force that binds protons and

neutrons together in nuclei.
solar wind A radial How of particles (mostly electrons

and protons) from the Sun.
subduction zone
Either of two points along tiie ecliptic at which

the Sun reaches its maximum distance north or south of
the celestial equator.
solstice
A description of mechanics
and electromagnetic theory formulated by Einstein that
explains that measurements of distance, time, and mass
special theory of relativity
are affected by the observer's motion.
The
spectral analysis
identification of chemicals by the
appearance of their spectra.
spectral class
classification
of stars according to the
The photograph of
layer in the
atmosphere
of a planet di-
location
where
colliding tectonic
plates cause the Earth's crust to he pulled
down
into the
mantle.
subdwarf
sequence
stars
of lower luminosity than mainof the same spectral type.
star
subgiant A star whose luminosity is between that of

main sequence stars and normal giants of the same spectral t\pe.
summer
Sun
is
Sun
solstice
he point on the
ecliptic
where the
farthest north of the celestial equator.
The
about which the Earth and other planets
star
\c\o\\e.
appearance of their spectra.
spectrogram
above the troposphere.
rectly
a spectrum.
sunspot
A temporary
cool region in the solar photo-
sphere.
spectrograph
device for photographing a spectrum.
spectroscope
device for directly viewing a spectrum.
spectroscopic binary star A double star whose binarv

nature can be deduced from the periodic Doppler shifting of lines in its spectriun.
spectroscopy
The
speed
spicule
sphere.
The
rate at
narrow
study of spectra.
which an object moves.

jet
of rising gas in the solar chromo-
sunspot cycle The semiregular 11-year period with

which the number of simspots fiuctuates.
supergiant
slar of very
superior conjunction
being behind the Sun.
trons.
spiral
arms Lanes of interstellar gas, dust, and young

wind outward in a plane from the central re-
stars that
gions of a galaxy.
high luminosity.
configuration of a planet
supermassive black hole A black hole whose mass exceeds a thousand .solar masses.
supernova
outburst during which a star sudbrightness roughly a millionfold.
stellar
denly increases
spin A small amount of angular momentum possessed
by certain particles such as electrons, protons, and neu-
The
its
surface gravity The weight of a unit mass at the surface of an object such as a planet.
synchrotron radiation The radiation emitted by

charged particles moving through a magnetic field; nonthermal radiation.
A flattened, rotating galaxy with pinwheel-like spiral arms winding outward from the galaxy's
nucleus.
synodic month
standard candle An object whose known luminosity can

be used to deduce the distance to a galaxy.
synodic period The interval between successive occurrences of the same configuration of a planet.
standard time Local mean solar time at a standard meridian adopted for convenience at surrounding geo-
graphical areas.
spiral galaxy
star
self-luminous sphere of gas.
witli respect to
fhe period of revolution of the Moon

the Sun; the length of one c\cle of lunar
phases.
Tauri stars Young variable stars associated with intermatter that show erratic changes in luminosity.
stellar
Tauri wind
star.
.\
flow of particles
away from
Tauri
(of a comet) Gas and dust particles from a toinet"s

nucleus that have been swept awav from the comet's
head bv the radiation pressure of sunlight and ihe solar
wind.
tail
telescope
An
instrument for viewing remote objects.
Earth's
The
passage of a celestial body across the meridpassage of a small object in front of a larger ob-
1 he pressuie and temperature

substance can simultaneously exist as a solid,
triple point
which
liijuid,
and
gas.
tropical year
troposphere
turbulence
Ju-
The period of revolution of the Earth

about the Sun with respect to the vernal equinox.
a particular star does,
in
at
Trojan asteroid One of several asteroids that share

piter's orbit about the Sun.
temperature (excitation) The temperature of a star determined from the strengths of various spectral lines that
originate in atoms with different stages of excitation.
The
lowest level in the Earth's atmosphere.
Random motions
in a gas
or liquid.
UBV
system .A system of stellar magnitudes involving

measurements of starlight intensit) in the ultraviolet,
Absolute temperature measured
temperature (Kelvin)
is
ject.
temperature (effective) The temperature of a blackbocU

that would ladiate the same total amount of energy that
Temperature measured on
at 32 and boils at 212.
during which the Sun
umbra.
ian; the
temperature (color) The temperature of a star determined bv comparing the intensitv of starlight in two
wavelength bands.
temperature (Fahrenheit)
scale where water freezes
solar eclipse
completely hidden by the Moon, or a lunar eclipse during which the Moon is complelelv immersed in the
transit
temperature (Celsius) Temperature tneasured on a

scale where water freezes at 0 and boils at 100.
total eclipse
blue,
and
visible spectral regions.
centigrade degrees.
Electromagnetic radiation of wavelengths shorter than those of visible light but longer than
those of X rays.
ultraviolet radiation
temperature (kinetic) Temperature directly related to

the average speed of atoms or molecules in a substance.
terminator
face of the
The line dividing day and night on

Moon or a planet; the line of sunset
the suror sun-
umbra
The
central, completely
dark portion
of a
shadow.
rise.
universal time
terrae
Cratered regions of the Moon; lunar highlands.
Any of the planets Mercury, Venus,

Earth, or Mars, and sometime also the Galilean satellites
terrestrial planet
and
universe
Pluto.
Local
mean time
at
the prime meridian.
universal constant of gravitation The constant of proportionality in Newton's law of gravitation.

All space
along with
all
the matter and radia-
tion in space.
thermal energy The energy associated with the motions

of atoms or molecules in a substance.
the Earth
thermal equilibrium
outflow of heal
in a
A balance between the input and

system.
Two doughnui-shaped regions around

where many charged particles (protons and
'Van Allen belts
electrons) are trapped by the earth's magnetic field.
variable star
thermal radiation The radiation naturallv emitted bv

any object not at absolute zero.
thermonuclear reaction
from a
moving rapidU
in a
3-K cosmic microwave background
I'holons
varies.
VLBI Very long baseline interferometr\ a meihotl of

connecting widely separated radio telescopes to make
observations of very high resolution.
wavelength
thermosphere A region in the Earth's atmosphere

tween the mesosphere and the exosphere.
whose luminosity
The
visual binary star A double star in which the two components can be resolved through a telescope.
ol
reaction resulting
high-speed collision ol nuclear particles

because they are at a high temperature.
star
point on the ecliptic where the Sun

crosses the celestial equator from south to north.
thermodynamics The branch of physics dealing with

heat and the transfer of heat between bodies.
thermonuclear fusion A reaction in which the nuclei
atoms aie fused logethei at a high temperature.
vernal equinox
Ihe distance between two successive peaks
wa\e.
be-
from e\er\
white dwarf A low-mass star that has exhausted all

thermoiuiclear fuel and contracted to a size roughly
equal to the size of the Earth.
.i
hlackbodN sj)ecirum at nearlv 3 K;
the cooled-off radiation from the primordial fin-ball that
Widmanstatten patterns
filled all space.
(citain types of meteorites.
its
part of the sk\ with
tidal force
.\
gravitational forie wliose strength and/or
dircdion varies over

the bod\.
hofU and thus tends to deform
(Crystalline structure
seen in
Wien's law A relationship between the temperature of

biaikbodv and the wavekiigth al which it emits the
greatest intensity of radiation.
Gloisan
winter solstice The point on the ecliptic where the

Sim reaches its greatest distance south of the celestial
zenith The point on the celestial sphere op]5()sitc to the

direction of gravity.
equator.
zero-age main sequence (ZAMS) The main sequence of

voinig stars that have just begun to burn hydrogen at
Electromagnetic radiation whose wavelength

between that of ultraviolet light and gamma rays.
ray
year
Sun.
The
Zeeman
lines
is
period of revolution of the Earth about the
effect
splitting or
because of a magnetic
broadening of spectral
field.
their cores.
zodiac A band of twelve constellations around the sky

centered on the ecliptic.
Answers
Chapter
Chapter 7
10'
3.
a.
7.
8 minutes
8.
8700
b.
1.7
d.
= W/r. mantle = 699c:

About 220 million years ago
10'
Coil-
crust
= 1%
km
Chapter 2
2.
4.
5.
he equator
south pole; 232; December 21
The
Due
Chapter 11
east
11.
10 Mt,; 10
'-^
Lq
Chapter 3
AU in 1984; 26 square AU in five years

Average distance = 100 AU: farthest distance = 200 AU
13. Yes; sidereal period = synodic period = 2 years; Mars
(sidereal [leriod = 1.88 yr; synodic period = 2.14 vr)
5.2 square
11.
12.
Chapter 12
11.
During 5
billion vears. 2
percent of the Sun's hvdrogen
converted into helium
Chapter 5
7.
7S
8.
2500 A; 8 times brighter than the
9.
The Sun
is
.Sun
8 times brighter
Chapter 16
10.
One supernova
about evcrv 75,000 vears
Chapter 6
5.
hc\ are
llie
sunplesl
sl.iblc Lhcirjital
hvdrogen (the most abundaiu

abundant elements (excluding
9.
The planet's spectral lines
caused l)v the planet's motion
cunibinations ol
element) with the next three most
helium which
i.s
inert)
should exhibit a Dopplcr shilt

along its orbit; telluric lines arc-
Chapter 17
1500 Mly
not shifted
9.
The planet would be made of ice; density = 1 g/cm'; the

planet's diameter is 1.77 times larger than the Earth's diameter
5000 km/sec (for these answers

50 km/sec/Mpc)
10.
10.
fl
15 kn)/.scc/Mly
is
Illustration credits
1
p. 1: Photography by D. F.
Mahii of the Anglo-Ausirahan Observatory trom original negatives by the U.K.
1.2-m Schmidt telescope, copyright
1980 Royal Observatory, Edinburgh;
Fig. 1-2: Courtesy of Scientific American
Books, NASA, and USNO; Figs. 1-3. 1-4,
Chapter
NASA;
1-5, 1-11:
Fig. 1-7:
USNO;
Fig.
Fig. 6-14: Courtesy of the Univerof Arizona and |PL.

Chapter 7 p. 100, Figs. 7-1, 7-4. 7-9,
Board;
sity
7-11, 7-12, 7-19, 7-20, 7-21. 7-22, 7-23,

7-24, 7-25, 7-29, 7-30, 7-31, 7-32:
Fig. 7-3:
New Mexico
Courtesy of S. M.
7-16: Courtesy of M. Tharp
Larson; Fig.
1977; Fig. 7-26:

Meszaros, NASA; Fig.
Lick Observatory; Fig. 1-9: Copyright R. J. Dufour, Rice University;
and Bruce Heezen
Palomar Observatory.
Chapter 2 p. 1 1 Copyright Anglo-
7-27:
1-8:
Fig. 1-10:
Australian Telescope Board; Fig. 2-1:

British crown copyright, reproduced with
permission of the Controller of the Brittanic Majesty's Stationary Office; Fig. 2-2:
(left) Drawing from Elijah Burritt's Alias,
Janus Publications, Wichita Omnisphere
Earth-Space Center, 220 S. Main, Wichita, Kansas, (right) Courtesy of R. C.
Mitchell, Central Washington University;
Fig. 2-4:
NASA;
USNO;
Figs. 2-11, 2-22, 2-24:'
Fig. 2-13: Lick
2-21: Courtesy of
Chapter 3
p. 30:
Observatory; Fig.
M. Harms.
NASA;
Fig. 3-8: Palo-
mar Oliservatorv; Figs. 3-10, 3-11: Yerkes

Obseivatory; Fig. 3-13: Copyright 1986
3-14: Lick
J. B. Marling, Lumicon; Fig.
Obser\atorv.
Chapter 4
p. 47:
NOAO;
Fig. 4-8;
Yerkes Observatory; Fig. 4-12: NOAO;

Fig. 4-14: Courtesy of The MultipleMirror Telescope Observatory; Fig. 4-15:
California Institute of Technology; Figs.
4-16, 4-17: Courtesv of NRAO/AUl; Figs.
NASA;
NRL,
4-18, 4-19, 4-21, 4-22, 4-23, 4-24:

Fig. 4-20: (a)
G. R. Carruthers,
JPL, NASA; (c) Mount Wilson and

Las Campanas Observatories; Fig. 4-25:
(a,c) E. C. Krupp. Griffith Observatory,
(b) JPL, NASA, (d) based on Uhuru data.
Chapter 5 p. 65: Courtesy of Bausch
and l.oiiib; Figs. 5-5, 5-10: Mount Wilson
and Las C:ampanas Observatories.
Chapter 6 p. 79: Photogiaphv by D. F.
Maliii of the Anglo-Australian Observatory from original negatives by the U.K.
1980 Royal Observatory, Edinburgh:
(b)
Figs. 6-3, 6-4, 6-5:
NASA;
Fig. 6-6:
Mount Wilson and Las Campanas Observatories; Fig. 6-8:
Palomar Observatory;
Fig. 6-12: Lick Observatory; Fig. 6-13:
Copyright
Anglo-Australian Telescope
NASA;
State University
Observatory; Fig. 7-8:
Courtesy of
S. P.
USGS;
USNO:
Fig. 7-28:
TASS;
Fig. 7-35;
Arecibo Observatory.
Chapter 8 p. 1 3
Courtesy of S. P.
Meszaros, NASA; Fig. 8-1: Copyright
The University of Texas McDonald
Observatory; Figs. 8-2. 8-3. 8-6. 8-7. 8-8,
Fig. 7-36:
8-11. 8-12, 8-14, 8-15, 8-16, 8-17, 8-18.

8-19. 8-20, 8-21. 8-23, 8-24, 8-26, 8-33.
8-34. 8-35: NASA; Fig. 8-13: Lowell Ob-
New Mexico
servatory; Figs. 8-29, 8-30:
State University Observatory.
Chapter 9
p. 157, Figs. 9-4, 9-8. 9-9,

9-11. 9-12, 9-13, 9-14, 9-15, 9-16, 9-18,
9-19, 9-20, 9-21, 9-23, 9-24, 9-25, 9-26,
9-27:
NASA;
tory; Fig. 9-2:
Fig. 9-1:
Yerkes ObservaState Univer-
New Mexico
sity; Fig. 9-3: (left)
NASA,
(right) Lick
Observatory; Fig. 9-7: Lick Observatory;

Fig. 9-17: S. P. Meszaros, NASA; Fig.
9-29: Yerkes and McDonald Observatories; Fig. 9-30: Courtesy of J. Christy and
R. Harrington.
Chapter 10 p. 183, Fig. 10-15: Copyright Max Planck Institut fiir
Aeronomie, Lindau/Harz; Figs. 10-1,
10-5: Yerkes Observatory; Fig. 10-6: Meteor Crater Enterprises, Arizona; Fig.
10-7: Courtesy of W. Alvarez; Figs. 10-8,
CourRonald A. Oriti; Fig. 10-17: Johns

Hopkins University. Naval Research Laboratory; Figs. 10-18, 10-19: Lick Observatory; Fig. 10-21: Palomar Observatory;
Fig. 10-22: New Mexico State University
10-9, 10-10, 10-11, 10-12. 10-13:
tesy of
Observatory; Fig. 10-23: TASS, Sovfoto.

Chapter 11 p. 202: Copyright .^ngloAustralian Telescope Board; Fig. 11-7:
Palomar Observatory; Fig. 11-12; Yerkes
Observatory; Fig. 11-17: Lick Observatory.
p. 221, Fig. 12-18: NASA;
Fig. 12-1: Celestron International; Fig.
Chapter 12
12-4:
San Fernando Observatory;
12-6, 12-14, 12-16:
NOAO;
Figs.
Fig. 12-8:
575
IlluitraUon credits
C. Keller, Los
tory; Fig. 12-9:
Alamos
NASA
Scientific
Labora-
and the High
Alti-
tude Observatory: Fig. 12-10: N.-\S.-\ and

Harvard College Obseryatory; Fig. 12-11:
Project Siratoscope. Princeton L'niyersity;
Fig. 12-12: Mount Wilson and Las Campanas Observatories; Fig. 12-15: Naval
Research Laboratory.
Chapter 13 p. 239. Figs. 13-2. 13-5.
13-ti.
13-10. 13-11. 13-19: Copyright
Anglo-.Australian Telescope Board; Fig.

13-1: Photography by D. F. Malin of the
Anglo-Australian Observatory from original negatives by the U.K. 1.2-m Schmidt
telescope, copyright 1980 Royal Observatory. Edinburgh: Fig. 13-4: (a)
IRAS
14-3; Copyright Anglo-Australian Telescope Board; Fig. 14-5: Courtesy of R. B.

Minton; Fig. 14-7: Mount Wilson and Las
tory, copyright
Campanas
panas Observatories: Figs. 17-6, 17-15:

NOAO; Fig. 17-8: Yerkes Observatory;
Observatories; Figs. 14-8, 14-
Palomar Obseryatory, copyright

California Institute of Technology; Fig.
14-9: Photography by D. F. Malin of the
Anglo-.Australian Observatory from original negatives by the U.K. 1.2-m Schmidt
telescope, copyright 1980 Royal Observatory, Edinburgh; Fig. 14-10: (a) S. S.
Murray, Harvard Smithsonian Center for
12:
Astrophysics, (b) NRAO/AUI {VLA observations by .A. .Angerhofer, R. .A. Perlev, B. Balick, D. K. Milne); Fig. 14-15:
NASA: Fig. 14-19: NR.AO/AU1 (VLA
photograph produced by NLR/ROG,
observations by R. M. Hjellming, K.
using spline fitted survey data [IR.AS was

developed and operated by the Nether-
Johnson);
lands .Agency for .Aerospace Programs

(NIVR). N.ASA, and the U.K. Science
and Engineering Research Council

(SERC)). (b) optical photography by D.
tesy of J. Kristian,
Campanas
F.
Malin of the Anglo-.Austialian Observafrom original negatives by the L'.K.

1980 Roval Observatory, Edinburgh;
Fig. 13-8: Lick Observatory; Figs. 13-9.
13-16; USNO; Fig. 13-13: Copyright
National Geographic Society-Palomar
Sky Survey, reproduced by permission of
the California Institute of Technology.
Chapter 14 p. 260. Figs. 14-14. 14-20:
Lick Observatory photograph; Figs. 14-2,
tory
J.
Chapter 15 p. 282: Courtesy of

L. Smarr and J. Hawley; Fig. 15-5: Cour-
Mount Wilson and Las
Observatories.
Anglo.Australian Telescope Board; Fig. 16-1:

Steward Observatory, L'niyersity of .Arizona; Fig. 16-3: Harvard Observatory;
Chapter 16
Fig. 16-4:
16a:
p.
292: Copyright
NOAO;
USNO;
Figs. 16-6, 16-10, 16-
Fig. 16-9:
G. Westerhout; Figs. 16-16b, 16-I6c:
NASA;
Fig. 16-17:
Courtesy of K.
and M.J. Claussen.

Chapter 17 p. 306,
Fig. 17-2:
Figure 17-1: Courtesy of
Humphries;
Fig. 17-3;
V.
17-7, 17-21:
Lo
USNO;
Lund
Palomar Observa-
1959 California
Insti-
Mount Wilson and Las Cam-
Figs. 17-9, 17-10, 17-16, 17-19:
Palomar
Observatory; Figs. 17-12, 17-13: Copy-
R. J. Dufour, Rice University;

Courtesy of Francois SchweiCerro Tololo Inter-.American Obser-
right
Fig. 17-20:
zer,
vatory; Fig. 17-23: Lick Observatory.
Chapter 18 p. 324: Courtesy of J. F.

Hawley and L. L. Smarr; Figs. 18-1, 18-2,
18-7: Palomar Observatory; Figs. 18-3,
18-8, 18-10: NOAO; Fig. 18-5: Copyright
I960 National Geographic SocietyPalomar Sky Survey, reproduced by permission of the California Institute of
Technology: Fig. 18-9: J. E. Grindlay,
Harvard Smithsonian Center for .Astrophysics; Fig. 18-11; E.J. Schreier, Harvard-Smithsonian Center for .Astrophysics;
Figs. 18-12, 18-14:
NR.AO/AUI;
Fig.
18-15: Courtesy of T. D. Kinman.

Fig. 18-17: Courtesy of
NOAO;
T. Stephenson. H. C. Arp, J. J. Lorre;

Fig. 18-21: Courtesy of J. F. Hav\ley,
L. L. Smarr.
Chapter 19
Courtesy of
tute of Technology; Figs. 17-4, 17-5,
p. 34 1 Courtesy of Lawrence Berkeley Laboratory; Fig. 19-3:

Bell Laboratories; Fig. 19-6: Courtesy of
John Mather, NASA.
Appendixes p. 358: S. P. Meszaros,
NASA.
Index
absolute magnitude. 204-205

absolute zero, 6566
absorption line, 71-72
accretion,
Barnard, E.
Barnard
E.,
object,
240
240
barred spiral galaxy, 310-311

Barringer Crater, 189-190
94-96
275-276
alpha particle, 72
Alvarez, Luis, 190
270
horn antenna, 346
186-188
belts (on Jupiter), 132-133, 138
Bessel, Friedrich, 264
Betelgeuse, 255
Big Bang, 7, 90, 344-345, 351-355
Big Crunch, 349
Alvarez, Walter, 190-191

acids (in meteorites), 194
Andromeda galaxy, 308
Big Dipper, 2, 12
binary asteroid, 189
binary star, 212-218, 273-276,
accretion disk,
acceleration,
Bell, Jocelyn,
40
Bell Labs
330-331
Adams, John Couch, 41
active galaxy,
belt asteroid,
Allende meteorite, 194

Almagest, 32
amino
angstrom, 49
Angstrom, A.
eclipsing,
J.,
49
Anglo-Australian Observatory,
angle, 2
139-142
ring,
Aristarchus, 21-22, 27, 32

asteroid,
83-84, 182-191, 193

between, 188189
collisions
discovery of, 185

asteroid belt, 186-188
influence of Jupiter on, 187-188
asthenosphere, 113
astronomical unit, 6
Atkinson, Robert, 222
atmospheric pressure, 103
atom, 65, 72-76, 88-90

atomic number, 88
atomic structure, 72-76
aurora, 125126
autumnal equinox, 16-17
average density, 83, 347-348
Baade, Walter, 269

Babcock, Horace, 235
Balmer, Johann Jacob, 73
Balmer lines, 73-75, 207, 326
Balmer
series,
7475
bar (pressure), 103
215-216
spectrum, 214
visual,
angular diameter, 2
angular measure, 2
angular momentum, 94, 287
annular eclipse, 26
anorthosite, 165
anticyclone, 135
Aphrodite Terra, 117
aphelion, 36
apparent magnitude, 203-204
arc minute, 2
arc second, 2
287-289
216-218, 273
spectroscopic,
212
blackbody, 67-69
blackbody curves. 67-69
blackbody radiation, 67-69
black hole, 7, 44, 267, 279, 282-289
Blanford, Roger, 338
BL Lacertae object, 333-335
214
Bode, Johann
blueshift,
Elert,
184
Bode's law, 184
Bohr model
(of atom),
7375
Bohr, Niels, 73-75, 207

Bohr orbits, 74
Bok, Bart. 240
Bok globule. 240
Boltzmann. Ludwig, 66
bounded
universe. 347
Brahe, Tycho, 35, 272
165166
B ring, 139-141
brown oval (on Jupiter). 135-136
Bunsen burner. 70
breccia,
Bunsen. Robert. 70. 86

Burbidge, Geoffrey, 333
burster, 7,
278-279
caesium (discovery of), 70-71

Callisto, 85, 157, 167-169, 173-174, 176
carbonaceous chrondite, 193
carbon burning, 264-266
Cassegrain focus, 53
Cassini, G. D., 107, 138
Cassini division, 138, 140-141
Cassiopeia A, 268-269
celestial coordinates, 14-15
celestial equator, 14-15
in
H 15
celeslial pole. II.

celestial
Celsius, .'\nders.
Celsius scale,
relativistic, 347-351
coude focus. 53
Crab Nebula. 7. 270-272
65
65-66
Centaurus A. 332
Centaurus X-3. 273-275
center of mass. 213
crater (lunar), 163
ring. 139
34835
274-275
critical density,
309-310
256-257, 294. 308
central bulge (of galaxy), 295,
critical surface.
Cepheid
current sheet, 149
variable,
Ceres. 83. 184-185. 189

Cerro Tololo Inter-American Observatory,
47
Chandrasekhar limit, 264. 269, 279, 282

Chandrasekhar, S., 264
charge-coupled device (CCD). 87
Charon, 178-179
chemical differentiation. 122-123, 160
Chichen
Itza,
chromosphere, 228-229
40-41
circunistellar disk, 93
circumstellar shell, 255
vs.
312314
irregular.
313
rich vs. poor. 312

cocoon nebula, 242
color index. 206
Coma cluster. 314
(of a comet). 196

comet. 194-199
structure of, 196
Comet Brooks, 197
Comet Enke, 199
constellation,
deceleration parameter, 348-351
14-15
deferent.
31-32
De
Rei'olulwnilnis
de
Sitter.
Orbium Ceteslmm. 34
Willem. 347
diurnal motion, 14
direct motion, 31-32
Doppler, Christian, 214
Doppler
113,224-225
effect)
dust (interstellar).
dust tail. 197-198
as a
1.
293
elliptical galaxy. 3
effect. 161
ol,
101-103
of the solar system, 80-83
member
core of, 123-124

crust of, 110
interior of,
magnetic Held
micro-
of,
124
magnetosphere of. 124-125

mantle of, 123-124
surface of, 110-114
earthquake, 123
ecliptic.
(see
equations of stellar structure, 224

equinox, 1617
precession of, 18
equivalence principle, 283-284
Eratosthenes, 21-22
E ring, 142
ergosphere, 287
Eros.
188-189
Europa, 5, 85. 157. 167-169, 172-173

event horizon, 285-286
evolutionary track, 241, 243, 253, 263
expanding balloon analogy, 342343
exponent, 51
eyepiece lens, 51
Fahrenheit. Gabriel. 65
Fahrenheit scale, 65-66
first quarter moon. 1920
eclipse.
wave background)
lunar,
cosmic particle hori/on, 344

cosmic singularity. 344
cosmological rcdshift. 343
solar.
-82
15-16
22-27
21. 24-26
prediction of. 27
24-27
eclipse path,
problem. 352
51-52
25-26
53
focus (of ellipse), 35

focus (of lens), 5152
focus (of mirror), 53
force, 40, 353
Fraunhofer, Joseph von. 70. 86
Friedmann. .\Iexandre, 347
F ring. 140-141
lull
moon. 19-20
191-192
fusion crust,
316
30 1-301
galactic cannibalism. 7.
122-124
eccentricity. 8
185
31-32
epicycle,
local length (of mirror),
ring. 142
atmosphere
346-347
Planets,
focal length (of lens).
Earth. lOOff
coronal streamers. 229

cosmic background explorer (CX)BK).
Minor
epicenter, 123
llatness
binary star)
Eagle Nebula, 243
coronagraph. 229
coronal hole. 230
cosmic censorship. 286

cosmic microwave background
Doppler
shift (see
star, (see
dynamo
convection cell, 226

convective envelope. 225
convective zone. 225
Copernicus. N., 32-35, 39
core helium burning. 252
core hydrogen burning, 249
corona (of Sun). 229-230
Epiiemerides of
double
dwarf
emission line, 71-72

emission nebula, 243
Enceladus, 142-143
Encke division, 142
Encke, Johann Franz, 142
energy level diagram, 75-76
double-line spectroscopic binary, 213

double radio source, 332-333
D
momentum, 271
90-92, 240. 345
214-215, 256, 296, 326,
effect,
343, 346
Doppler
continuous spectrum. 64, 71-72

continuum. 328
convection.
dark nebula, 240
disk (of galaxy), 295

195. 197
of.
elongation, 33-34
ellipse, 35, 40-41
elliptical galaxy, 310-311
dinosaurs, extinction of, 190-191
12-13
continental drift,
abundances
Cygnus A, 325, 332

Cvgnus loop, 248
Cygnus X-1, 288-289
diffraction grating, 87
condensation temperature, 92
conduction, 224
confinement (of quarks), 355
conic section, 40-41
conjunction, 3334
conservation of angular
clement, 88-92
cvclone, 135
Dicke. Robert. 345

differential rotation (of Jupiter), 133
differential rotation (of galaxy). 298
differential rotation (of Sun), 231, 235
coma
Comet Hallev. 41. 183.

Comet Ikeya-Seki. 197
Comet Kohoutek. 196
Comet Mrkos. 198
Comet West. 198
spin of, 296

electron volt, 69
307
D..
degeneracy, 252
degenerate-electron pressure, 252, 282
degenerate-neutron pressure, 269, 282
degree (angle). 2
degree (temperature). 65-66
delta Cephei. 256
density wave. 299-301
circle,
regular
Heber
Curtis,
declination.
12
Chiron, 83-84
chromatic aberration, 51-52
cluster (of galaxies),
Eddington, Arthur, 222

Ein.stein, Albert, 39, 42-44, 68. 74. 222.
283-284, 342. 347
Einstein Observatory. 60-61, 329
electromagnetic radiation, 49, 65
electromagnetic spectrum, 49-50, 64
electromagnetic theory, 42, 48-49
electron, 73, 89
cosmology. 341-355
sphere. 1415
galactic center, 295,

galactic cluster.
245
galactic nucleus. 295.
galaxy,
301-304
7-8
galaxy collisions, 315-317

galaxy mergers, 316-317
(;alilean satellites,
Galilei, Galileo,
167-176
37-39, 47. 48, 167, 231,
293
Galileo Regio, 173
gamma
ray.
49-50. 69
Gaiuma Ray Observatory (GRO).

Gamow, George. 345
61
Ganymede,
85, 157, 167-169,
173-176
general theory of relativity, 4344. 283284. 342'

geocentric cosmology. 3 -32
1
geyser. 171
giant elliptical galaxy. 311. 332
details
of M87, 331, 335-337
giant molecular cloud,
Humanson,
246-248
globular cluster. 253. 294-295

gluon. 353-354
Gondwanaland.
Goodricke, John, 256

grand unified theory (GUT), 252254
granulation, 226-227
granule. 226-227
gravitation. 17. 30,
39-44, 283ff
gravitational deflection of light,

gravitational radiation.
graviton.
gravity
43-44
287
Herculis, 274
impact crater, 189-190
last
inferior conjunction, 3334

inflation, 351-353
Laurasia,
352
infrared radiation, 49-50, 69
Great Red Spot. 132-135

greenhouse effect, 105
G ring, 142
Gum, Colin, 267
Gum Nebula. 267-268
Infrared Astronomical Satellite (IR.AS).
57-58
Andrew, 136-137
232-233
195, 197
halo (of
H-alpha, 73
Haro, Guillermo, 244
Hawaii, 1 14
head-tail source,
HEAO,
333-334
heliocentric cosmology,
32-33
helium, 90-91
discovery of, 71
helium burning, 251-253. 266
helium
flash,
252-253
helium-shell Hashes, 261
Helix Nebula, 262

Herbig, George, 244
244-245
Hercules X-1, 272-275
Herbig-Haro
158159
interferometry, 56-57
314
International Ultraviolet Explorer (lUE),
58-59
object,
Herschel. William. 41. 49. 107, 184. 293.
307
Herschel, John, 307
Hertz, Heinrich, 40
Hertzsprung, Ejnar. 208

Hertzsprung-Russell diagram. 208-211.
213. 241, 244, 253-254, 256, 263
H-gamma, 73
highlands (lunar), 163
high-pressure system, 134-135
H II region, 243-245, 296-297, 299
Himalaya Mountains, 114
Hipparchus, 31
Horsehead Nebula, 1, 240
horizontal-branch stars. 254
hot-spot volcanism, 114
H-R diagram (see Hertzsprung-Russell
diagram)
Hubble, Edwin, 307-311, 317, 342, 351
J.,
41-42
64ff
photons, 68-69, 74
speed of, 48, 49, 50, 65
wave nature of, 48-49
as
light curve,
light year,
293
interstellar medium, 246
lo, 5, 85, 157, 167-172
ion, 75
60
51-52
light, 47ff,
interstellar extinction,
Hawley, John F., 338

H-beta, 73
HDE 226868, 288-289
1 1
law of cosmic censorship, 286

law of equal areas, 36
Leavitt, Henrietta, 293-294
"legend" of the Ephemera, 239-240
Leverrier, U.
256
intergalactic gas, 256,
Comet, 41, 183,

galaxy), 295
187-188
quarter moon, 1921
lens,
Ingersoll.
instability strip,
intercrater plains (on Mercury),
Halley's
points,
Lamaitre, Georges, 347

Large Magellanic Cloud (LMC), 249, 311-
inflationary epoch,
gravitation)
Hale, George Ellery,

Hallev, Edmund, 41
Kirkwood gaps, 187-188

Kruger 60, 212
Kuiper Airborne Observatory (KAO),
57-58
Lagrange
241
hyperbola, 40-41
HZ
kinematic wave. 300

King, Ivan, 337
Kirchhoff, Gustav, 70-71, 86
Kirchhoffs laws. 71
Kirkwood, Daniel, 187
312
44
353-354
(.see
Milton. 317
Huygens, Christian. 48
hydrogen, 90-92, 222
hydrogen atom. 73-75. 296
hydrogen burning. 222-223, 241, 266
hydrogen envelope (of a comet), 197
hydrostatic equilibrimn, 148, 223-224,
Kepler's first law, 35-36

Kepler's second law, 36
Kepler's third law, 36-37. 212-213. 217,
298, 304
287
gravitational slowing of time.

gravitational wave.
Hubble classification, 309-311

Hubble constant, 318-319, 321. 328
Hubble now. 317
Hubble law. 317-319. 325. 327-328. 342.
348-349
Hubble Space Telescope (HST). 58-60.
321. 339
216-218, 256
light-gathering power, 54
hmb
darkening, 227
Lin, C.
C, 299-300
ionization, 75
Lindblad, Bertil, 298

line of nodes, 22-23
ionopause, 126-127
liquid metallic
ion
tail.
197-198
192-193
hydrogen, 148
lithosphere,
1 13
X-3, 289
iron tneteorite.
LMC
irregular galaxy, 31 1-3 12

Ishtar Terra, 117
Local Group, 312-313

long-period comet. 197
isotope. 89
Lord Rosse, 306-307
|ansky. Karl. 55
Lovelace, Richard, 337-338

low-pressure system, 134135
jet (in galaxies and quasars). 327. 336

Jovian planet, 83-84, 131-153
formation of, 94-96, 132
Juno, 185
Jupiter, 5, ISlff
as a member of the solar system, 8084
of, 132-137, 143-145

chemical composition of, 145. 146
atmosphere
to Saturn, 143-149
excess heat from, 146
internal structure of, 147148
niagnetosphere of. 149
inoon's of (see Galilean satellites)
compared
oblateness of. 147

rotation of, 133
Kalnajs, Agris
J.,
300
Kant. Immanuel, 307

Keeler gap, 142
Keeler, James, 142
Kelvin, Lord, 65
kelvin scale,
65-66
Kepler, [ohannes. 35-37. 39, 81, 2(S9.

272, 341
luminosity, 205
luminosity class. 2 1
lunar eclipse. 21-25
lunar month. 20
Lyman-alpha. 74
Lyman series. 74
Lynden-Bell. Donald. 335
magnetic-dynamo model. 235

magnetogram. 234
niagnetopause. 124125
magnetosheath, 124215
magnetosphere. 124125
magnitude. 203-205
apparent, 203
absolute, 204
magnitude scale, 203
magnification, 51
magnifying power, 51
main sequence, 209-211

main sequence lifetimes, 250
main sequence star, 210. 241. 243-244.
249
major axis (of an ellipse). 35
mare. 162-167
formation of. 167
(lisio\ei\
mare basalt. 16,">

Mare Imbriuiu, 162
Mare Traiiquililatis. 163
marginally bound iniiverse. 348
Margon. Bruce. 276
satellites ol.
Mariner
Mars. lOOlf
as a
member
ot the solar system,
80-83
282
atmosphere of.
polar cap of, 20- 2
Nrw
New
rotation of. 107
Newton.
101-103. ]07-10'.l
178-179
telescope appearance. 150

Nereid. 178-179
neutrino. 225
neutron. 73. 89. 266
neutronization, 266
neutron star, 266. 269-276. 278-279.
158
10.
Penzias, Arno, 345-346

periodic table. 88-89
period luminosity relation, 256257.
II
(.1.
interior ol. 151
Aslronomy. 35
General Catalogue, 307
mesosphere. 102-103
metal-poor stars. 254
Isaac. 4. 37-42, 48, 52. 213.

283. 341-342
Newtonian cosmology. 341-342
Newtonian focus, 53
Newtonian mechanics, 4, 42, 44, 82
Newtonian reflector, 53
Newton's first law. 38-39
Newton's second law, 39-40
Newton's third law, 40
Newton's universal law of gravitation. 40
new moon. 1920
N galaxy. 333
no-hair theorem, 287
north celestial pole, 1415
northern lights, 125-126
nova, 177-178
Nova Herculis 1934. 278
nuclear density. 266
nucleus (of an atom). 72-73
metal-rich stars, 2.54
nucleus (of a comet). 195
surface of. 108-110. 114-122
water on. 108, 118-122

Martians, 107
mass loss (in stars). 255

mass-luminosity relation, 213
Mauna Loa, 14
maximum elongation, 3334
1
Maxwell.
J.
C.
42. 48-49. 140
Maxwell .Montcs. 117

Mercurv. 157-162
as a
member
of the solar system. 80-83
interior of. 160
magnetosphere of, 161

42-43. 284
orbit of,
rotation of. 160
surface of. 158-160
meteor, 191-192
meteor shower. 194195
OBAFGKM. 207
OB association. 243.
meteorite. 191-194
meteoritic swarm. 198
objective lens.
microwave background. 345347. 355

our motion through. 346
Mid-Atlantic Ridge. 111-112
Milky Way, 7. 29211
Miller. Joseph, 335
51-52
ellipse).
35
Olympus Mons. 114-115
asteroid)
open
luster.
photometry. 205-206
photon. 68-69, 74
photosphere. 226-229
Piaz/i. Guiseppe. 184
pixel. 87
Planck, Max, 68. 74
Planck's law, 69
Planck time, 345, 351. 354-355

planetary nebula. 260, 262-263
planetesimal, 94
planets. 80fi
average density of, 83

diameters of. 83
formation of. 9297
masses of. 83
80-82
orbits of.
plasma. 233
plate tectonics. 111
Plato.
114
37
244-245
178-179
Pleiades,
Pluto,
pole star, 14
population 1 stars. 254, 256

population II stars. 254. 256
powers-of-ten. 2-3
precessing ellipse, 42-43. 300
17-18
prime focus, 53
precession.
283-284
prism. 48. 86
Prometheus. 170
prominence. 233234
proton. 73, 89
protostar.
240-242
protosun, 94
protoplanet. 95
Ptolemy, 31-32
pulsar. 7.
269-273
245
33-34
opposition.
perturbation, 299
phases (of the Moon), 19-21
photodisintegration, 266
principle of equivalence.
observable universe. 344

oceanic trench. 133
ocean rift. 133
octahedrite. 192
Olbers. Heinrich. 185. 341
Olbers' paradox. 342. 344
Oort cloud. 198

Oort. Jan. 198. 298. 301
millibar. 103
Mimas. 142
minor axis (of an
minor planet {see
247. 296-297. 299
oblateness. 147
meteoroid. 189
micron, 68
microwave. 49-50. 69
293-294
perihelion, 36
quadrature. 33
Miranda, 153
optics. 47ff
quantum
mirror. 52-55
missing mass problem.
molecule. 90
Orion, 13
Orion Nebula. 7. 247
oxygen burning. 264-266
quark. 353354
quasar, 8, 277, 324, 326ff
quasistellar radio source (sir (|uasar)
Paloniar Obser\atorv. 54
Pallas. 185. 189
radar. 12.5-126
explcM.ilion of. 161-l(i5
formation
167
interior of, 164
Pangaea,
radial velocity curve. 215-21(i
orbit of. 17
parsec, 6
Parsons, William. 306
Moon.
85. 157. 159.
Moon
of,
of,
18-21
rocks.
month.
Mount
Mount
17,
161-166
20
Kverest,
14
.St. Helens. 118-119

Multiple-Mirror Telescope. 55
parallax,
Paschen
Pasdien
I'auli.
264-266
member
of the solar system, 80-84
radial velocity.
203-204
series,
radiative zone,
75
Wollg.uig. 252
P.
|.
E..
pcniunbra. 2421)
of siuispol. 231
peiuuubral eclipse, 24
224-225
225
radio waves, 49-50. 55. 69

radio telescope. 55-57
ray (lunar). 163
Reber. Grote, 55-56, 325
24-26
74-75
Pauli exclusion principle.
215-216
radiative diffusion.
alph.i.
345
Penrose. Roger. 286
Neplune. 41. 150-151

as a
partial eclipse.
Peebles.
nebula. 6
neon bin ning.
353
162-167
22-25
eclipses of.
phases
314-315
field theory.
252
recombination, 243
red giant, 209-21 I, 249-255. 260
redshift. 214. 317-318, 342-343
cosmological, 343
red supergiant, 261
lelleding telescope, 52-55
relletlion.
5253
Slipher, V. M., 317
reflection grating, 87
retlection nebula,
reflector,
245
52-55
refracting telescope,
refraction.
refractor,
5052
50-51
50-52
Small Magellanic Cloud (SMC), 294, 312

Smarr, Larry L., 338
sodium (energy-level diagram of). 76
solar corona, 229-230
solar flare, 125,
cosmology, 347
residual polar cap (on Mars), 121
resolving power, 54
retrograde motion, 31-32
rho Ophiuchi region, 242
Solar
14-15
233
Maximum
solar nebula,
Mission (SMM), 236
92-96
236
solar seismology,
formation
of,
92-97
229-230
124-125
163
140
Roche, Edward, 274
solar wind, 97,
Roche
south celestial pole, 11. 1415

southern lights, 125
space station. 9
riUe,
ringlets,
277
Rontgen, William, 49
lobe, 274,
Rosse, Earl of.
306-307
rotation curve (of galaxy), 298, 315
Rougoor, G. W., 301

RR Lyrae variable, 256-257, 294
rubidium (discovery of), 71
Russell, Henry Norris, 209
Rutherford. Ernest, 72
16-17
Sombrero Galaxy, 309
relativity, 43,
Sagittarius
325
West, 303-304
.Sandage, Allan, 319, 326, 349

Sargent. Wallace. 335-336
Saturn, 131, 138-149

as a member of the solar system, 80-84
atinosphere of, 143-145
compared to Jupiter, 143-148
excess heat from, 146-147
internal structure of, 147-148
magnetosphere of, 149
oblateness of, 147
138-143
satellites of, 142-143
Schmidt, Maarten, 326
Schwarzschild, Karl, 285
Schwarzschild radius, 285
Schweizer, Frangois, 317
sea-floor spreading, 111-113
seasons, 15-17
Sccchi, Angelo, 206
rings of,
magnetic splitting
arm
330
Seyfert galaxy, 330-331

Shapley, Harlow, 294, 307
Shapley-Curtis debate, 307-308, 317

shell helium burning, 261, 265
shell hydrogen burning, 250, 261, 265
sheperd satellite, 143
shock wave. 124-125, 247, 301, 338
in galaxy, 301
short-period comet, 197
Shu, Frank, 299-301
sidereal month, 20
sidereal period, 34, 36
siderophile element. 190
silicon burning, 265-266
singularity, 285-286, 344
Sirius,
264
299
202ff
binary.
212-218
240-249
205-206
colors of,
of,
distances
260ff
to,
203-204
evolution of, 239ff
205208
Stefan-Boltzmann constant, 67
Stefan-Boltzmann law. 66-67
Stefan's law (see Stefan-Boltzmami law)
stellar association, 245
stellar evolution,
240ff
stone (meteorite), 191-192

Stonehenge, 12
stony-iron meteorite, 192193
stony meteorite, 191-192
stratosphere, 102-103
strong nuclear force, 353-354
subduction zone,
summer
solstice,
1 1
16-17
Sun, 22 Iff
as a blackbody, 69
chromosphere
of,
terrae (lunar). 163

terrestrial planet, 83, lOOff
Thales, 27
Tharsis rise, 115, 118
thermal equilibrium, 224, 241
thermal pulses. 261
thermonuclear fusion, 222-223

thermosphere, 102-103
Third Cambridge Catalogue, 325
Thomson, William, 65
Titan, 84-85, 157, 176-178
Titius, Johann, 184
Toomre, Alar, 315
Toomre, Juri. 315
birth of.
death
5657
5557
reflecting. 52-55
refracting, 50-52
ultraviolet, 58-59
X-ray, 60-61
temperature, 65-66
temperature scale, 65-66
radio.
296
(of a galaxy), 246, 296,
Gustav, 319
infrared,
spiral galaxy, 297, 309-311

SS433, 276-277
standard candle, 320-32
standing shock wave, 338
stars.
335-339
264-269
gamma- ray, 60-61
228-229
296
spin-flip transition,
spiral
194, 248,
telescope, 47ff
spicule,
spectra of, 206-208

surface temperatures of,
Stefan, Josef, 66
Seyfert, Carl,
Tammann,
spectrograph. 86-87
spectroscope, 70
spectroscopic binary, 215
spectroscopy, 85
spectrum, 48, 64, 70-72, 86-88
speed of light, 48, 49, 50, 65
seismograph, 123
3537
283-
206-208
magnitudes of, 203-205

radii of, 210
axis,
222,
7,
supernova remnant, 248-249, 267-269

synchrotron radiation, 301, 333
synodic month, 20
synodic period, 34
233
of,
seeing disk, 54
seismic wave, 123
semimajor
supernova,
86-88
spectral line, 64,
spin,
233-234
superinassive black hole, 304,
spectral analysis, 70
spectral type (stellar).

Sagittarius A, 303,
spectrum of, 86
sunspots on, 231-234
Sun-grazing comet, 198
sunspot cycle. 231-232. 235
sunspot maximum, 232
sunspot minimum, 232
superior conjunction, 33-34
supergiant star, 225
supergranule, 228-229
solstice,
Space Shuttle, 4
special theory of
284
226
of,
sunspot,
solar system, 79ff
solar transient,
internal structure of,
location in galaxy, 295
223-226
photosphere of, 226-227
rotation of, 231-232
2127
solar eclipse,
regolith, 164
relativistic
composition of, 249

corona of, 229-230
energy source in, 222-223
model
solar cycle, 235
refractory element, 167
right ascension,
228-229
total eclipse.
24-26
158
transmission grating, 87
triple point, 177
Triton, 85, 157, 178-179
transit.
Trojan asteroids, 187-188

troposphere, 102
Trumpler, R. J., 293

T Tauri stars, 244-245
T Tauri wind, 97, 245
Tunguska event, 99
tuning fork diagram, 311
turn-off point, 254
twenty-one centimeter radiation, 296
I Cepheid, 256-257, 294
II Cepheid, 256-257
Type
Type
Type
Type
tail,
II tail,
197-198
197-198
UBV
filter, 205-206
Uhuru. 273. 28
ultraviolet radiation.
of a sunspot. 23
universe, 347
universal constant of gravitation. 40
unbounded
universe. 34 Iff
of.
Wilson, Robert, 345-346

winter solstice, 16-17
Wolf, Max, 185
Verne, Jules, 9
Vesta, 185, 189
Viking Mission, 107
Virgo cluster, 312-313
our motion toward, 346
volatile element, 167
volcano, 114-115, 118-119, 169-172
vortex, 136
Young, Peter, 337

Young, Thomas, 48
spectrum
344345
deceleration of, 348-351
density of matter in, 347-349
fate of, 348-349
flatness of, 351-352
geometry of, 349-35
shape of, 349-351
Uranus. 41. 150-153
member
of the solar system. 80-84
discovery of, 4
interior structure of, 151
rings of, 153
rotation of, 153
seasons on, 151-152
telescopic
56
Very Large Array (VLA), 57
vernal equinox, 1415
rotation of.
344
creation of,
as a
surface of. 103, 117-120, 125-126

surface temperature on, 105106
very-long-bascline-interferointery (VLB I),
335
Venus, 37. lOOff
as a member of the solar system. 80-83
atmosphere of, 101-106, 119
cloud layers of, 105-106
phases of, 37-38, 104
velocity dispersion,
(i9
umbra. 24-26
age
waxing gibbous moon, 1920

weak nuclear force, 353355
weakon, 353-354
Wegener, Alfred, 1 1
Wells, H. G., 9
Whipple, Fred L., 195
white dwarf, 209-210, 263-264, 272-279
white ovals (on Jupiter), 135136
Widmanstatten, Count Alois von, 192
Widmanstatten patterns, 193
Wien's law, 66-67, 69, 205
Wien, Wilhelm, 67
Vela pulsar, 272
49-50.
appearance
ot,
of.
125-126
90
Wasserburg, Gerald
Valles Marineris.
Van Allen
1 16
radiation belts.
Veil Nebula,
267-268
124-125
ray,
49-50, 69
zap crater, 166
Zeeman,
Zeeman
150
Valhalla Basin. 176
194
waning crescent moon, 1921

waning gibbous moon, 1921
wavelength, 49-50. 65
waxing crescent moon. 1920
Pieter,
effect,
233
233
zenith, 17
zero-age main sequence (ZAMS). 253

zodiac, 17, 31
zones (on Jupiter), 132-133, 138

Zwickv. Fritz. 269
NOZIHOH MH3HXHON
/\>
'v:^
>
H
W
w
2:
K
O
s
N
o
,*^.-
?!*
ORION
=^RIGEL
5;^
LEPUS
COLUMBA
CANOPUS
SOUTHERN HORIZON
THE NIGHT SKY
IN
WINTER
Latitude of chart is 34N, but it is

practical throughout the continental
United
Chart time
States.
(local standard time):
vertically
and
Mid-December.. 11
Mid-January
9
it
so the direction you
facing shows at the bottom.
are
Mid-February.... 7
To
use:
turn
Hold chart
Star Chart from
GRIFFITH OBSERVER,
Griffith Observatory,
Los Angeles
pm
pm
pm
NOZIHOH NH3HXHON
SOUTHERN HORIZON
THE NIGHT SKY

United States.
To
Hold chart
vertically
and
direction you
are
use:
turn
it
so
the
Star Chart from
GRIFFITH OBSERVER,
IN SPRING
Chart time
(local standard lime):
Mid-March
Mid- April
Mid-May
Griffith Observatory, Los Angeles
11pm
9
7
pm
pm
NOZIHOH NHSHXHON
HVXS HIHON.
SIHVlOd
>
H
W
o
N
O
K
a:
o
2
N
o
2
Sf OKl'll
SOUTHERN HORIZON
THE NIGHT SKY

United States.
IN
SUMMER
Chart time
(daylight saving time):
Mid-June
To
use:
Hold chart
vertically
and
turn it so the direction you are

Star Chart from
GRIFFITH OBSERVER,
Mid-July
Mid- August
Griffith Observatory,
Los Angeles
Midnight
10
pm
pm
NOZIHOH NHSHXHON
>
H
W
no
z
o
N
as
a:
o
2
N
o
H
(/)
2;
SAGITTARIUS^
FOMALHAUT "^
./* GRUS
SOUTHERN HORIZON
THE NIGHT SKY

United States.
IN
AUTUMN
Chart time
(local standard lime):
vertically
and
Mid-September
Mid-October
turn it so the direction you

are
Mid-November
To
use:
Hold chart
Star Chart from
GRIFFITH OBSERVER,
Griffith Observatory, Los Angeles
11
9
7
pm
pm
pm

Discovering The Universe

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Discovering The Universe

Uploaded by

Copyright:

Available Formats

Discovering the universe

Digitized by the Internet Archive

DISCOVERING THE UNIVERSE

W. H. Freeman and Company

a science teacher extraordinaire!

The northern hemisphere of Uranus

25, 1986, as the spacecraft left the planet behind

This view of Uranus

set forth on a Sf-year cruise to Neptune. Voyager

wide-angle picture was taken

million kilometers (about

The pale blue-green color

atmosphere That gas absorbs red wavelengths

of the planet results

600,000 miles) from Uranus when

leaving the predominant hue seen here. The white at the

caused by a high-altitude haze Voyager 2

Book design: Valerie

and Sylvia Woodard

Cover design: Lisa Douglis

Library of Congress Cataloging-in-Publication Data

Discovering the universe.

ISBN 0-7167-1784-C (pbk)

Illustration credits are listed

book may be reproduced by any mechanical, photographic, or electronic process,

phonographic recording, nor may

United States of America

a retrieval system, transmitted, or otherwise copied for public or

private use, without the written permission of the publisher

Astronomy and the universe

Discovering the heavens

Gravitation and the motions of the planets

Our solar system

The Jovian planets

Quasars and active galaxies

Cosmology: The creation and fate

popularity of astronomy as a general science course is due in large

today: the creation of the universe, the formation of the Earth

about everyone. In today's increasingly technological \v(jrld, we face a

the Unn'erse offers a

move outward from

the planets to the stars

galaxies, older issues are

supplanted by newer questions and observations, including those from

A discussion of the formation of the solar system in Chapter 6

care that students find this book a pleasure to read.

dynamic, compelling, and relevant, and no book about it should

University deserves special recognition for his

It is a pleasure to here announce the availal)ility ol

Manual prepared by Thomas H. Robertson,

contains chapter commentaries, quizzes, and suggestions

for lecture topics

The Computerized Test Bank prepared by Allan T. Clark, The University

For more information and to request copies of these supplements please

with us on the preparation of this edition:

University of Central Florida

Yong Hak Kim

I was fortunate to have received many constructive comments and useful

Broward Community College

Pennsvlvania State Universitv

Mt. San Antonio College

Astronomical Society of the Pacific

In spite of the care exercised in preparing this text, errors