Cosmology 101

Recent Titles in the

Science 101 Series
Evolution 101
Janice Moore and Randy Moore
Biotechnology 101
Brian Robert Shmaefsky
Cosmology 101
Kristine M. Larsen
Science 101
GREENWOOD PRESS
Westport, Connecticut
r
London
Library of Congress Cataloging-in-Publication Data
Larsen, Kristine M., 1963
Cosmology 101 / Kristine M. Larsen.
p. cm.(Science 101, ISSN 19313950)
Includes bibliographical references and index.
ISBN 9780313337314 (alk. paper)
1. CosmologyPopular works. I. Title.
QB982.L37 2007
523.1dc22 2007000425
British Library Cataloguing in Publication Data is available.
Copyright
C
2007 by Kristine M. Larsen
All rights reserved. No portion of this book may be
reproduced, by any process or technique, without the
express written consent of the publisher.
Library of Congress Catalog Card Number: 2007000425
ISBN-13: 9780313337314
ISBN-10: 0313337314
ISSN: 19313950
First published in 2007
Greenwood Press, 88 Post Road West, Westport, CT 06881
An imprint of Greenwood Publishing Group, Inc.
www.greenwood.com
Printed in the United States of America
The paper used in this book complies with the
Permanent Paper Standard issued by the National
Information Standards Organization (Z39.481984).
10 9 8 7 6 5 4 3 2 1
Copyright Acknowledgments
The author and publisher gratefully acknowledge permission for use of the
following material:
All gures courtesy of Bridgette Alsbury.
In grateful recognition of all those who have been my teachers.
Contents
Series Foreword xi
Preface xiii
Introduction xv
Where Did It All Come From? xv
What Is Science? xvii
Science and Religion xix
1. Luminous Matters: Light and Stars 1
Cosmology as a Unique Practice 1
The Electromagnetic Spectrum 3
Magnitudes 8
Telescopes 9
Spectroscopy 14
The Classication of Stars 18
The HR Diagram 21
The Interstellar Medium 23
The Lives of Stars 24
Star Formation 24
Stellar Evolution 27
Cepheids and the Period-Luminosity Relationship 31
Stellar Corpses 34
2. The History of Cosmology 37
The Rise of the Geocentric Universe 37
The Copernican Revolution 41
Keplers Laws of Planetary Motion 43
Galileo, Newton, and the Telescopic Revolution 45
Why Is the Night Sky Dark? 48
viii Contents
The Problem of the Nebulae 49
The Great Debate 53
The Development of Hubbles Law 55
3. Warped Space-time: General Relativity and Modern
Cosmology 59
General Relativity 59
Cosmological Models 62
The Big Bang and Steady-State Theories 68
Lem aitre and the Primeval Atom 68
Gamows Evolutionary Model and the Steady-State 69
The Rise of the Big Bang 70
4. Measuring and Mapping the Universe 75
Normal Galaxies 75
The Milky Way 79
The Local Group 82
Active Galactic Nuclei 84
Dark Matter 88
Large-Scale Structure 91
The Cosmic Distance Ladder 93
Standard Candles 97
The Age of the Universe 101
Measuring the Hubble Constant 104
5. The Evolution of the Universe 109
The Standard Model of Particle Physics 109
The Fundamental Forces and Particles 109
Unifying the Forces 112
The First Fifteen Minutes 116
The Dark Age 118
The Fate of the Universe 121
Inationary Cosmology 122
Problems with the Big Bang 122
Old Ination 124
New Ination 126
Eternal Ination 127
Dark Matter Redux 129
Dark Energy 131
The Accelerating Universe 131
Candidates for Dark Energy 132
The Concordance Model 134
Contents ix
Acoustic Waves and the Early Universe 134
Results from the Wilkinson Microwave Anisotropy
Probe (WMAP) 136
Other Observations 139
6. The Road Goes Ever On: Continuing Speculations 141
Multiple Dimensions 141
String Theory 141
M-theory and Braneworlds 143
Cosmic Strings 146
The Fine-Tuned Universe 147
The Anthropic Principle 148
Multiverses 150
The Search for Extraterrestrial Intelligence 151
Life as We Know It (LAWKI) 152
Talking to Aliens 154
Conclusion 157
Glossary 161
Selected Bibliography 173
Index 183
Series Foreword
What should you knowabout science? Because science is so central to life
in the twenty-rst century, science educators believe that it is essential
that everyone understand the basic foundations of the most vital and far-
reaching scientic disciplines. Science 101 helps you reach that goal
this series provides readers of all abilities with an accessible summary
of the ideas, people, and impacts of major elds of scientic research.
The volumes in the series provide readerswhether students new to
science or just interested members of the lay publicwith the essentials
of a science using a minimum of jargon and mathematics. In each
volume, more complicated ideas build upon simpler ones, and concepts
are discussed in short, concise segments that make them more easily
understood. In addition, each volume provides an easy-to-use glossary
and an annotated bibliography of the most useful and accessible print
and electronic resources that are currently available.
Preface
The author of this volume has been charged with a seemingly impossible
taskexplaining the entire life and make-up of the universe in around
200 pages. Although I have tried my best to live up to the spirit of
this assignment, clearly a few details have had to have been left out
in the name of brevity. Although these few beginning sentences have
been typed with tongue rmly implanted in cheek, the truth is not far
away. Cosmologists have really taken upon themselves the study of all
aspects of this immense and ancient universe, aided by ever-improving
technology, a seemingly limitless source of imagination and ingenuity,
and the scientic method. Perhaps those are the truly basic themes of
this text. You, the reader, will be my companion on an exhilarating,
and admittedly sometimes confusing, journey through space and time.
Together we will discover how the human mind, limited by the size of
its enclosing skull and its location on the surface of the Earth at this era
in time, have sought the answers to the most ancient and fundamental
questionswhere did the universe come from and where is it going?
To study cosmology in isolation from its history is certainly to ignore
the scientic method, and we certainly shall not make that mistake here.
We must also never forget the important concept that science is done by
scientists, some of whom are more colorful in personality than others.
As members of the human race, individual scientists can also make
mistakes, can sometimes turn a blind eye toward results that contradict
their own research, and can make disparaging comments about the
research of others that might not be considered scholarly. Welcome
to the true world of science, warts and all. Some of what you will read
here is highly speculative, and a (hopefully very) few items of discussion
may be out of date by the time you hold this volume in your hands. If so,
xiv Preface
consider it yet another valuable lesson in the scientic method, courtesy
of this well-meaning and conscientious teacher.
While I have largely circumvented the mathematical nuts and bolts
of cosmology in the telling of this tale, it is impossible to avoid numbers
altogether. In our voyage through space and time we will necessarily
encounter numbers both inconceivably small and astoundingly large. It
is for this reason that you will soon be introduced to the mathematical
shorthand scientists use to manage numbers of extreme size, known as
scientic notation. Otherwise, this book would waste far too many pages
devoted to strings of zeros!
This project would not have been possible without the valuable input
of Ronald Mallett and Emma Keigh, and would have been far less in-
teresting without Bridgette (Brie) Alsburys diagrams. You all have my
eternal gratitude for all your help.
Now lets start exploring!
Introduction
WHERE DID IT ALL COME FROM?
Midday, March 29, 2006, the Libya-Egypt border. A cheer slowly erupted as
the shadow of the moon raced across the landscape from west to east,
enveloping the gathered crowd of 14,000 men, women, and children.
With one last, brilliant gasp of lightthe diamond ring effectthe sun
was swallowed by the dark disk of the new moon, plunging the now-
applauding witnesses into an eerie, 360-degree twilight. Astronomers
(both professional and amateur), public servants and military men of all
ranks, politicians from several nations (including Hosni Mubarek, the
president of Egypt), and general devotees of exotic adventures stood
shoulder to shoulder as the heavens graced us with a show unlike any
othera total solar eclipse. For that all-too-brief instant intime, divisions
of race, religion, language, social class, and political afliation were ut-
terly irrelevant and forgotten, as we stood in awe on that isolated desert
plateau, equal citizens of a vast and fascinating cosmos. Afar-too-eeting
four minutes later, the sun reappeared with an equally evanescent dia-
mond ring, signally the end of totality. Well, Rahu gave back the sun,
I kidded with my companions, making reference to the ancient Hindu
myth for the cause of solar eclipses. After what I had just witnessed, it
was more obvious to me than ever why our ancestors across the globe
revered the heavens, worshipping the stars, planets, sun, and moon as
gods and goddesses, and generally sought explanations for the forces of
nature in a creative and colorful pantheon of demons and deities.
This universal need to explainthe world and its workings forms the ba-
sis of the uniquely human activity called myth. These traditional, ancient
stories were crafted by cultures according to their own interactions with
the environment, and sought to explain the beginnings of the world,
xvi Introduction
the natural patterns of birth and death, and outlined the culturally ac-
ceptable behavior of that particular group of people. These stories have
been passed down largely through oral traditions from generation to
generation, which explains the normal variations seen in most myths.
Among the most fundamental myths are those that seek to explain the
creation of the earth and heavens, a cultures creation myth. As Barbara
Sproul (1979, 1) explains, The most profound human questions are
the ones that give rise to creation myths: Who are we? Why are we here?
What is the purpose of our lives and our deaths? How should we under-
stand our place in the world, in time and space? Understandably, many
religions include a creation myth among their traditions.
Although there is a rich variety of creation myths from around the
world, there are certain common threads or archetypes which are seen
repeatedly. The central theme is usually the creation of the universe and
its various parts. Some models have a one-time creation, while others
explain the universe as being a giant cycle of creation and destruction.
Some envision the universe as being created fromnothingness, or chaos.
The creator can be seen as a single god or goddess, or a group of beings.
The process of this creationcanbe a celestial war, the deathof a sacricial
god, the separation of earth from sky, or a special song, dance, or series
of spoken words (such as the famous Let there be light of the Old
Testament).
For example, in a Chinese myth, a great egg hatched after the passing
of numerous centuries, bringing forth from his lengthy sleep the giant
creature Pan Gu. The top of the shell became the heavens, while the
bottomof the shell became the earth. He kept the two separate using his
ever-growing body. After the heavens became xed into place, Pan Gu
died. His breathgave rise to the wind, his voice gave birthto thunder, and
his eyes became the sun and moon. According to Cherokee legend, the
world was originally covered with water. Through the action of Lock-
chew, an industrious crawsh, mud was brought up from below the
waters to make land, which was in turn dried by the apping of the
buzzard Yah-tees wings. The human race sprang from drops of blood
from T-cho, the Sun goddess. Even ctional cultures can have a creation
myth, such as the elves in The Silmarillion, J.R.R. Tolkiens sweeping
epic prequel to The Lord of the Rings. After tackling the formation of the
universe, and most importantly Earth itself, most myths then explain the
origin of the sun, moon, and stars, and nally human beings and other
citizens of the world. Some myths continue further, offering a vision
of the end of the world, and the ultimate triumph over evil and/or
the death of the gods, such as the RagnarokDoom of the Godsof
Norse myth.
Introduction xvii
Thanks to meticulous observations of the heavens, both with the un-
aided eye and with increasingly powerful technology (including the
invention of the telescope, photography, and spectroscopy), coupled
with mathematical techniques and advances in our understanding of
the basic forces of nature (such as gravity), a scientic understanding
of the universe, called cosmology, has developed over the past two mil-
lennia. Derived from the Greek kosmos (order), cosmology is the
scientic study of the history and structure of the universe. Sometimes
considered a special branch of astronomy, it draws from myriad other
areas of study, including mathematics, physics, and chemistry. Although
cosmology traces its origin to those ancient and poetic explanations of-
fered in creation myths, modern cosmology is indisputably a discipline
of science, and follows the scientic method in exploring different possible
models of the history of the universe.
What Is Science?
Science is a methodology used to explore the natural world and is
based on the twin pillars of observation and theory. Countless observa-
tions are made of the world around us, both passively collecting infor-
mation from natural occurrences, as well as actively performing specic
experiments in the laboratory which probe nature at a deeper level.
Based on those observations, scientists develop hypothesesproposed ex-
planations which can then be tested and further rened (or refuted).
As a hypothesis is continually tested, it will become accepted by more
scientists if it is found to successfully explain observations and experi-
mental outcomes, and becomes a theory. Unfortunately, in the common
vernacular, the word theory has become synonymous with a guess or
a hunch, while a scientic theory is actually neither.
Regardless of specic discipline, scientic theories have several com-
mon attributes: they are consistent, both internally and with already ob-
served phenomenon; they are falsiable (a concept widely explored by
philosopher Karl Popper); they make predictions that can be observed
with current or future technology; they tend to be relatively simple or
elegant (although it might seem the opposite to the casual observer);
nally, a theory can never be proven absolutely true, therefore there
is no end to scientic endeavor. A true scientic theory is always open
to being disproved, and the burden of proof is continually placed on
the scientist. It is vitally important that observations and experiments,
as well as theoretical calculations, be repeatable by various independent
groups of scientists, to minimize error and bias. This is commonly called
peer review. It is this peer review that assures the fundamental integrity
of science, if not most scientists. In the words of astronomer Vera Rubin
xviii Introduction
(1997, 219), Science is competitive, aggressive, demanding. It is also
imaginative, inspiring, uplifting. At its very core, the process of science
is based on gathering knowledge and revising our understanding based
on that knowledge. There is no such thing as faith in science, no
sacred truthseverything is open to revision. No matter how suc-
cessful a theory may be, it is always possible that new technology or
theoretical investigations could spell its demise. The late astrophysicist
John Bahcall wistfully remarked that every time we get slapped down,
we can say Thank you Mother Nature, because it means were about to
learn something important (Lemonick and Nash 1995, 84).
An important distinction needs to be drawn between the scientic
method, which is designed to be as free from prejudice as possible, and
scientists, who, as human beings, are, unfortunately, sometimes suscep-
tible to personal bias. For example, the big bang theory was disdained by
many scientists in the 1940s and 1950s because they felt it was too theo-
logical (in that it suggested a beginning of the universe rather than an
eternal state of being). There can even be fads in science, hypotheses
which become suddenly popular for a time, only to be abandoned just
as quickly. Astronomer Donald Fernie has lamented this herd mentality
of scientists, especially his colleagues, whom he compares to antelope,
heads down in tight formation, thundering with rm determination in a
particular direction across the plain. At a given signal fromthe leader we
whirl about, and, with equally rmdetermination, thunder off in a quite
different direction, still in tight parallel formation (Singh 2004, 378
379). Scientists can even become personally attached to their theories,
being convinced of the beauty of the idea and reticent to abandon it.
One particularly sarcastic comment is that a competent astrophysicist
can reconcile any theory with any newobservation, while an even more
cynical colleague extended this claim, asserting that the astrophysicist
often need not even be competent (Rees 2000, 39).
On occasion, the scientic establishment has even marginalized new
ideas and observations which contradict the canon of the eld. An
example is the measurement of the expansion rate of the universe made
by Wendy Freedman and her research group in the 1990s using the
Hubble Space Telescope. When her measurements contradicted then
current assumptions about the age of the universe, some astronomers
dismissed her work out of hand. Despite the human foibles of scientists,
science as a process does prevail, as in the case of Freedmans work,
which was subsequently vindicated by further observations. The key is
that because the scientic community relies on peer review, claims of
either newobservations or theories are repeatedly checkedby colleagues
Introduction xix
at other institutions. In this way, claims that have no basis, or outright
frauds (such as the announcements of successful human cloning by
South Korean scientist Hwang Woo-suk) are eventually revealed. This
self-correcting mechanism is one of the key aspects of science.
Science and Religion
In the words of theorist Paul Davies,
No science is more pretentious than physics, for the physicist lays claim
to the whole universe as his subject matter. . . . Physicists, like theologians,
are wont to deny that any system is in principle beyond the scope of their
subject. (Davies and Brown 1988, 1)
Based on comments such as this, as well as issues in the news, such as
debates over stem cell research and the teaching of Intelligent Design,
it is understandable that there is widespread confusion concerning the
relationship between science and religion. In general, the relationship
between religion and science can be respectful or adversarial. One point
of view is that they ask fundamentally different questions and seek differ-
ent answers. In this model, there is no cause for conict. Diametrically
opposed to this is the viewpoint that there will always be tension between
the two elds because they seek to explain the same world through com-
pletely different lenses. An example of the rst perspective is found in
the writings of Tenzin Gyatso (2005, 206), the Fourteenth Dalai Lama,
who explains that
Many aspects of reality as well as some key elements of human existence,
such as the ability to distinguish between good and evil, spirituality, artis-
tic creativitysome of the things we most value about human beings
inevitably fall outside the scope of the [scientic] method.
Physicist Stephen Hawking agrees: Love, faith, and morality belong to a
different category to physics. You cannot deduce howone should behave
from the laws of physics (1993, 173).
In this vein, paleontologist Stephen Jay Gould (1999, 5) suggested a
principle of respectful noninterference . . . by enunciating the Principle
of NOMA, or Non-Overlapping Magisteria. Gould was careful to dene
the distinct magisteria as follows:
Science covers the empirical realm: what is the universe made of (fact) and
why does it work this way (theory). The magisterium of religion extends
xx Introduction
over questions of ultimate meaning and moral value. These two magisteria
do not overlap, nor do they encompass all inquiry (consider, for example,
the magisterium of art and the meaning of beauty). To cite the old clich es,
science gets the age of rocks, and religion the rock of ages; science studies
how the heavens go, religion how to go to heaven. (Gould 1999, 6)
In this way, there is no inherent conict between science and religion.
Abb e Georges Lematre, one of the founders of the big bang model,
similarly wrote that
Once you realize that the Bible does not purport to be a textbook of science,
the old controversy between religion and science vanishes. . . . There is no
reason to abandon the Bible because we now believe that it took perhaps
ten thousand million years to create what we think is the universe. Genesis
is simply trying to teach us that one day in seven should be devoted to rest,
worship, and reverenceall necessary to salvation. (Guth 1997, 55)
Framing the relationship between science and religion in this way brings
attention to a point frequently lost in the ongoing debates, namely
that rather than being antireligous, science is unreligious. Likewise, in
cosmological discussions, physicists, such as Stephen Hawking (1993,
99), openly admit that there is still room for religion:
And even if there is only one unique set of possible laws, it is only a set
of equations. What is it that breathes re into the equations and makes a
universe for them to govern? . . . Although science may solve the problem
of how the universe began, it cannot answer the question: Why does the
universe bother to exist?
With the respectful relationship of NOMA, however, also comes re-
sponsibilities on both sides. Religion has no basis to impose on scientic
theory (such as Young Earth Creationism, or Intelligent Design), while
as Gould (1999, 910) warns, scientists cannot claimhigher insight into
moral truth from any superior knowledge of the worlds empirical con-
stitution. Clearly failure on either side to respect the boundaries of
their magisteria leads to conict. Problems arise between science and
religion when the point of view is taken that each one has a lock on
explaining the one true reality. For example, a religious group or move-
ment may ignore the scientic method and cling to beliefs about the
world which are plainly contradicted by observations (such as the age of
Earth). As the Dalai Lama admonishes, spirituality must be tempered
by the insights and discoveries of science. If as spiritual practitioners we
Introduction xxi
ignore the discoveries of science, our practice is also impoverished, as
this mind-set can lead to fundamentalism (Gyatso 2005, 13). On the
other side, scientists may claim that the scientic method, rather than
being silent on supernatural possibilities, negates the possibility of God.
Both points of view are equally counterproductive.
The benets of a harmonious, integrated relationship between sci-
ence and religion have been expressed by scientists and religious g-
ures alike. On one side, it could create a bridge between science (and
scientists) and the larger human experience. As the Dalai Lama offers,
Since the emergence of modern science, humanity has lived through an
engagement between spirituality and science as two important sources of
knowledge and well-being. Sometimes the relationship has been a close
onea kind of friendshipwhile at other times it has been frosty, with
many nding the two to be incompatible. . . . science and spirituality have
the potential to be closer than ever and to embark upon a collaborative en-
deavor that has far-reaching potential to help humanity meet the challenges
before us. We are all in this together. May each of us, as a member of the
human family, respond to the moral obligation to make this collaboration
possible. This is my heartfelt plea. (2005, 208209)
On the other side, religious practitioners have found that an un-
derstanding of science fosters an appreciation for the wonders of the
universe and can actually deepen their faith. Carl Sagan (1996, 2930)
concurs, writing that
Science is not only compatible with spirituality; it is a profound source of
spirituality. When we recognize our place in an immensity of light-years and
in the passage of ages, when we grasp the intricacy, beauty, and subtlety of
life, then that soaring feeling, that sense of elation and humility combined,
is surely spiritual. . . . The notion that science and spirituality are somehow
mutually exclusive does a disservice to both.
Cosmology owes its origins to spiritual notions about the natural world;
it is tting that it still peacefully coexists with that aspect of human
activity in this modern age.
1
Luminous Matters:
Light and Stars
COSMOLOGY AS A UNIQUE PRACTICE
Cosmologists are at a serious disadvantage compared with other scien-
tists. Biologists can directly experiment on specimens in the laboratory,
raising generations of mice, bacteria, corn plants, or fruit ies, carefully
monitoring how they interact, age, and evolve. Geologists can sample
rocks from around the world, created during different periods in the
earths history, and conduct tests in the lab to determine their age and
composition. Chemists can mix together numerous compounds in the
laboratory in varied proportions and under different conditions, and
repeat the same experiments over and over again. Astronomers in gen-
eral, andmost specically cosmologists, canonly, as your mother warned,
look but dont touch. Limited by what Peebles and Ratra (2003, 560)
call this astronomers Tantalus principle, we cant gather a sample of
stars and bring them into the laboratory to dissect. Neither can we build
our own stars in the laboratory and watch them age and die over a year
or two (the time of the average funding grant). It would be nice if the
universe would repeat the experiment of the big bang before our eyes
so we could test our theories more directly, but since that might necessi-
tate our own destruction, perhaps we should humbly accept our passive
position as mere observers some 13 billion years after the fact without
complaint. The point is that, since cosmologists are largely limited to
observing what events the universe deigns to show us, in whatever fre-
quency and style it decides to do that, creative means are required to
wring every bit of information out of each light beam which comes our
way. This is, of course, one of the reasons why cosmology has exploded
as a eld over the past century as a direct result of improvements in
technology.
2 Cosmology 101
The study of cosmology requires being conversant in a new language,
that of mathematics. For example, cosmology is a science of very big
(and very small) numbers. Scientic notation is a shorthand way of
taking care of all those nasty zeroes (thereby cutting down the length of
astronomy textbooks). For example, six million is written as 6,000,000
or 6 10
6.
On the other hand, six millionths is written as 0.000006,
or 6 10
6
. In each case, the exponent (or power of ten) counts the
number of places the decimal point is moved either to the right or left
of its usual position. Most scientists utilize one of two standard sets of
metric system units to describe physical properties such as length, mass,
and time. These are the so-called mks (meters/kilograms/seconds) and
cgs (centimeter/gram/seconds) systems. Astronomers and cosmologists
are notorious for augmenting these standard systems with unique units
they have contrived in an effort to make their equations prettier. Two
of the most important are the astronomical unit or au (dened as the
average distance between the Sun and Earth, 93 million miles or 150
million kilometers) and the parsec. The denition of a parsec is difcult
to understand out of the context of its development, so we will leave
that for future discussion. For the moment, know that one parsec is
206,265 au, dened in such a way as to exactly cancel out a factor of
(you guessed it) 206,265 in an important astronomical equation! The
parsec (or pc) is the unit astronomers and cosmologists use to discuss the
distances to stars. For larger distances, kiloparsecs (kpc) or thousands
of parsecs, and megaparsecs (mpc), millions of parsecs, are routinely
used. Astronomers also use the star they understand the best, namely
our sun, as a common yardstick against which to measure other stars.
For example, a solar mass unit measures the mass of a star relative to the
mass of the sun.
But what about the so-called light year commonly used in science c-
tion? Although there is certainly some affection attached to the term, it
is not commonly used by astronomers when talking among each other,
as it was somewhat articially dened, does not simplify equations, and
actually leads to confusion when communicating with the general pub-
lic, who commonly believe it to be a measure of time and not of length.
As dened, a light year is the distance a beam of light could travel in one
year. Traveling at a speed of 186,000 miles (300,000 km) per second, the
corresponding distance amounts to about six trillion (6 10
12
) miles
(9 10
12
km). One parsec is 3.26 light years. By comparison, the nearest
star (the Sun), is 93 10
6
miles away. It takes a beamof sunlight just over
8 min to reach Earth, and nearly six hours to reach Pluto. It takes that
same beam of sunlight over 4 years to reach Alpha Centauri, the nearest
Luminous Matters 3
star system outside of our own. Our intrepid ray of solar radiation would
spend over 24,000 years trying to reach the center of our Milky Way
Galaxy, and more than 2 million years crossing intergalactic space to
reach the Andromeda Galaxy, the nearest large spiral galaxy to our own.
If your head is beginning to ache, dont despair, but be warned that
the pain is about to be ratcheted up just a tad bit more. Turn these
numbers around for a second and think about what they really mean.
If it takes sunlight 8 min to reach Earth, when you look up at the Sun,
you are not seeing it as it really is this moment, but how it was 8 min
ago. If the sun were to really be swallowed by the demon Rahu, we
wouldnt know until 8 min after the crime had been committed. Pluto
would continue to receive old sunlight for six hours before noticing
something was amiss, and any Alpha Centaurians would only notice after
4 years time that a certain yellow star had suddenly disappeared. Any
observers in the Andromeda Galaxy would continue to see our sun for
over 2 million years after its demise! This means that the news from
Earth is always woefully out of date when seen from cosmic distances,
and our view of the universe is also past history. The farther out we look
into space, the farther back we see in time. This concept is sometimes
called look-back time, and is routinely the source of wonder and some
bewilderment for poor unsuspecting astronomy students. Such is the
nature of cosmology. As we shall see on our journey together through
the universe, it is not only queerer than we suppose, but queerer than
we can suppose (Yulsman 2003, 38).
THE ELECTROMAGNETIC SPECTRUM
Light is usually thought of as a wave produced by oscillating electric
and magnetic elds, and is therefore called an electromagnetic wave.
Unlike other waves with which people are normally familiar (such as
sound or water waves), light does not require a medium to transmit
through, but instead can travel through an absolute vacuum. Like other
waves, electromagnetic waves have several measurable properties:
1. Light travels at a specic speed, called the speed of light. In a
vacuum, this is about 186,000 miles per sec (300,000 km/sec). In
his special theory of relativity (1905), Albert Einstein demonstrated
that this is the ultimate speed limit for objects traveling in the
universe (an inconvenience for cosmic travelers commonly ignored
in science ction).
2. Light waves come in different sizes, called wavelengths. The wave-
length is the distance measured between two consecutive peaks (or
4 Cosmology 101
two consecutive valleys) in a light wave. The possible wavelengths
of light vary from smaller than an atom to larger than a football
eld. The array of possible wavelengths is commonly called the
electromagnetic spectrum as shown in Figure 1.1.
3. As a light wave travels by an observer, she can count the number
of waves (cycles) that pass by her each second. This is called the
frequency. If you multiply the wavelength of a light wave by its
frequency, you always obtain the speed of light. Therefore, light
with long wavelengths has a small frequency, and light of a short
wavelength has a large frequency.
4. The height of the peaks and valleys of the wave gives the amplitude.
The intensity of the light wave is related to its amplitude.
5. A nal property of light is its polarization. Think about playing jump
rope with a friend. As your friend holds one end still, you can shake
the rope and set it into a wave-like motion. Depending on how you
shake the rope, you can make the waves seem to travel side to side,
up and down, or in any other direction (or plane) you wanted.
If you carefully made the waves move in the same plane all the
time, you will have polarized the rope waves in that direction. Some
astronomical objects emit light polarized in a specic plane, while
others emit light which is unpolarized (i.e., it can oscillate in any
random direction).
Our rst glimpse of the range of the electromagnetic spectrum began
with Isaac Newton in 1671. When he passed seemingly white sunlight
through a triangular chunk of glass called a prism, it was spread out into
a rainbow of different colors which he projected on the wall. He named
this effect the spectrum, after the Latin word for an apparition. What he
really observed were various wavelengths of visible light, the portion of
the electromagnetic spectrum which the human eye can directly detect.
The different colors are bent by different amounts as they travel through
the prism, and therefore spread out into their component wavelengths,
or colors. Today both prisms and diffraction gratings (glass or other
transparent material etched or colored with many close, parallel lines)
are used to decompose light into its component wavelengths. The amaz-
ing amount of information astronomers can glean from such efforts
(called spectroscopy) will be discussed in a following section. Visible light
varies in wavelength from 400 (violet) to 700 (red) nanometers (nm),
where there are 1 billion (10
9
) nm in 1 m. For comparison, the average
width of a single human hair is about 8,000 nm. It is no accident that our
eyes are most sensitive to this part of the electromagnetic spectrum. The
sun, our star, puts out the greatest amount of light at the yellow-green
Luminous Matters 5

Figure 1.1 The Electromagnetic Spectrum
region of the spectrum (which is why it appears yellow to our eyes). This
is also the portion of the spectrum where our eyes are most sensitive.
Through the process of evolution, our eyes have developed to take ad-
vantage of the suns most basic properties. Fortunately for astronomers,
a great deal of the visible light from celestial objects passes through our
atmosphere, allowing us to study our solar system, stars, and galaxies
from the surface of the earth through this atmospheric window.
In 1800, William Herschel found that a thermometer placed just be-
yond the red end of a visible spectrum would warm up, demonstrating
the presence of energy at wavelengths just slightly too long for the hu-
man eye to see. He named these caloric rays, but today we refer to this
as the infrared (IR) portion of the electromagnetic spectrum. Objects
that are not hot enough to glow, but still radiate heat (such as human
bodies and blacktop parking lots on a summer day) emit these wave-
lengths of light. They range from about 700 to 100,000 nm, or 0.7 to
1000 micrometers (m). Night vision goggles and so-called heat-seeking
missiles rely on detecting IR radiation. Heat lamps used to keep food
warm at restaurants also utilize these wavelengths. Stars cooler than the
sun, planets, and clouds of dust, especially those shrouding baby stars,
naturally emit IR. Some wavelengths in this range reach the ground,
but the longer ones are absorbed in the atmosphere, especially by water
vapor, so IR astronomy is best done from mountaintops, balloons, or
satellites above the atmosphere.
Slightly longer than IR waves are microwaves, which occupy the 1 mm
through 3 cm portion of the electromagnetic spectrum. Although we
6 Cosmology 101
normally connect microwaves with popcorn and nachos on demand,
the universe itself emits important wavelengths in this range. This cosmic
microwave background, accidentally discovered in 1965, is a ghostly relic
from the early history of the universe, and provided important evidence
for testing the original big bang model and its subsequent revisions.
The longest wavelengths of the electromagnetic spectrum are those
referred to as radio. This encompasses a variety of human technolo-
gies, including AM, FM, and shortwave radio, radar, and television. The
longest wavelengths considered useful by astronomers are in the range
of 300500 m (the length of several American football elds). The cen-
ters of certain galaxies emit a tremendous amount of radio waves, lead-
ing astronomers to hypothesize the cause to be black holes a million
times the mass of our sun voraciously consuming entire stars and clouds
of gas and dust. Although the existence of radio waves was predicted
by James Clerk Maxwell in the mid-1800s, they were not produced in
the laboratory until 1888 (by Heinrich Hertz), and celestial radio waves
were not detected until 1931. With radio waves, we reach the end of the
long wavelength, low frequency, or red end of the electromagnetic
spectrum.
Returning to visible light, let us explore those wavelengths just a bit
too short for the human eye to see. In 1801, J. Ritter discovered that
silver chloride turned black when exposed to the rays of the suns spec-
trum just beyond the violet wavelengths. This invisible radiation is now
called ultraviolet (UV). With wavelengths between 1 and 400 nm, these
rays interact with human tissue to cause sunburns, skin cancer, wrinkles,
and cataracts. Fortunately for life on Earth, the stratospheric ozone layer
absorbs a signicant amount of the UV produced by the sun. However,
destruction of the ozone layer through human activity (such as the cre-
ation of CFCs, or chlorouorocarbons, the spraying of certain fertilizers,
and jet contrails) is increasing the amount of UV light that reaches the
ground, resulting in a complementary increase in skin cancer rates. In
space, UV rays are produced by objects signicantly hotter than the sur-
face of the sun, such as massive stars, newly formed white dwarfs, and the
centers of galaxies. Because our ozone layer absorbs UV, astronomers
must rely on satellites to study these wavelengths.
In 1895, physicist Wilhelm Roentgen accidentally discovered invisible
rays that could pass through most human soft tissues and allowthe bones
of living beings to be studied from outside their bodies. In 1901 he was
awarded the Nobel Prize for discovering what he called x-rays. While the
benets of emergency room x-ray tests and the occasional dental x-ray
are without dispute, repeated exposure does increase risks for cancer
Luminous Matters 7
(hence the lead apron and questions about possible pregnancy that
accompany x-ray photography, and why the technician leaves the room
or stands behind a screen during the actual test). Fortunately for life
on Earth, ozone (O
3
) and molecular oxygen (O
2
) in our atmosphere
absorb most of the x-rays emitted by the sun and other celestial objects.
By the same token, astronomers who wish to study x-rays emitted by
the suns million-degree corona, hot gas in clusters of galaxies, or gas
being devoured by neutron stars or black holes, must rely on satellite
observations. The wavelengths of x-rays are very small, ranging from0.01
to 1 nm. In comparison, the diameter of an atom is about 0.10.5 nm.
The shortest waves in the electromagnetic spectrum are the gamma
rays ( rays), so named because they were the third type of radiation
discovered to be emitted by nuclear reactions. Physicist Paul Villard
accidentally discovered them in 1900 while experimenting with ra-
dioactivity. With wavelengths smaller than an individual atom (10
7
to
0.01 nm), these rays can be used to sterilize medical equipment and
food (by killing bacteria) and can be used to kill cancer cells (in so-
called radiation treatment). However, uncontrolled exposure can also
kill healthy tissue, leading to radiation sickness, cancer, and death. Once
again, Earths atmosphere protects us from these harmful rays, but pre-
vents astronomers from making ground-based observations. Satellites
allow astronomers to study rays from solar ares and the centers of
galaxies. At this point, we have reached the end of the short wavelength,
high frequency, or blue end of the electromagnetic spectrum. Looking
back on this survey of the electromagnetic spectrum, it becomes obvious
that astronomers must observe the universe at all possible wavelengths
in order to learn as much as possible about various celestial bodies. At
the same time, astronomers must do most of these observations from
space because of the shielding nature of our atmosphere in absorbing
many wavelengths of light (especially the dangerous ones!)
A natural question to ask at this point is why are the harmful forms
of light clustered in the short wavelength end of the spectrum? The
answer was discovered by physicist Max Planck around 1900. He found
that light has a dual naturealthough we have been characterizing it as
a wave, it can also act as a particle (which we call a photon). This wave-
particle duality is one of the basic tenets of quantummechanics, the branch
of physics which studies the realm of the very small, including atoms.
Planck found that there was a relationship between the wavelength of a
light wave and the energy its photon carried. The shorter the wavelength
(or higher the frequency), the bigger the punch the photon can deliver.
Therefore a stream of photons of radio light can bombard your skin
8 Cosmology 101
(and constantly does!) and you will not suffer any ill effect. However, a
small number of ray photons carry enough energy to break apart DNA
and can cause cells to become cancerous or die.
MAGNITUDES
Returning to visible wavelengths, it is obvious to even the most casual
stargazer that stars differ from one another in brightness. This appar-
ent brightness depends on two important factorsthe stars distance
and its energy output. The original scale used to classify stars by their
apparent brightness was developed by the ancient Greek astronomer
Hipparcos and popularized centuries later by Ptolemy in his naked-eye
catalog of over 1,000 stars, published circa a.d. 140. The stars were
divided into six classes, or magnitudes, with the rst magnitude stars
being the brightest. After the invention of the telescope, the magnitude
scale was expanded to include seventh, eighth, and dimmer magnitude
stars. In the eighteenth century, several difculties with this antiquated
scale became impossible for astronomers to continue to ignore. Unfor-
tunately, as is commonly the case with astronomers, rather than throw
out the old system and develop an improved and more logical classi-
cation from scratch, it was decided to patch up the old scale and make
it workable, becoming the eternal bane of subsequent generations of
astronomy students.
One of the difculties was that the objects contained in the rst mag-
nitude represented far too great a spread in apparent brightness. Just as
a homeowner might knock out a wall to expand a room, astronomers
knocked out a mathematical wall, and expanded the magnitude scale to
the left on a number line, rst to magnitude zero and then to negative
numbers. Therefore, the more negative the magnitude of an astronom-
ical object, the brighter it appears. For example, the sun is magnitude
26.7, the full moon is 12.6, Venus is 4.4 at its brightest, and Sirius,
the brightest star in the night sky, is 1.4. The second problem was
that the difference in brightness between two consecutive magnitudes
wasnt consistent. Since the average was about a factor of 2.5, it was
suggested by astronomer Norman Pogson that the difference between
magnitudes be standardized to 2.512, the fth root of 100. Therefore, a
magnitude 1.5 star appears 2.512 times brighter than a magnitude 2.5
star, while a magnitude 2.7 star appears (2.512)
5
or 100 times brighter
than a magnitude 7.7 star.
So far we have only discussed how bright a star appears, or its appar-
ent magnitude. As previously noted, this depends on both the distance
and true energy output (luminosity) of the star. The intensity of light
Luminous Matters 9
decreases as it moves farther from its source, according to an inverse-
square law. For example, since Jupiter is about ve times farther from
the sun than Earth, the sun appears 5
2
or 25 times dimmer from Jupiter
as compared to Earth. Since Pluto is about 30 times farther from the
sun than Earth, the sun appears 30
2
or 900 times dimmer. So a star that
appears dim in our sky might actually put out more energy than the sun
but be located quite far away. Inorder to discuss the actual energy output
of a star, astronomers introduce the concept of absolute magnitude. This
is how bright a star would appear if it were located exactly 10 pc (32.6
light years) away. For example, although the suns apparent magnitude
is a whopping 26.7, its absolute magnitude is a disappointing +4.8, too
dim to see from many light-polluted suburban skies.
Since absolute magnitude is dened with distance in mind, by com-
paring a stars apparent and absolute magnitudes we can determine
its distance using a vitally important equation called the distance modu-
lus formula. For readers intrigued by mathematics, the exact relation is
M m = 55log D, where M is the absolute magnitude, m is the appar-
ent magnitude, and D is the distance to the star, measured in parsecs.
Therefore, if astronomers can somehow determine the true brightness
of a celestial object, its distance can be calculated. This is a very power-
ful technique which will become the basis for measuring much of the
universe.
TELESCOPES
As strange as it might sound, the facts surrounding the invention of
the telescope are still under debate by historians. Credit is normally given
to Johannes Lippershey, a Dutch optician, who applied for a license for
the instrument on October 2, 1608. Soon after, Galileo Galilei heard
about the invention and built one of his own, and became the rst
astronomer to make detailed telescopic observations. The name telescope,
fromthe Greek teleskopos (far-seeing), was coined by Greek poet Ionnes
Demisiani in 1612 (Webb 1999, 42).
Galileos telescope was a refractor, whereas most professional modern
telescopes are reectors. Refraction is the bending of light as it passes
fromone material to another (such as air to glass, and air to water). This
is why the part of a shing pole submerged in a pond looks bent. This is
also how a prism can take sunlight and spread it into its component col-
ors. Different colors (wavelengths) are bent by differing amounts as they
pass through a prismor lens. A telescope that uses a large (diameter, not
thickness) lens to gather light is called a refractor. As seen in Figure 1.2,
the light rays pass through the lens and are bent, cross at the focal point,
10 Cosmology 101
Figure 1.2 Basic Telescope Design
and continuing on to a second, smaller lens called the eyepiece. The
light rays are bent once more and soon after pass into your eye, causing
an image to be created on your retina. All that bending tends to make
the image upside-down, which is not a concern to astronomers (since
the universe has no real up or down), but can be quite disconcerting
to the rst-time observer when looking at the moon (or your neighbor-
hood)! The distance between the center of a lens and the focal point
is called the focal length, and is a fundamental property of that lens.
For those readers who liked to burn leaves with a magnifying glass as a
child, the focal length is the distance between the glass and the leaf, such
that the sunlight appears most concentrated (and the leaf burn most
easily). The magnication of the telescope (how many times bigger it
makes objects appear) can be found by dividing the focal length of the
large lens (called the objective lens) by the focal length of the eyepiece.
To change the magnication, an eyepiece of a different focal length
could be utilized. Theoretically, any telescope can magnify any number
of times using the proper eyepiece (although useful eyepieces with fo-
cal lengths shorter than about one-half inch are increasingly harder to
manufacture and tend to be costly). This must mean that there are other
properties of a telescope which make it better than another.
Any telescope has three powerslight gathering ability, resolution,
and magnication. The rst two depend on the aperture, or diameter of
the objective lens, with bigger denitely being better. As we have seen,
magnication depends on the focal lengths of the two lenses. The most
important of these three powers is the ability to gather light. A telescope
is essentially a light bucket, gathering as much light as possible, in order
to collect enough photons to make a distant object visible. If not enough
light is gathered, no observation can take place! Just as a bigger bucket
Luminous Matters 11
can gather more water, a bigger lens can gather more light, and creates
a brighter image. Once an image bright enough to see has been made,
the ability to see details, or resolution, becomes important. The smaller
the detail that can be seen, the more information that can be gathered.
Although in theory the resolution of a telescope depends solely on
its aperture, in practice our atmosphere creates additional distortions
which smear out the image of a star or galaxy and prevent ne details
from being seen. Telescopes in satellites beyond our atmosphere avoid
this problem. Once a bright, detailed image has been collected, its size
can be manipulated through magnication. This is the easiest of the
three powers to change, but is actually the least important. All too often,
the rst-time telescope buyer is only aware of magnication, and can be
manipulated by salespersons accordingly.
Although well-made refractors give outstanding images, there are
problems inherent in their design. As stated previously, different col-
ors of light are bent by different amounts as they pass through a lens.
Therefore, different colors will actually focus at slightly different points,
causing the image of bright stars to take on a rainbow-like appearance.
This is called chromatic aberration, and plagues cheap refractors. This
can be overcome by using the proper combination of lenses, but it
obviously increases the price of high quality instruments. The second
problem is more serious. As anyone who wears eyeglasses can attest, the
wider the lens, the heavier it is. But unlike eyeglasses, where plastics can
be easily substituted, telescope lenses must be made of glass. This limits
the size of the largest refracting telescope, for if a lens is too large, it will
actually warp under its own weight. The largest refractor in the world
is the 40-inch wide instrument installed at the Yerkes Observatory in
Williams Bay, Wisconsin, in 1897.
In 1668 Isaac Newton invented a different type of telescope, one that
utilized the reection of a large mirror to gather light. The primary mir-
ror of a reecting telescope or reector acts like the objective lens of a
refractor, gathering light and determining the theoretical resolution of
the telescope. Magnication, resolution, and light gathering ability are
determined in a manner similar to a refractor. Reectors have an addi-
tional complication. Because the light bounces off the primary mirror
rather than passing through it, the focal point is located inside the tube
of the telescope. This presents a problem, as stufng ones head inside
the telescope tube in an attempt to view the image prevents the image
from forming in the rst place! Instead, a small secondary mirror is
added to send the light outside of the tube where it can be easily viewed.
Different types of reectors vary in how the secondary mirror is used. In
12 Cosmology 101
Newtons original design (shown in Figure 1.2), the light comes out of a
small hole cut in the side of the tube, where the eyepiece holder and fo-
cuser are mounted. This Newtonian design remains the simplest to build
(as this author can attest) and the most cost-effective telescope design,
and is a favorite among amateur astronomers for these very reasons.
Because mirrors reect all wavelengths the same, they do not suffer
from chromatic aberration. In addition, because the weight of the mir-
ror is more fully supported than in the case of a lens (which is held only
at its edges), huge reectors can be built. However, after a certain size,
even with the support provided by the mount, the mirror will tend to
warp. The largest unwarped primary mirror in the world is the 200-inch
Hale telescope at Mt. Palomar Observatory outside San Diego (1948).
Today, much larger telescopes can be made, either out of segmented
mirrors (such as the 400-inch, or 10-m, Keck Telescopes in Hawaii) or
with exible mirrors equipped with special support mounts that recon-
gure either the primary or secondary mirrors curvature as needed
to compensate for the distortions. The same technology can be used
to compensate for distortions caused by the atmosphere, a technique
known as adaptive optics. The signals fromseveral telescopes can also be
combined in a process known as interferometry, resulting in an image
which would otherwise have required a single telescope of unrealistic
size to obtain. The largest telescope in the world, the VLT (Very Large
Telescope) in Chile combines the signals from four 8.2-m reectors,
mimicking the images from a single 16-m mirror. It is important to note
that if rigid mirrors are not carefully ground to the proper shape, they
will not focus correctly. Such a spherical aberration plagued the Hubble
Space Telescope until special corrective optics were incorporated into
its cameras.
Modern astronomical research is no longer done by a human be-
ing directly looking through the eyepiece of a telescope (affectionately
called eyeball-on-glass). Since the late 1800s, technology has enabled
astronomers to record the light received from celestial bodies, allowing
it to be studied at a later time and in greater detail. These technologies
can also collect more light than the human eye, allowing for dimmer
and dimmer objects to be studied. The rst great technological advance
of this type was obviously photography. The rst photograph of a star was
made in 1850 by J.A. Whipple at Harvard College Observatory. Over the
next two centuries, photography greatly expanded our understanding
of the solar system, stars, and galaxies. In a rather practical sense, it also
opened the door to women in astronomy. As it was deemed improper
and impractical for women to spend time in a dark, cold observatory,
Luminous Matters 13
the fact that the photographs could be studied in the comfort and safety
of a laboratory or library allowed women to work at major observatories
for the rst time in the late 1800s.
The charge-coupled device, or CCD, was invented in 1969 by Willard
Boyle and George Smith at AT&T Bell Labs, again revolutionizing as-
tronomy. These sensitive detectors use a computer chip to record almost
every photon that strikes them, converting the information into an elec-
trical signal which is processed by a computer. Using CCD technology,
backyard amateur astronomers are now obtaining images of planets and
galaxies superior to those obtained photographically a century ago by
the worlds largest telescopes.
Wavelengths other than visible light have their own types of tele-
scopes, based on the principles of optics outlined above. For example,
radio telescopes are generally constructed as large parabolic dishes that
collect radio wavelengths and focus them to a detector. Radio astron-
omy can be done in the daytime as well as on cloudy days (at some
wavelengths). Remember, though, that radio wavelengths are much,
much longer than those of visible light, and it turns out that the res-
olution of a telescope depends on the specic range of wavelengths
being collected. This means that radio telescopes must be huge in com-
parison to optical telescopes, and in practice dont achieve the same
level of detail. Radio sources are also typically very faint, so the addi-
tional light gathering ability of a large telescope is required. The largest
single dish radio telescope is the 1,000 ft Arecibo dish, built into a nat-
ural bowl-shaped valley in Puerto Rico in 1963, and was featured in the
James Bond movie Goldeneye. The technique of combining signals from
several telescopes (interferometry) was actually rst used with radio
telescopes.
The most important radio wavelength in astronomy is 21 cm, caused
by cool clouds of hydrogen gas, or HI regions. An atom of hydrogen con-
sists of one electron orbiting one proton. Each particle is said to possess
the property of spin, which for the sake of simplicity we will (incorrectly)
consider here as being like the rotation of the earth around its poles.
The electron and proton can either spin in the same direction, or in
opposite directions. When the two particles spin in the same direction
(parallel), the atom has slightly more energy than in the case where
they spin in opposite directions. The electron doesnt pick this higher
energy state on its own, but can be coaxed into it by collisions with other
particles. When the electron changes its direction (in order to return to
a lower state of energy), it gives back the energy by emitting a photon
of radio light with wavelength 21 cm. This special radiation is critical
14 Cosmology 101
in mapping the structure of the Milky Way, where the presence of dust
blocks visible light, making it otherwise impossible to see large parts of
our own galaxy.
SPECTROSCOPY
Spectroscopy is the technique of intentionally breaking down the light
from a source into its component wavelengths, using a prism or diffrac-
tion grating. It allows scientists to determine many physical properties
of a light source, such as its composition, motion, and temperature. In
1814 Josef von Fraunhofer carefully analyzed the visible spectrum of the
sun, and found that the rainbow of colors was interrupted by 10 promi-
nent dark lines he dubbed A through H, and an additional 574 fainter
lines between the B and H lines. This absorption spectrum became the
basis of the chemical analysis of the composition of the sun.
In 1859, Gustav Kirchhoff developed three important rules governing
spectra illustrated in Figure 1.3, all based on the observations of himself
and Robert Bunsen:
1. A hot, glowing solid, liquid, or dense gas emits a continuous spec-
trum, consisting of a solid rainbow of all colors (wavelengths). An
example is the metal lament in a standard incandescent light bulb.
2. A hot, thin gas emits light only at certain, particular wavelengths,
creating an emission spectrum, and these wavelengths are unique
to the composition of that particular gas.
3. If a continuous spectrum is passed through a cool, thin gas, the
gas will absorb only certain wavelengths, creating an absorption
spectrum. The wavelengths that are absorbed are identical to those
emitted by the same gas when heated.
Because the wavelengths emitted and absorbed are unique to that
chemical element, the presence (or absence, inthe case of anabsorption
spectrum) of a particular spectral line inthe light coming froma celestial
object allows astronomers to determine its chemical composition. A set
of spectral lines is as unique to an element as a set of ngerprints is to a
person. In 1864, William Huggins was able to show that some nebulae
inhis time a generic termfor any object that appeared fuzzy whenviewed
in a telescopeproduced emission spectra and are therefore hot clouds
of gas in space. He was also able to show that they were comprised of
the same elements that make up the sun. Spectroscopy yielded other
surprises closer to home. For example, helium (named for Helios, the
Greek god of the sun) was discovered in the spectrum of the sun in
1868, nearly 30 years before it was found on Earth.
Luminous Matters 15
Figure 1.3 Kirchhoff s Rules of Spectra
In order to understand why spectroscopy allows for chemical identi-
cation, we must return to the eld of quantum mechanics. In 1912,
Danish physicist Niels Bohr developed a model of the atom that resem-
bled the solar system. In this model, the negatively charged electrons
orbit the positively charged nucleus in specic, allowed orbits, each with
its own inherent energy. The simplest atom to consider is hydrogen. It
also makes up about 75 percent of the universe (by mass), so it is a very
important atom to understand. A single electron orbits a single proton
due to the electrical attraction of their opposite (but equal) charges.
Because of this attraction, the electron normally occupies the closest
allowed orbit to the proton, known as the ground state.
In order to occupy other orbits which are farther from the nucleus,
also called excited states, the electron must absorb energy (in the form
of a photon). But not any old amount of energy will do! It takes a specic
amount of energy for the electron to jump to the rst excited state,
and unless the electron absorbs a photon of just the right energy (with
just the right wavelength), there will be no transition. The absorbed
photon will create an absorption line in the spectrum of the light pass-
ing through this collection of atoms. Likewise, there is another specic
amount of energy (and hence another corresponding wavelength of
light) associated with a transition to the second excited state, the third,
16 Cosmology 101
and so on. Since there are many, many possible excited states, there are
many possible transitions, each with its unique associated wavelength of
light. The spacing of the orbits (also known as energy levels) is xed
by the atomic properties of hydrogen, and is different for other ele-
ments, thus resulting in unique spectral lines being associated with every
element.
Once in the excited state, the electron prefers to release its acquired
energy and return to the ground state. In order to do this, it must emit
a photon of light of the exact same energy (and wavelength) that it
absorbed to make the jump in the rst place. This creates the unique
emission lines created by a cloud of hot, thin gas. Many electrons are ex-
cited due to the high temperature, and in order to return to the ground
state they must emit energy, thereby creating the emission spectrum.
Some transitions are more easily accomplished under conditions
found on Earth and are therefore more commonly seen. However, there
are certain transitions that are very difcult on Earth, but much easier
in the near-vacuum of space. These transitions were unexpected by as-
tronomers and when discovered in the spectra of celestial objects were
named forbidden lines. An important example is a green spectral line in
the spectrum of oxygen atoms that have lost two of their eight electrons
(doubly ionized oxygen). It was originally attributed to an unknown el-
ement dubbed nebulium, since it was discovered in the spectra of hot
clouds of gas in space.
Among the contributions to quantum mechanics of physicist Max
Planck was the realization that a truly continuous spectrum over all
wavelengths is only possible for an idealized object known as a blackbody.
This perfect emitter is also a perfect absorber, and hence should be
utterly black when cold. Although blackbodies and their radiation are
idealizations, stars are decent approximations (at least near the visible
portion of the spectrum), as is the cosmic microwave background.
Blackbodies have three important properties that are of use to
astronomers:
1. They emit light over a wide range of wavelengths, with varying
amounts of energy at different wavelengths. The graphical repre-
sentation of this is called the Planck curve.
2. They emit most intensely at one particular wavelength, which is
related to their temperature. This results in a peak in the Planck
curve. The hotter the object, the shorter the wavelength of maxi-
mum emission. This explains why hot stars are bluish, cooler stars
are reddish, and middle of the road stars like our sun appear
Luminous Matters 17
yellow. The relationship is called Wiens law, and gives astronomers
a way to estimate the surface temperature of the star.
3. The hotter and larger the blackbody, the greater its total energy out-
put, or luminosity. The luminosity is related to its surface area (ra-
dius squared in the case of a sphere) and its temperature raised to
the fourth power. This StefanBoltzmann law relates true brightness,
surface temperature, and size of a star. If the rst two properties
are known, the third can be estimated.
A nal property of stars and galaxies that can be gleaned from a
study of light is motion toward or away from Earth (known as the radial
velocity). In 1842, Austrian scientist Christian Doppler announced that
the wavelength of any wave (be it a water wave, sound wave, or light wave)
becomes shorter if the wave source were approaching the observer, and
would become longer if the source were receding. This Doppler effect is
most obvious for sound waves, where the change in wavelength creates
a change in the pitch (not volume) of the sound. If an ambulance is
approaching you, its siren becomes more shrill (its pitch increases) as its
wavelength shortens, but when the same ambulance rushes away from
you, the pitch lowers (becomes more bass) as the wavelength stretches
out. Likewise, when a star or galaxy approaches Earth, its light is shifted
toward shorter values, or toward the blue end of the electromagnetic
spectrum. This is called a blueshift. A receding star or galaxy has its light
waves lengthened, shifting toward the red end of the spectrum (termed
a redshift). The amount of the shift directly depends on the speed of
the object. Therefore, it is possible to calculate both the direction of
the motion (toward or away) as well as the speed. For example, the
bright star Sirius was found to be approaching Earth by William and
Margaret Huggins in 1868 based on a 0.015 percent blueshift in its
absorption spectrum, as compared to the suns spectral lines (Singh
2004, 244).
Although astronomers passively receive the shifted light from objects
in space, on Earth we can actively bounce wavelengths (usually in the
microwave or radio part of the spectrum) off a moving body and by
noting the change in wavelength of the light that returns, estimate the
speed of the moving body. This is the basic principle behind the Doppler
radar used by meteorologists to study the motion of precipitation inside
clouds, as well as radar guns used by police to catch speeders. It is
indeed amazing to think that the same technology that studies the spi-
raling formation of tornados can be used to examine the rotations of
galaxies.
18 Cosmology 101
THE CLASSIFICATION OF STARS
Although many people are under the impression that all stars are
white, a careful survey of the night sky will convince them that this is
really not the case. Some stars appear yellowish, like our sun, while
others exhibit varying intensities of blue, orange, or red. As we have
already seen, the color of stars is determined by their surface temper-
ature, through Wiens law. Stars also differ in size and true brightness
(luminosity or absolute magnitude), with these properties related to the
temperature through the StefanBoltzmann law. Astronomers use these
properties to classify stars into various groupings, and then utilize this
information to study their lives and deaths.
It bears repeating that most of what astronomers learn about stars is
achieved by studying their spectra. This includes composition and mo-
tion. In this vein, Annie Jump Cannon lauded the universes Patient
light! Shining down on humanity these countless centuries until man
became clever enough to wrest from its vibrations the secrets so closely
concealed (1941, 56). It is therefore natural to ask if stars exhibit differ-
ent types of spectra and if so, what inherent properties are responsible
for these differences. The answer was rst noted by Fraunhofer himself
around 1814. Not only did he study the spectral lines of the sun, but
he found that some bright stars had similar spectra, while others had
spectra with differing appearances. The rst signicant classication sys-
tem was done by Jesuit priest and astronomer Father Angelo Secchi in
the 1860s. His equipment was crude by modern standards, and he was
limited to making drawings at the eyepiece (as Fraunhofer had done),
but he was still able to divide stellar spectra into ve basic classes. Type
I stars, including Sirius, were bluish-white in color (and therefore hot-
ter than the sun). Their spectra exhibited stronger hydrogen spectral
lines than did the sun (although Secchi did not understand this). Type II
stars were similar to the sunyellowstars of mediumtemperature whose
spectra exhibited relatively strong lines now known to be due to calcium
and other metals. Type III stars were cooler orange and red stars. Their
spectra had curious bandsclose forests of linesnear the blue end of
the spectra. These are now known to be due to molecules of titanium
oxide (TiO). Type IV stars were very red and had bands of lines due to
carbon (which Secchi himself realized). Type V stars were very peculiar
and rare, exhibiting strong emission rather than absorption lines.
Secchis classication was very inuential in his time, and further re-
nements required better technologynamely photographs of spectra
which could be studied and repeatedly compared. The rst photograph
Luminous Matters 19
of a spectrum (of the bright star Vega) was taken by New York physician
and amateur astronomer Dr. Henry Draper in 1872. After achieving this
task, he and his wife, Anna, set a more ambitious goal of photographing
and classifying stellar spectra in more detail than Secchi had been able
to achieve. Unfortunately all those chilly nights in his personal observa-
tory led to Drapers death in 1882 from pneumonia at age 45. After her
husbands death, Anna Draper tried to carry on the project with a series
of unsatisfactory assistants, but nally gave up, and after exchanging a
series of letters with the Director of the Harvard College Observatory
(HCO), E.C. Pickering, she decided to endow the HCO with sufcient
funds to carry out the work in her husbands honor. The result was the
Henry Draper Memorial, which produced the most detailed catalogs of
stellar spectra ever compiled.
Pickering had begun hiring women computers in 1879, low-paid
workers who did much of the labor-intensive calculations and measure-
ments of the mountain of photographic plates generated by the various
ongoing research projects at the HCO. With the funding provided by
Mrs. Draper, Pickering was able to hire so many women that they be-
came known as Pickerings Harem. Beginning in 1886, the women
computers at Harvard began measuring the positions of spectral lines
for thousands of stars, using the suns spectrum as a standard. The ini-
tial work was mainly done by Nettie Farrar, who left HCO at the end of
the year to be married. She was replaced by Pickerings former maid,
Williamina Paton Fleming, who, according to an often-repeated story,
was originally hired as an observatory assistant by Pickering some years
before in a t of exasperation, complaining of his male assistants that
his housekeeper could do a better job. Fleming became the supervisor
of a number of women computers, and in 1898 became the rst woman
to receive an ofcial HCO appointment, Curator of Astronomical Pho-
tographs. The rst Henry Draper Catalog was published in 1890 and
contained the classied spectra of 10,351 stars. Pickering and Fleming
had developed a new system of classication in the preparation of this
work, an alphabetical system from A to M (deleting J, to avoid confusion
with I) which listed the spectra in order of increasing complexity, and
decreasing relative strength of the hydrogen lines. Letters NQ were
reserved for stars of peculiar spectra.
By this time, Pickering had already begun taking advantage of techno-
logical improvements and organized more detailed evaluations of stellar
spectra. In 1888 he hired Drapers niece, Antonia Maury, to reclassify
stars in the northern half of the sky utilizing this improved technology.
Maury, a graduate of Vassars astronomy program (under the direction
20 Cosmology 101
of Maria Mitchell, the rst American woman astronomer), bristled un-
der Pickerings orders to merely improve Flemings system, and instead
developed her own unique classication. Not only did she separate the
spectra into 22 numerical classes, but she reordered some of Flemings
classes, putting O and B before class A. More importantly, she rated
the spectra on a second dimension, the appearance of the spectral lines
themselves. Normal spectral lines were class a, while class b spectral lines
were hazy, and class c were unusually narrow. Maurys catalog and sys-
tem were eventually published in 1897, but went virtually ignored for
a decade. The HCO instead ofcially sanctioned the system devised by
fellow Harvard computer Annie Jump Cannon.
Cannon, a Wellesley physics graduate, was hired by Pickering in 1896
to classify the southern stars. She agreed with Maurys rearrangement of
the original Fleming classes, but ignored Maurys line-width scheme. She
also omitted classes which had been shown to be artifacts of the photo-
graphic plates, resulting in the now famous spectral system OBAFGKM,
often remembered by astronomy students through the rather politically
incorrect mnemonic device Oh Be A Fine Girl Kiss Me. The system
was later rened by Cannon to include subtypes denoted by the num-
bers 09 within each spectral class. It should be noted that over the
course of her four decades at the HCO, Cannon personally classied
the spectra of over 350,000 stars. Her classication system appeared in
1901, and was endorsed by the International Astronomical Union in
1922. It was ordered that any future extensions or changes utilize her
system as their foundation. In the past decade, the spectral classication
scheme has been accordingly expanded to include even cooler objects,
the L (1997) and T (1999) classes. Approximately half the L class ob-
jects are true stars. The rest, and all the T objects, are brown dwarfs, failed
stars that began life with too little mass to generate energy via nuclear
fusion.
Cannons classes list stars in order of decreasing surface tempera-
ture (as determined by their color), although the actual temperatures
corresponding to each spectral class were determined by Cecilia Payne-
Gaposchkin in 1925 as part of her ground-breaking Ph.D. thesis. It was
the rst thesis completed in astronomy at Harvard, and in it she also
demonstrated that the overall composition of most stars is similar
approximately 75 percent hydrogen and 25 percent helium (by mass).
The idea that the relative strength of spectral lines of different elements
in a star is due to temperature and not composition may seem contra-
dictory, but actually follows directly from the same concept of energy
levels in an atom that gives rises to the lines in the rst place. Since the
Luminous Matters 21
spacing of the energy levels varies from element to element, there is
some range of temperatures at which the electrons of a given element
most efciently and most often jump to excited energy states. However,
if the energy (temperature) is too great, electrons can actually be re-
moved from the atom, called ionization. This will weaken the relative
strength of the excited state spectral lines. If the energy (temperature)
is too low, the electrons wont make many jumps at all, and the strength
of the spectral line will also be weaker.
THE HR DIAGRAM
But what of Maurys second dimension? In 1905, Danish astronomer
Ejnar Herzsprung discovered that red stars come in two very different
varietieslow luminosity dwarfs located in our astronomical neighbor-
hood, and high luminosity giant stars which can be seen over large
distances. Interestingly, these red giants were exactly the same stars
which Maury had labeled with her c-characteristic. Herzsprung wrote
to Pickering, calling her discovery,
the most important advancement in stellar classication since the trials by
Vogel andSecchi. . . . To neglect the c-properties inclassifying stellar spectra,
I think, is nearly the same thing as if the zoologist, who has detected the
deciding differences between a whale and a sh, would continue classifying
them together. (Jones and Boyd 1971, 240)
Pickering countered that the spectra did not show enough detail to
verify Maurys claim. During the same period of time, the preeminent
American astronomer Henry Norris Russell had independently discov-
ered that stars could be found in different size types. It was Russell who
rst plotted a diagram demonstrating this, and today the Hertzsprung
Russell diagram forms the backbone of stellar astronomy. What Maury
had unknowingly discovered was that swollen giant stars have lower
atmospheric pressures than smaller stars, resulting in the different ap-
pearance of their spectral lines.
The HRdiagram(shown in Figure 1.4) is generally plotted using spec-
tral class, temperature, or some measure of a stars color on the horizon-
tal (x-axis), and luminosity, absolute magnitude, or some other measure
of a stars intrinsic brightness on the vertical (y-axis). Dim stars are plot-
ted near the bottom, bright stars near the top. If absolute magnitude is
used, this means that large positive numbers (such as +15) are near the
bottom and large negative numbers (such as 7) are near the top. Hot-
ter stars appear onthe left side of the diagramandcooler stars towardthe
22 Cosmology 101

Figure 1.4 The Hertzsprung-Russell (HR) Diagram

right. It is found that over 90 percent of stars appear on a fairly narrow
swath starting in the upper left corner (bright and hot) and moving to-
ward the lower right corner (dim and cool). This is where normal stars
are found, those that Hertzsprung called dwarf stars, more commonly
knownas mainsequence stars. Notethat thesestars arenot necessarily small,
as our sun belongs to this category; they are, however, much smaller than
the giants. If our sunwere replacedby a redgiant, it wouldswallowupthe
orbits of all the planets through Earth. Stars not found on the main se-
quence are dying stars. For example, a small clump of stars are located in
the lower left sectionof the HRdiagram. These are very hot, yet very dim,
meaning they must be small in size. These are stellar corpses approxi-
mately the same mass as the sun, called white dwarfs. Stars in the upper
right part of the diagram are relatively cool, but very bright, meaning
they are large in size. This is where the giants and supergiants dwell.
A detailed study of the HR diagramand the various size classes of stars
was published by William Morgan and Philip Keenan, assisted by Edith
Kellman, at Yerkes Observatory in 1943. The resulting MK or Yerkes
system was a two-dimensional system, like Maurys, which introduced a
Luminous Matters 23
luminosity class in addition to the regular spectral class. The luminosity
classes are designated by Roman numerals, and are as follows:
Ia bright supergiants
Ib supergiants
II bright giants
III giants
IV subgiants
V dwarfs (main sequence)
VI subdwarfs
For example, the sun is a G2V star. The reason for the large number
of luminosity classes was rst hinted at in the work of Nancy Roman at
Yerkes Observatory. She noticed that some stars had very few elements
heavier than hydrogen and helium, although their surface temperatures
appeared the same as normal stars. This was some of the rst evidence
that stars occur in different populations by composition.
THE INTERSTELLAR MEDIUM
Interstellar space is commonly thought of as being completely devoid
of matter. Fortunately this is not the case, for without the interstellar
medium (ISM) new generations of stars would not be possible. Clouds
of gas and dust are generically termed nebulae, and are commonly as-
sociated with either the birth or deaths of stars. They form a sort of
astronomical Rorschach inkblot test, and often carry whimsical names
such as the Horsehead, the Crab, and the Cats Eye.
Interstellar dust grains are believed to be composed of carbon
(graphite) and silicates, the former being created in the atmospheres of
a particular type of cool supergiants formerly called carbon stars (Sec-
chis Type IV stars). Because the size of these dust grains is in the same
range as the wavelengths of visible light, light interacts with the dust
in interesting ways. Light from distant stars is generally dimmed by the
intervening dust, an important effect called extinction. It causes stars and
other celestial objects to appear dimmer than they should, introducing
error into distance calculations which rely on a comparison of apparent
and absolute magnitude. If a star appears dimmer than it should, it will
be assumed to be more distant than it truly is. Blue light is preferentially
scattered from the dust because of the size of its wavelength, resulting
in two complementary effects. First, blue light from hot stars scatters off
24 Cosmology 101
clouds of dust, turning these reection nebulae blue. Second, since the
light from the star has lost some of its blue component, the starlight is
deblued, or appears redder than it should. Like extinction, this inter-
stellar reddening must be taken into account when studying the properties
of stars. If the dust is dense enough, it will prevent any visible light from
traveling through, creating a dark nebula which is usually only seen
when positioned in front of a luminous nebula.
The other component of the ISM is gas, mainly hydrogen, either neu-
tral (cool) or ionized (hot). HI regions are clouds of cool hydrogen which
are detectable via the 21 cm radiation they emit. HII regions are reddish
clouds of hot hydrogen. Ultraviolet light from nearby hot, young stars
ionizes the hydrogen. The electron is eventually recaptured by its nu-
cleus (proton), gradually winding its way back to the ground state. As
the electron falls from the third energy level to the second it emits a
characteristic red wavelength called hydrogen alpha, which is respon-
sible for the red color of these emission nebulae. Planetary nebulae are
shells of gas emitted from dying stars, while supernova remnants are the
shredded remains of exploded stars, returning enriched material to the
ISM for recycling into a new generation of stars.
Approximately half of the mass of the ISM is contained in dense
clouds composed mainly of molecular hydrogen (H
2
). These molecu-
lar clouds also contain over 100 different compounds including water,
formaldehyde, alcohol (both the rubbing and drinking varieties), sugar,
and even amino acids, the building blocks of life. Exceedingly cold (only
5 degrees above absolute zero), these gigantic blobs of gas can contain
10 million times the mass of the sun in a diameter of dozens of parsecs.
Molecular clouds are the stellar nurseries in which new stars form. But
how can a diffuse, distended cloud of gas become a compact nuclear
inferno?
THE LIVES OF STARS
Star Formation
Star formation is believed to be kick-started by some sort of trigger,
perhaps the shock wave from the supernova explosion of a nearby dying
star, or the intense stellar wind rushing out from neighboring hot and
bright O- and B-type stars. A small knot of material within the cloud is
compressed, and gravity takes over. As the forming protostar collapses its
atoms collide, generating greater and greater heat. In order to conserve
angular momentum, any initial rotation the protostar has also increases,
just as a gure skater spins faster by pulling in her arms. This creates
Luminous Matters 25
a serious problem for the would-be star. Unless it nds a way to shed a
signicant portion of its angular momentum, calculations show that a
newborn star would spin faster than the speed of light, which clearly is
not allowed. By comparison, as Galileo discovered, the sun spins roughly
once a month. The key is a protostars magnetic eld, which can act like
a brake, slowing down its rotation. Otherwise, the protostar might be
ripped apart by its own spin! As the protostar continues to contract
and heat up, a disk of material forms around it, which can eventually
form planets, as in our own solar system. In fact, there is so much
excess material associated with star formation that it typically takes place
shrouded in a cocoon of dust and gas, hiding it from direct view until
the stars stellar wind turns on and blows it away. Astronomers normally
study protostars through the IR produced from the cocoon, which itself
is warmed by the protostar it contains.
When the protostars core temperature reaches about a million Kelvin
(10
6
K), fusion occurs for the rst time. Fusion is a nuclear reaction in
which small atomic nuclei are combined to form a larger one, releasing
a tremendous amount of energy. Todays thermonuclear weapons (hy-
drogen bombs) are an uncontrolled example. Back to the Futures Doc
Brown may have safely harnessed fusion to power his time machine,
but real scientists have not been so fortunate. The original atom bombs
and nuclear power plants are examples of a different type of nuclear
reactions called ssion. In this process, a large nucleus, such as uranium
or plutonium, is split into smaller pieces, also releasing energy.
In the protostars rst bout of fusion, deuterium, or heavy hydrogen,
fuses to form helium-3, a light form of helium with a nucleus made of
two protons and one neutron. This temporarily halts gravitys collapse of
the protostar, and in fact the object temporarily swells to several times its
nal size. But deuterium is quite rare, so this phase in the protostars life
is brief by stellar standards. Once the deuteriumin the core is exhausted,
gravity returns to center stage, and the protostar resumes its collapse. If
it is about sixty times the mass of Jupiter, or 6 percent the mass of the
sun, its core reaches the 2 10
6
K necessary to fuse lithium, the third
element on the periodic table. Once again, the collapse is temporarily
halted for a few million years, until the protostars limited supply of this
rare material is exhausted as well.
The collapse now resumes under gravitys relentless urging, and if
the protostar is smaller than 75 times the mass of Jupiter (7.5 percent
the mass of the sun), no further fusion cycles are possible. Its core
simply cannot generate enough heat to overcome the natural repulsion
of positively charged protons for each other. Such objects are failed
26 Cosmology 101
stars, or brown dwarfs. They continue to collapse until the electrons get
as close together as they possibly can, lling up all the lowest energy
states. This degeneracy pressure holds up the brown dwarf against further
collapse. Once formed, the brown dwarf can do nothing more than
become cooler and dimmer over time.
A happier fate awaits protostars above the critical mass limit of 0.075
solar masses. When the core nally reaches about 10
7
K, protons (hydro-
gen nuclei) are moving fast enough to overcome their normal electrical
repulsion for each other, and instead fuse together to form helium,
releasing considerable energy in the process. The energy stabilizes the
interior against the pull of gravity, and a star is born. It is at this point
that the new star occupies a point on the main sequence of the HR dia-
gram. The zero age main sequence (ZAMS) is a line on the HR diagram
representing the original position of stars of different masses, with the
most massive at the top left and the least massive on the bottom right.
The more massive the star, the higher its temperature, and the faster it
uses up its fuel to generate that energy, resulting in a greater brightness.
The downside is that it dies sooner.
Our discussion has focused on single stars, but it is clear that any
molecular cloud contains enough material to create many, many stars.
About half of all sun-like stars and one quarter of lower mass stars are
members of binary systems, in which two stars orbit each other. Multiple
births in the form of triplets and quadruplets of stars are, as in the case
of humans, much less common. Larger groupings of stars, with dozens
or even thousands of members, also exist. These are OB associations,
open clusters, and globular clusters. OB associations are, as their name
implies, loose groupings of dozens of very hot and short-lived O and
B stars, spread across 100 pc. The most famous example is the stars of
Orions belt. The mutual gravitational pull of the individual stars is not
sufcient to hold the association together, and they tend to fall apart
in only a few million years (assuming that their member stars dont die
rst!). Open clusters are also irregularly shaped, but include from several
hundred to a thousand fairly young stars within a diameter of several
dozen pc or less. The Pleiades, or Seven Sisters, in Taurus is a classic
example. Open clusters are more tightly bound than OB associations,
but stars can still wander away over time as the cluster is tugged upon by
other stars or giant molecular clouds. OB associations and open clusters
are sometimes called galactic clusters, as they are found in the spiral
arms of the Milky Way.
Globular clusters represent a very different sort of cosmic critter. These
are spherical swarms of hundreds of thousands to a million older stars
Luminous Matters 27
packed into a diameter of less than 50 pc. In a typical backyard telescope
they have the general appearance of a small cotton ball, with some
individual stars resolvable at the edges. Most globular clusters occupy a
more or less spherical halo aroundthe plane of our galaxy, andrepresent
some of the oldest stars known. Because of their age, globular clusters
are important laboratories for the study of stellar evolution.
Stellar Evolution
When gazing skyward, it is easy to believe, as the ancients did, that the
stars are eternal and unchanging. But as was sometimes noted, stars can
dramatically change in brightness, and in doing so seemingly appear
and disappear from view. The ancient Chinese noted the occasional
appearance of guest stars, while Renaissance astronomer Tycho Brahe
coined the term nova, the Latin term for new. Despite these rare
yet spectacular events, the evolution of an individual star is generally so
slow that a single generation of astronomers cannot hope to see much
in the way of change in the properties of any individual star. How, then,
can astronomers claim to understand the evolution of stars from cradle
to grave with any accuracy? The key is the large number of stars we have
to observe, each at a slightly different point in its life cycle. Consider
the case of a group of ant botanists. Compared to the lifespan of an
individual tree, a single ant lives for a trivially short period of time. But
by observing many trees of different species at different points in their
developmentseeds, seedlings, saplings, mature trees, dying trees, and
fallen dead treesthe ant scientists could piece together how trees are
born and die. Astronomers follow the same procedure in studying stars.
Main sequence stars represent the stable portion of a stars life. It is
in hydrostatic equilibrium, where the inward pull of gravity is perfectly
balanced by the outward push of the energy it generates via nuclear
fusion. A main sequence star converts hydrogen into helium in its core.
Because stars are 75 percent hydrogen and 25 percent helium(by mass),
it is tempting to think that by the end of a stars life it becomes 100
percent helium. But note the previous phrase in its core. Only in the
core of the star is the temperature high enough for fusion to occur. This
means that unless the star can effectively circulate its material (called
convection), only the hydrogen in the core is available for fusion. Very
lowmass stars (Mand Lclass mainsequence stars under 0.4 solar masses)
are completely convective, able to circulate their material from core to
surface. Not only are they exceedingly fuel-efcient in that they are
miserly in their energy generation and resulting brightness, they are
also able to use all the fuel at their disposal. These underachievers of the
28 Cosmology 101
stellar world may not dominate our visible night sky, but they represent
the vast majority of all stars in our galaxy, and their estimated lifespan is
10 trillion yearsnearly 1000 times the current age of the universe. As
you might expect, we will revisit this startling concept. Main sequence
stars over 0.4 solar masses are limited to using only the hydrogen initially
in their cores, as convection only occurs in the outer portion of the star
near the surface.
To quote an Iron Maiden song, as soon as youre born, youre dy-
ing. For stars this is most certainly true. From the moment of its rst
appearance on the ZAMS, the very act of shining changes the chemistry
inside a star. There are two general fusion cycles which stars can use to
convert hydrogen nuclei into helium nuclei. Stars with masses similar
to the sun and lower have cooler core temperatures, relatively speaking,
and rely on the protonproton cycle to generate energy. Although there are
several branches of this reaction, each with several steps, the main idea
is that four hydrogen nuclei (four protons) fuse to form one nucleus of
helium (also called an alpha particle), which is composed of two pro-
tons and two neutrons. In addition, several other particles, including
neutrinos, are produced. This process generates a tremendous amount
of energy through the most famous equation of all time: E = mc
2
. Re-
call that the speed of light (c) is a rather large number300,000 km/s.
The equal sign in the formula means that energy can be converted into
mass, and vice versa, with the speed of light squared as the conversion
factor. Because of this, a small amount of mass can be converted into
a tremendous amount of energy. This is similar to certain currency ex-
changes; for example a few dollars can be converted into an impressive
number of rupees. The helium produced by the fusion (and the very
light waste products) has less total mass than the four protons that went
into the reaction, with the excess mass being converted into energy. As
a result, the sun and all main sequence stars are constantly losing mass
as they generate energy. For example, the sun is losing over 4 million
tons of mass every second! At its present rate of energy production, the
sun is not quite halfway through its estimated 10- to 11-billion-year main
sequence lifespan.
With a core temperature of about 15 million Kelvin, the sun gener-
ates 90 percent of its energy using the protonproton cycle. The other
10 percent is due to the carbon or CNO cycle. Stars with several times the
suns mass or greater are able to exploit more fully this more efcient
fusion cycle, which utilizes carbon as a catalyst or agent of change. The
CNO refers to an overview of the cycle, in which a carbon nucleus is
converted into nitrogen, oxygen, and then back to carbon again. In
Luminous Matters 29
the process, four protons are again converted into one helium nucleus,
more energy, and light waste particles. In order for the carbon cycle to
be effectively utilized by a star, the core must be hotter (preferably 20
million Kelvin or more) in order for the reactions to take place, meaning
a higher mass star, and of course carbon must initially be present.
For millions or billions of years (depending on its mass), a star happily
churns out energy through the fusion of hydrogen into helium. But
when all the hydrogen in the core is exhausted, the main sequence
portionof a stars life is spent. The details of what happens next depends,
not surprisingly, on the mass of the star. First we will consider a low mass
star, dened as eight solar masses or less. When hydrogen fusion ceases
in its core, there is nothing to keep gravity at bay, and the helium core
begins to contract and heat up. This extra heat raises the temperature
of a shell of hydrogen surrounding the core to the requisite 10 million
Kelvin and hydrogen begins fusing into helium. But this is no cause
for celebration. The star is forever banished from the main sequence,
and this is just part of the stars dying process. The interior of the star
is not stable, and the outer layers expand out into space, swelling the
stars size to as much as the diameter of Earths orbit. As the stars outer
layers move farther from the hydrogen fusion they cool, and take on a
reddish tinge. This is the red giant phase of the stars life, and it occupies
a position on the HR diagram called the red giant branch (as shown in
Figure 4.7 on p. 104).
As the core continues to collapse and heat up, it eventually reaches
another critical temperature. At 100 million Kelvin, helium fuses to
form carbon. In this triple-alpha cycle, three helium nuclei are converted
into one carbon nucleus and energy. An additional helium nucleus can
also combine with the carbon to form a nucleus of oxygen. Hydrostatic
equilibrium is reestablished inside the star, and it retreats from the red
giant portion of the HR diagram, taking a location on the horizontal
branch. There it remains, for about 20 percent of the amount of time
it had spent on the main sequence, until the helium in the core is
exhausted. Gravity regains the reins, and for the second time the star
becomes a red giant. The increased core temperature converts the shell
of helium created in the hydrogen shell burning stage into a new source
of fusion (helium shell burning), while a shell of hydrogen just outside
this is now hot enough to begin a second episode of hydrogen shell
burning. During this double shell burning stage the star occupies the
asymptotic giant branch of the HR diagram. The stars mass is too low
for its gravity to compress it enough for temperatures necessary to fuse
carbon and oxygen to be reached, so this is the end of the road. The
30 Cosmology 101
tenuous outer layers of the star are repeatedly puffed off into space,
forming a planetary nebula, and the core contracts until the degeneracy
pressure of its electrons halt the collapse, forming a white dwarf. It should
be explained that the term planetary nebula is yet another example of
astronomical obfuscation. The name does not imply that the objects
are related to planets, but rather refers to their roundish shape, which
reminded nineteenth century astronomers of planets.
Recall that low mass red dwarf stars (under 0.4 solar masses) are fully
convective and can convert all their hydrogen into helium. But at the
same token they never achieve high enough temperatures to convert
helium into carbon, nor do they undergo shell burning. Therefore they
never become red giants, but instead slowly become brighter and hotter
as they use up their hydrogen stockpiles. Near the end of their main
sequence lives they may actually look very similar to our sun, but only for
a brief moment, incosmic terms, before collapsing toforma white dwarf.
We now turn our attention to massive stars, more than eight times
the mass of the sun. Unlike their lighter cousins, these heavyweights
generate high enough temperatures in their cores to fuse carbon and
oxygen into heavier elements. In fact, the heavier the star, the higher
its possible core temperatures, and the heavier the elements it can fuse.
These stars will reach the red supergiant section of the HR diagram
in their later evolution, swelling to a thousand times the diameter of
the sun. If our sun were replaced by one of these behemoths, it would
easily engulf the orbits of Mars, the asteroid belt, and in some cases even
Jupiter. Once helium fusion completes, the star permanently becomes
a red supergiant, as each new fusion cycle lasts for a briefer period of
time than the last, and the exterior of the star does not have sufcient
time to readjust. Elements through iron are created as the star dances
its way down the periodic table, creating the elements necessary for
the existence of planets and life. But iron is the end of the road as far
as fusion cycles go. It takes more energy to coax iron to fuse than is
liberated in the fusion reaction. So how were the heavier elements in
the universe created?
The problem is that heavy elements have a large number of protons,
and their large collective positive charge strongly resists any other pro-
tons (or helium nuclei) getting too close. Neutrons, being neutral, are
not rejected. Nature takes advantage of this through the s-process and
r-process, named slow and rapid for their relative speeds. A heavy nucleus
can capture as many neutrons as it wants, until it becomes radioac-
tively unstable, and decays into a different element, perhaps one that
Luminous Matters 31
cannot be created inside stars by any other means. The s-process occurs
in the outer layers of red giant stars, and is a back-door method for low
mass stars to contribute to the chemical evolution of the universe. Once
the new heavy elements are created, they are blown into space in the
planetary nebula, enriching the ISM and the next generation of stars.
The r-process occurs in the explosion of stars, called a supernova. The
study of how chemical elements are created inside stars is called stellar
nucleosynthesis and was pioneered in a seminal paper published in 1957
by E. Margaret and Geoffrey Burbidge, William Fowler, and Fred Hoyle.
As we shall see, in the early history of the universe only hydrogen and
helium were created. Every atom of oxygen you breathe, every atom of
gold you wear, and every atom of carbon in your cells was created inside
a dying star that existed long before the sun. In the famous words of
astronomer Carl Sagan, We are star stuff.
Cepheids and the Period-Luminosity Relationship
During a stars death throes, when it is moving between fusion cycles,
it crosses an area of the HR diagram called the instability strip. As the
name implies, stars in this section of the HR diagram are inherently
unstable, and will actually pulsate. Two important types of pulsating
variable stars can be found here, RR Lyrae stars and Cepheids. RR Lyrae
stars are lower mass stars that pulsate in a day or less. Cepheids, named
after Delta Cephei, the rst example discovered, are higher mass stars
located further up on the HR diagram, and take several to dozens of
days to complete one cycle of variation. Stars that vary in brightness
for any reason are monitored, and once sufcient data is gathered,
the apparent magnitude versus time is graphed, leading to a pattern
called a light curve. The shape of the light curve is unique to that
type of variable, and is used by astronomers to understand why and
how that particular type of star varies, by comparing it to computer
models.
The women computers at the Harvard College Observatory not only
classied stellar spectra a century ago, but also discovered and moni-
tored thousands of variable stars using photographic plates. Among the
most prolic of the variable star discoverers at Harvard was Henrietta
Swan Leavitt. One of her studies involved identifying variable stars in
the Large and Small Magellanic Clouds, now known to be small satellite
galaxies orbiting the Milky Way. Among the more than1700 variable stars
she identied there, she found a number of what we now call Cepheids.
In the Small Magellanic Cloud she was able to plot light curves for
32 Cosmology 101

Figure 1.5 Cepheid Variables and the Period-luminosity Relationship

twenty-ve Cepheids, and noted that the brighter the star appeared, the
longer its period of pulsation. As she explained in her 1912 paper,
there is a simple relation between the brightness of the variables and their
periods. . . . Since the variables are probably at nearly the same distance from
the Earth, their periods are apparently associated with their actual emis-
sion of light, as determined by their mass, density, and surface brightness.
(Leavitt 1912, 3)
In other words, as seen in Figure 1.5, period is related to the absolute
magnitude of the stars. This period-luminosity relationship is one of the
most powerful relations in astronomy, and after her death in 1921,
she was nominated for the Nobel Prize in physics by an astronomical
colleague who was unaware of her demise. For a random Cepheid,
once its period is measured, its absolute magnitude can be estimated,
and from that the distance calculated (through the distance modulus
formula). The only problem is that the exact numerical relationship
between absolute magnitude and period must be rst determined using
Cepheids of known distance.
The rst person to attempt such a calibration was Ejnar Hertzsprung,
who was also the rst scientist to use the generic label Cepheid to
describe these stars. In 1913 he used the observed motions of thirteen
Cepheids in the Milky Way too distant for parallax measurements as a
statistical way to estimate their distances. From this he determined a
mathematical expression for the period-luminosity relation and used it
to estimate the distance to the Small Magellanic Cloud. Although his
calibration wasnt correct, it was an important rst step. As we shall see,
the period-luminosity relationship is now the most reliable means of
Luminous Matters 33
determining the distances to galaxies, so long as individual Cepheids
can be viewed. The rst widely used calibration was done by Harlow
Shapley, later the director of the HCO. Unbeknownst to Shapley and
the astronomical community, it contained several errors, and resulted in
erroneous calculations of distances to galaxies and the size of the Milky
Way for several decades.
The most fundamental error in Shapleys calibration was discovered
in the 1940s by Walter Baade. Using the largest telescopes in the world,
the 100-inch at Mount Wilson and later the 200-inch at Mount Palomar,
he studied individual stars in the Andromeda Galaxy (M31). He found
that the spiral arms had more O and B stars, while the nucleus had
more red giants, and named these two types of stars Population I and
Population II, respectively. He also noted that the Population I stars
were similar to stars in the spiral arms of the Milky Way, while the
Population II stars were reminiscent of stars in globular clusters. Baade
had stumbled upon the same division in stellar chemistry as Nancy
Roman had in her luminosity class studies, namely that stars of different
compositions have different intrinsic brightnesses and occupy slightly
different locations on the HR diagram, even for the same spectral class.
For example, main sequence stars are population I stars, while normal
stars of population II occupy a parallel but slightly lower line on the HR
diagram, the subdwarfs. In discussing stellar chemistry, astronomers use
the term metallicity, which compares the mass of elements heavier than
helium to the overall mass of the star. For example, the metallicity of the
sun is 0.02, so 2 percent of the suns mass is in elements heavier than
helium.
Baade found 300 Cepheids in M31 and found that they come in
Population I and Population II varieties, with slightly different shapes to
their light curves, and importantly, different period-luminosity relations.
Type I Cepheids, called classical Cepheids, are Population I stars and
are about 1.5 magnitudes brighter than type II (Population II) Cepheids
(now called W Virginis stars). By unknowingly lumping together both
types of Cepheids into his calibration of the period-luminosity relation-
ship, Shapley introduced signicant error into his distance calculations.
As with many astronomical concepts we have encountered thus far, the
naming system of the populations was done before the physics behind
the differences was understood. Population II stars have lower metal-
licities because they represent an earlier generation of stars, and were
made of material that had not been recycled and enriched as much as
that contained within more recent generations of stars (Population I).
Extending this logic, there should exist a rst generation of stars,
34 Cosmology 101
referred to as Population III in keeping with the history of the naming
of populations, which contains no enriched material. This primordial
generation of stars would have been comprised of pure hydrogen and
helium, and as we shall see, understanding them presents unique chal-
lenges to astronomers.
Stellar Corpses
Once a star reaches the end of its fusion cycles, its nal state of
rest depends on its mass. Low mass stars collapse to around the size of
the earth, forming a white dwarf. Further collapse is prevented by the
degeneracy pressure of the electrons. As we have noted, the degeneracy
pressure is created by the electrons being forced as close together as
nature will allow, having lled up all available energy states. This is a
result of the Pauli exclusion principle, one of the many predictions of
quantum mechanics. Simply put, no two electrons can occupy the same
quantum state, just as no two students are allowed to occupy the same
seat in a classroom. Gravity can cram together the electrons available
states, just as the seats in a classroom can be shoved closer together,
but there comes a point where neither can be moved closer together.
This halts the gravitational collapse of the star and holds up the white
dwarf from further collapse. The strength of the electron degeneracy
pressure can only counter gravity if the star dies with a mass under 1.4
solar masses, named the Chandrasekhar limit after Nobel Prize laureate
Subrahmanyan Chandrasekhar who rst calculated it. Stars that die with
greater masses explode as a supernova. Note that it is the stars nal mass,
not initial mass, that matters. There are various ways that a star can shed
mass during its nal stages of life, such as a planetary nebula.
There are actually two main classes of supernovae, which explode for
different reasons. The rst, type Ia, involve a white dwarf andits redgiant
binary partner. Mass will owfromthe bloatedredgiant, adding material
to the white dwarf. Since white dwarfs are on a strict weight regiment,
bound by the Chandrasekhar limit, they will resist accumulating too
much mass. As the material spirals onto the white dwarf, it rst forms
an accretion disk. If enough material accumulates, it can ignite, blowing
off sufcient mass to avoid reaching the critical limit. This temporarily
increases the apparent brightness of the system, creating a nova. If too
much material accumulates and the Chandrasekhar limit is reached,
the carbon and oxygen in the white dwarf ignite in a sudden burst
of fusion, and the star explodes, creating a type Ia supernova (SNIa).
In a type II supernova (SNII), a red supergiant explodes once it has
created an iron core and fusion has suddenly ceased. Note that what
Luminous Matters 35
Tycho Brahe originally named a nova was actually a supernova. The
difference between the two was not understood until the early twentieth
century.
After a type II supernova explosion, the remnant of the coreif any
collapses to a diameter of about 20 km. Since its mass is above the
Chandrasekhar mass limit, the degeneracy pressure of the electrons is
of no use. Rather than violate the Pauli exclusion principle and sit in
each others laps, the electrons combine with protons to formneutrons,
with a wind of neutrinos traveling near the speed of light screaming
out of the dying star hours before the explosion is seen at the stars
surface. This type of stellar corpse is therefore composed of neutrons,
and their degeneracy pressure prevents further collapse once the corpse
has compressed to the size of a large city. The rst neutron stars were
discovered by British graduate student Jocelyn Bell in 1967 as part of a
radio survey of quasars. She found unnaturally regular and rapid radio
signals emanating from some unidentied celestial sources, which were
briey dubbed LGM (little green men). These pulsars are now known
to be neutron stars rotating many times a second. Electrons spiraling
in their intense magnetic elds emit a characteristic type of radio waves
known as synchrotron radiation because it was rst seen in atomic collision
machines. These radio waves form two beams, at the north and south
magnetic poles of the neutron star, and as it rotates, beams of radio
waves sweep past Earth like a lighthouse beacon.
Neutron degeneracy pressure can prevent complete gravitational col-
lapse only if the mass is under about 2.5 solar masses. For dying stars
heavier than this, there is no force in the universe that can halt the
collapse, and the material implodes to a mathematical point called a
singularity, where the laws of physics, including general relativity, break
down. Such an objects gravitational eld is so strong (or in relativis-
tic language, it warps space-time to such a high degree) that not even
light can escape. Such an object was named a black hole by physicist John
Wheeler in the 1960s. Because of its very nature, a black hole is not
directly observed. Rather, its presence is discovered by its gravitational
effect on nearby material. For example, Cygnus X-1, the most famous
black hole, was discovered because of its interaction with its blue giant
companion star HDE226868 (the 226,868th star Annie Cannon clas-
sied as part of the Henry Draper Catalog extension). As material is
drawn off the normal star, it spirals into an accretion disk before plung-
ing down the gravitational drain into the black hole. The particles are
moving so fast that collisions heat the disk to a million Kelvin, creating
x-rays, which in turn led to the discovery of the systemby the Uhuru x-ray
36 Cosmology 101
satellite in the 1970s. Black holes of stellar size (typically less than 20
solar masses) are certainly fascinating creatures, but nature creates mon-
strous black holes of millions of solar masses, which lurk in the hearts of
galaxies. It is these which will occupy some of our attention in a future
section.
2
The History of
Cosmology
THE RISE OF THE GEOCENTRIC UNIVERSE
Cultures throughout the world developed their own creation myths, and
over time many were able to utilize careful observations to discern the
natural cycles of eclipses and apparitions of the planet Venus as the
Morning and Evening Star. Modern cosmology traces its history as a
continuous tradition of thought from the ancient Greek philosophers
(circa 600 b.c.) through the Renaissance to Isaac Newton. As with many
other vigilant observers of the world, the scholars of ancient Greece
understood that the earth was round. The fact that different stars were
visible as one moved closer to the North Pole, the observed sinking
of ships below the horizon as they sailed out from the coast, and the
round shadow Earth cast upon the moon during a lunar eclipse led
them to such knowledge. This may have rst been articulated by the
famed mathematician Pythagoras of Samos (b. circa 572 b.c.). Later
Greek historians attributed to him the understanding that the Morning
Star and Evening Star were both the same body. Pythagoras pictured
the heavenly bodies as being perfect spheres, moving in equally perfect
circles. As the sun, moon, and planets moved, they created an ethereal
music beyond the profane ear of mere mortals to discern, the famous
harmony of the spheres.
When asked what he felt was the reason for being born, Anaxagoras
(b. circa 500 b.c.) is reported to have replied, To investigate the sun,
moon, and heaven (Heath 1991, xxxiii). He understood that the moon
shines by reected sunlight, and with this knowledge correctly described
the causes of both solar and lunar eclipses. He is also attributed with
the concept that the stars are distant res, but he felt they were cooler
38 Cosmology 101
than the sun because they were found in a colder environment. Eudoxus
(b. 408 b.c.), a student of Plato, developed a geocentric, or Earth-centered,
model for the universe. Based on thirty-three rotating spheres, the
model sought to explain the apparent motions of the sun and moon
with three spheres each, while the ve known planets (Mercury, Venus,
Mars, Jupiter, and Saturn) were said to have four spheres of motion
each.
As complex as Eudoxus model may seem, an even more intricate
model of fty-ve concentric crystalline spheres was developed by famed
philosopher Aristotle (b. 384 b.c.). All heavenly bodies moved around
the earth, with the stars located on another, xed crystal sphere, located
some large distance beyond the orbit of Saturn. This celestial sphere was
spun by the Prime Mover, a mystical force that was interpreted by later
Christian writers as angels or even the hand of God. Aristotle argued that
the universe was perfect, unchanging, and eternal, without beginning
or end. It was this eternal nature of the heavens which formed part of
his argument for an Earth-centered universe. Aristotle offered that if
the earth moved, it would be unnatural and could not be sustained,
since objects seem to fall toward the center of the earth. He also argued
that Earth had to move toward the center of the universe, but since
everything appears to move toward Earths center, it must be the center
of the universe. He offered that if Earth were to orbit the sun or some
other body, the positions of the stars relative to each other would seem
to shift as we observe the stars from slightly different perspectives in that
orbit. Since this parallax had not been observed, Aristotle argued that
this proved that the earth could not move. Finally, Aristotle appealed to
everyday observationswe dont y off Earth, or even sense an overall
motion, so it appeared to him and many of his contemporaries that
Earth simply could not be moving.
Just as Aristotle noted that objects seemed to fall toward the center of
the earth, he saw re rise toward the heavens, and concluded from this
that Earth and the heavens must be made of two completely different
materials. Since it was commonly believed that Earth was made of four
basic elements (re, earth, air, water), he proposed a fth element,
called aether, which is perfect and unchanging, like the heavens. The
direct inuence of Aristotles writings can be traced through European
science until openly challenged by the observations of Galileo in the
seventeenth century.
A completely different view on the heavens was expounded by
Aristarchus of Samos (b. circa 310 b.c.). Using the geometry of lunar
The History of Cosmology 39
eclipses and the phases of the moon, he argued that the moon is about
36 percent the diameter of the earth, the diameter of the sun is about
seven times larger than the diameter of the earth, and that the distance
between Earth and the sun is about 19 times larger than the distance
between the earth and moon. His calculations were grossly inaccurate,
not surprising given they were based on naked-eye observations. The
actual values are approximately 27 percent, 109, and 388, respectively.
Despite these errors, he was successful in showing that the sun is much
larger than the earth, and advocated for a heliocentric, or sun-centered,
model for the universe.
Perhaps the most famous experiment of ancient times was Eratos-
thenes determination of the circumference of the earth. The director
of the famous library of Alexandria, Eratosthenes (b. circa 276 b.c.)
learned that on the rst day of summer, the sun passed directly over-
head at Syene (now called Aswan). He watched the sun at Alexandria
on the summer solstice, and found that it did not pass directly overhead,
but instead passed 7 degrees lower (a fact he measured by observing the
shadow cast by an upright stick). Not only was this difference caused by
the curvature of the earth, but because the earth is roughly a sphere, by
knowing the difference in angle between the two cities and the distance
between them, one can calculate the circumference of the earth. Since
7 degrees is about 1/50 of the 360 degrees in a circle, the distance be-
tween the two cities must be 1/50 of the total circumference. The answer
he derived, approximately 25,000 miles, is very close to the actual value
of 24, 902 miles. The largest source of error is the fact that the earth is
not a perfect sphere.
Among the greatest astronomers of antiquity was undoubtedly Hip-
parcos of Nicaea (now Iznik, Turkey). Between 150 and 125 b.c., he
was responsible for an impressive number of observations, inventions,
and theoretical models. The magnitude system previously described is
attributed to him, as is the discovery that Earth wobbles on its axis like
a top, causing the position of the sun relative to the stars to shift over
the centuries. This precession now places the sun in the constellation
of Pisces on the rst day of spring, or vernal equinox, soon to shift
into Aquarius (as lauded in the famed song by the Fifth Dimension).
He is frequently credited with the invention of trigonometry and the
astrolabe, a device used to measure the positions of the stars.
The work of Hipparcos was very inuential onPtolemy, whose Almagest
was considered a seminal work in astronomy through the sixteenth cen-
tury. Working in Alexandria, Egypt between a.d. 127 and 141, Ptolemy
40 Cosmology 101
wrote extensively on a detailed geocentric model of the cosmos. He
based his model on the postulates that
The heavens are spherical and move spherically; that the earth, in gure,
is sensibly spherical also when taken as a whole; in position, lies right in
the middle of the heavens, like a geometrical center; in magnitude and
distance, has the ratio of a point with respect to the sphere of the xed
stars, having itself no local motion at all. (Munitz 1957, 106)
This was a classically Aristotelian view, holding that the heavens were
perfect and unchanging (or immutable). However, when Ptolemy
tried to tackle the very practical problem of trying to model the ap-
parent motions of the sun, moon, and planets, like other geocentric
thinkers before him he found that simple circular orbits centered on
Earth were insufcient. He expanded an early model created by Hip-
parcos, and placed each planet on a small circle called an epicycle, which,
as shown in Figure 2.1, was then attached to a larger circle, called the
deferent. Unlike Hipparcos, Ptolemy then added a further fudge factor,
stating that the earth is not exactly at the center of planetary motion,
but that instead the true center of motion was a point called the equant,
which like Earth was offset from the center of the circular deferent. By
carefully adjusting the epicycle, deferent, and equant for each planet,
Ptolemy was able to model the observed motions of the planets in the
sky. Although their apparent motion was nonuniform (sometimes mov-
ing faster or slower relative to the background stars than on average),
as seen from the equant the motion would be perfectly uniform. This
prejudice in demanding the perfection of the heavens was to prove to
be the bane of astronomy for centuries to come.
Over the next few centuries, errors crept into the predicted positions
of the planets, but these were xed by adding additional epicycles to
the model. Thus observational accuracy was achieved at the expense of
elegance and simplicity. With the fall of the Roman Empire the works of
Aristotle and Ptolemy were largely lost in Europe, but were saved by the
Islamic world, whichachieved its Enlightenment betweenthe eighthand
tenth centuries. In fact, the name now used for Ptolemys work comes
from an Arabic term meaning greatest. When the Dark Ages ended in
Europe and the great works of the past were rediscovered, the geocentric
models of Aristotle and Ptolemy were embraced by the Catholic Church
as being in line with Christian doctrine. It was this alignment with the
powerful religious institutions of Europe which made the overthrow
of the erroneous geocentric tradition a lengthy processnamely the
Copernican Revolution.
The History of Cosmology 41
Figure 2.1 The Geocentric Model of Ptolemy
THE COPERNICAN REVOLUTION
Nicolaus Copernicus (14731543) was a canoninthe Catholic Church
inPoland, and evenearned a degree incanonlaw. Based onobservations
of the night sky and the continuous errors which had crept into Ptole-
maic predictions of planetary positions over the centuries, he proposed
a heliocentric model of the heavens. As laid out in his privately circu-
lated, brief work, Commentariolus (circa 1514), the Copernican model
was based on several axioms:
1. Stars appear to rise and set because of the rotation of the earth.
2. Earth, like all the planets, revolves around the sun in a circular
orbit.
3. The apparent backward motion shown by some planetscalled ret-
rograde motionis an optical illusion caused by the faster moving
Earth passing out the slower moving outer planets, such as Mars.
4. The lack of observable parallax is expected because the stars are
simply too far away.
A much longer treatise, De Revolutionibus Orbium Caelestium (On the
Revolutions of the Heavenly Spheres), remained unpublished until
just before his death. Georg Joachim Rheticus, a former student, took
charge of the publication, but the nal details fell to Lutheran theolo-
gian Andreas Osiander. In a controversial, unsigned preface, Osiander
42 Cosmology 101
attempted to appease the majority of readers, who would generally as-
cribe to the geocentric universe (partially for religious reasons). He
wrote,
it is not necessary that these hypotheses shouldbe true, or evenprobable; but
it is enough if they provide a calculus which ts the observations. . . . Let no
one expect anything inthe way of certainty fromastronomy, since astronomy
can offer us nothing certain, lest, if anyone take as true that which has been
constructed for another use, he go away from this discipline a bigger fool
than when he came to it. (Hawking 2002, 78)
Osianders preface appears even more an affront when compared to
the preface Copernicus himself had meant to include, which deantly
accosted possible detractors:
Perhaps there will be babblers who, although completely ignorant of math-
ematics, nevertheless take it upon themselves to pass judgment on mathe-
matical questions and, badly distorting some passages of Scripture to their
purpose, will dare to nd fault with my undertaking and censure it. I dis-
regard them even to the extent of despising their criticism as unfounded.
(Singh 2004, 43)
As Copernicus had predicted, there were those who attacked his
model on religious grounds. It was feared that a heliocentric universe
would remove humans from a place of centrality (as Gods special cre-
ation) and lead to the heretical view that humans were merely one part
of the natural world. Martin Luther himself attacked the Copernican
model, remarking that The fool wants to turn the whole art of astron-
omy upside-down (Singh 2004, 40).
For religious as well as scientic reasons, the Copernican model was
slow to be accepted. Not only was the idea of a moving Earth hard
to swallow for many people, but the model didnt predict the motions
of the planets with any better precision than the geocentric model. In
fact, Copernicus had to utilize small epicycles in order to achieve a
reasonable match with the observed heavens. The basic problem was
that Copernicus had fallen prey to the same prejudice which Aristo-
tle had held supremenamely the perfection of the heavens and re-
liance on circular motion. Copernicus wrote that we must observe that
the Universe is spherical. This is either because that gure is the most
perfect . . . or again because all the perfect parts of it, namely, Sun, Moon
and Stars, are so formed (Munitz 1957, 152153). So long as astronomy
The History of Cosmology 43
was a slave to this prejudice, attempts to discern the true motions of the
planets were doomed to failure. But Copernicus model was an improve-
ment over the geocentric model of Ptolemy and his successors in other
ways which were not immediately appreciated. Not only did it explain
the lack of observed parallax and provided a natural explanation for
retrograde motion (as opposed to the geocentric practice of tweaking
the various circles to mimic the effect), but it allowed for the calculation
of the distances of the planets in order from the sun, at least in units of
Earths orbital size (the astronomical unit).
KEPLERS LAWS OF PLANETARY MOTION
One famous observational astronomer who was troubled by Coperni-
cus model was Tycho Brahe (15461601). The Danish astronomer was
kidnapped by his uncle while still a toddler, under the excuse that Ty-
chos father already had two sons. This proved fortuitous for Tycho, who
received the best schooling available to nobility. He became interested
in observational astronomy as a teenager, after viewing a solar eclipse.
Unfortunately, less positive pursuits also engaged his time, and in 1566
he lost the tip of his nose in a sword duel. The wound never healed
properly, and Tycho took to wearing a metal replacement for the rest of
this life.
Tycho achieved fame through his meticulous observations of a new
star, which he dubbed a nova, in 1572. By observing the position of the
object relative to the other stars as seen from several locations, he found
that like the distant stars it did not seem to shift, or exhibit parallax. But
the moon, as seen simultaneously from different locations, does show a
shift in position relative to the stars. Tycho had therefore shown that this
nova was far beyond the moon, and must, in fact, be among the known
stars. The appearance of this previously unseen star seemed to y in the
face of the Aristotelian idea of the perfection and immutability of the
heavens, but its existence could not be denied, as it could be seen by
anyone bothering to gaze skyward.
King Frederick II of Denmark gave Tycho the island of Hven in
1576, where the astronomer built his observatory, Uraniborg, equip-
ping it with the nest naked-eye instruments for measuring the stars. For
20 years, Tycho made precise measurements of the motions of the plan-
ets. He came to appreciate some aspects of the Copernican model, but
took issue with the concept of a moving Earth as well as its seeming
contradiction with a literal translation of Scripture. In response, Tycho
proposed a compromise model in 1588, which had the sun, moon, and
44 Cosmology 101
stars orbiting Earth, but the planets orbiting the sun. This Tychonic
hybrid model achieved some level of popularity for several decades.
Following the death of his patron, Tycho received a position in Prague
as Imperial Mathematician in the court of Emperor Rudolph II in1599.
It was here that he took Johannes Kepler (15711630) as an assistant.
Kepler, a mathematics teacher in Graz, had come to the attention of
Tycho as the author of Mysterium Cosmographicum. A year later, Tycho
died, reportedly as the result of a drinking incident. Kepler succeeded
his mentor in the court, and by his own admission, quickly took advan-
tage of the absence, or lack of circumspection, of the heirs, by taking
the observations under my care, or perhaps usurping them (Hawking
2002, 631). Tychos legacy of observations afterward became the basis of
Keplers famous Rudolphine Tables of planetary positions.
Keplers main project, begun at the suggestion of Tycho, was an at-
tempt to explain the peculiarities in the observed motion of Mars. Like
astronomers before him, he was handicapped for years by a conviction
that the heavens must be based on perfect, uniform, circular motion. It
was only after attempting a noncircular shape, namely an ellipse, that he
cracked the problem wide open. The result was Keplers Three Laws of
Planetary Motion, whichhave ostensibly not changed since their original
publication:
1. The orbit of a planet (including Earth) is an ellipse, with the sun
at one focus. The other focus is a point of empty space. For those
not familiar with the geometry of an ellipse, the two foci are points
on either side of the center, along the wide side of the ellipse.
2. A line connecting a planet to the sun sweeps out an equal area
of space in an equal amount of time at any point in its orbit.
As a result, a planet moves fastest along its orbit when nearest
the sun, and slowest along its orbit when farthest from the sun.
Note that this clearly contradicts the Aristotelian idea of uniform
motion.
3. There is a simple relationship between the period of the orbit (P)
and a planets average distance from the sun (a), such that P
2
=
a
3
, where P is measured in years and a in au.
The rst two laws appeared in The New Astronomy (1609), while the
third appeared later, in The Harmony of the Spheres (1619). Keplers laws
were empirical; that is, strictly based on observations. Not only did they
explain Tychos decades of observations of Mars, but they could pre-
dict the future motions of the planets with greater precision than any
The History of Cosmology 45
previous model. The one obvious weakness in Keplers model was a lack
of theoretical underpinning. What was the force that kept the planets
orbiting the sun? Gravity, as the force we understand today, had not
been conceptualized at this time. Kepler proposed that the sun emitted
a type of magnetic force which interacted with the magnetic elds of
the planets to produce the orbits. Therefore, while Kepler had come
upon essentially the correct model of planetary motion, it took further
advances in theory, observation, and most importantly, technology, to
convince the astronomical community.
GALILEO, NEWTON, AND THE TELESCOPIC REVOLUTION
As previously noted, the rst astronomer to use a telescope in an ob-
serving program was Galileo Galilei (15641642). The Italian professor
of mathematics and physics rst used a crude refractor made of two
eyeglass lenses. He later learned how to grind his own lenses, and suc-
ceeded in making an instrument which magnied twenty times. Among
his important observations were craters and mountains on the moon
and sunspots, both of which contradict the ancient idea of the perfec-
tion of the heavens. By monitoring sunspots, an experiment which must
be done with extreme care to prevent blindness, he showed that the sun
rotated like Earth, only much more slowly (about once a month). He
was able to resolve the Milky Way into individual stars, and found that
the planets were not point-like as the stars were, but instead appeared
as small disks. This agreed with the prediction of Copernicus that the
stars were much farther than the planets.
Two additional observations greatly contributed to Galileo embracing
the Copernican concept. Firstly, he found that the planet Venus showed
a full range of phases like the moon, an observation which cannot be
explainedby a geocentric model, no matter the amount of tweaking. Sec-
ondly, he discovered four moons orbiting Jupiter, which demonstrated
that the commonly held belief that if Earth moved, the moon would y
off into space, was clearly wrong. It also proved that Earth was not the
center of all motion in the universe. His discoveries were published in a
pamphlet entitled Sidereus Nuncius (Starry Messenger) in 1610.
Galileo was fully aware of the religious concerns about removing Earth
from the center of the universe. Several years after Sidereus Nuncius, he
circulated a series of letters that argued against a literal interpretation
of the Bible when it came to astronomical topics. Soon after, in 1616, the
Vatican issued an edict against the teaching of the Copernican doctrine.
Galileo openly deed that edict in 1632 with the publication of Dialogues
Concerning the Two Chief World Systems. WritteninItalianrather thanLatin,
46 Cosmology 101
it was widely read, and its ridicule of the Aristotelian universe and open
support for Copernicus set him on a collision course with the Catholic
Church. In 1633 he was put on trial by the Inquisition for heresy, and
was found guilty. He was forced to recant his support of the Copernican
doctrine, and spent the rest of his life under house arrest. In 1979, Pope
John Paul II called for a commission to reevaluate Galileos trial, with the
result that in 1983 it was announced that Galileo should not have been
condemned. The commissions report was ofcially accepted by John
Paul II in 1992. Galileo is generally considered to be the last astronomer
to be harassed for espousing that Earth orbits the sun.
The same year as Galileos death, Isaac Newton (16421727) was born
in Lincolnshire, England. He spent 4 years as a student at Trinity Col-
lege, but returned to his family home when the university closed in 1665
for 2 years due to the plague. Temporarily unable to continue with for-
mal studies, Newton set to experimentation on his own, later resulting
in many important discoveries. Among his impressive list of contribu-
tions to science and mathematics are the reecting telescope, studies
of the spectrum of light, and calculus. His most famous written work,
Philosophiae Naturalis Principia Mathematica (The Mathematical Princi-
ples of Natural Philosophy), or Principia, as it is commonly known, was
not published until 1686. Newton was generally miserly when it came to
sharing his discoveries with the wider scientic community, and Principia
was produced through the insistence of Edmund Halley, who had used
Newtons techniques to determine the orbit of the comet that now bears
his name.
The rst of the two most widely known topics included in the Principia
is Newtons Three Laws of Motion, which still form the basis of the study
of mechanics and motion, and deservingly acquire their own chapter in
most introductory physics textbooks:
1. A body in motion remains in motion, and a body at rest remains so,
unless acted upon by an outside force.
2. The force required to change the motion is proportional to both
the mass of the object, as well as the size of the change in motion
(or acceleration). In mathematical form, F = ma.
3. Every action is accompanied by an equal and opposite reaction.
The second vital topic is the law of universal gravitation, which states
that all objects in the universe attract all other objects through a force
that decreases in strength as the distance between them increases via
an inverse-square relation (just as in the case of the intensity of light).
The History of Cosmology 47
If the moon were suddenly moved twice as far as it currently orbits,
the gravitational force between it and Earth would drop to one-quarter
its current value. The gravitational force also directly depends on the
masses of the objects in question. If the moons mass were to suddenly
double, the gravitational force between the earth and moon would also
double. In fact, using the law of universal gravitation, the three laws
of motion, and calculus, Newton was able to derive Keplers laws from
rst principle, an exercise routinely repeated by undergraduate physics
majors to this day.
With the law of universal gravitation, Newton completed the Coper-
nican Revolution. The motions of the planets were understood to be
caused by laws of nature, and the heavens worked by the same mundane
rules that governed falling bodies on Earth. But the theological tension
surrounding cosmology had not ended. Newton himself held unique
religious beliefs which had their root in Unitarianism. He wrote in the
Principia that there existed absolute, true, and mathematical time, of
itself and from its own nature, [that] ows equably without relation to
anything external, and by another name is called duration. . . . Absolute
space, in its own nature, without relation to any thing external, remains
always similar and immovable (Munitz 1957, 202). Absolute time and
space were intimately connected with God in Newtons mind. In an
unpublished manuscript, he wrote that an innite and eternal divine
power . . . extends innitely in all directions (Harrison 2000, 56). In
response to the idea that gravity could not act through empty space,
he argued that empty space is actually lled with spirit, and those who
denied the possibility of empty space actually denied the existence of
God.
Newtons thoughts concerning the universe as a whole were also
shaped by a series of letters exchanged with Richard Bentley, a cler-
gyman. When asked what would happen if a nite universe were lled
with evenly spread matter, Newton answered that it would naturally col-
lapse into a single spherical mass under the inuence of gravity, but
that if space were instead innite, the matter would clump together into
individual masses, like stars in the observed universe. This suggested
stability t with Newtons concept of a static and unchanging universe.
He believed that if there was a large-scale motion there would have
to be a center of that motion, something he disavowed. Unfortunately,
Newton was wrong on both accounts. An innite, static universe is an im-
practical balancing act, and is inherently unstable. Secondly, large-scale
motion does not require a physical center. Despite this instability being
pointed out by Bentley, Newton remained convinced that the universe
48 Cosmology 101
was static and innite, in a state of perfection reecting his concept of
the divine. In fact, he believed that it was the hand of God which kept
the universe in balance. This deep-rooted belief in a static universe was
to plague cosmology through the beginning of the twentieth century,
much as the perfection of the heavens had done so in the previous
millennium.
WHY IS THE NIGHT SKY DARK?
One of the ramications of an innite, eternal, static universe was
known before Newtons timenamely an apparent paradox. If there
were innite stars spread throughout innite, eternal space, then any-
where one would look in the sky, ones eye should intersect with the light
from some distant star. Therefore the night sky should not be dark, but
instead have an overall glow. This was not observed, hence the paradox.
Perhaps the rst clear articulation of this problemwas given by Johannes
Kepler, in his pamphlet written as a response to Galileos work, called
Conversation with the Starry Messenger:
You do not hesitate to declare that there are visible over 10,000 stars. The
more there are, and the more crowded they are, the stronger becomes my
argument against the innity of the universe. . . . [I]f they are suns having
the same nature as our sun, why do not these suns collectively outdistance
our sun in brilliance? Why do they all together transmit so dim a light to
the most accessible places? (Harrison 1987, 50)
Kepler was using this paradox as an argument against the innity of
the heavens. This was again articulated in Epitome of Copernican Astronomy
(1618), his nal major work, where he offered that the nite heavens
were enclosed and circumscribed as by a wall or a vault (Ibid.). A
century later, Edmund Halley attempted to solve the problem of the
dark sky by claiming that most stars in the innite universe are simply
not visible because their light is too feeble to reach us. Therefore, he
erroneously believed, we would see a dark sky.
A more correct analysis was done in 1744 by Swiss astronomer Jean-
Philippe de Ch eseaux, who believed that the solution was that most of
the distant starlight was absorbed by intervening clouds in space. This
solution was offered again in 1823 by German astronomer Heinrich
WilhelmOlbers, in his article On the Transparency of Space. Although
he discussed the problems with Keplers argument, he did not mention
de Ch eseauxs analysis, and in fact came to an identical solution. It is
because of the widespread readership of Olbers article, and his faulty
The History of Cosmology 49
citation of previous literature, that the problembecame known as Olbers
paradox. Although this solution became popular, it was easily proven
ineffective. John Herschel wrote in The Edinburgh Review in 1848 that if
interstellar material absorbed starlight, it would eventually become hot
itself and then radiate light from every point at every instant as much
heat as it receives (Harrison 1987, 99). This critique was reinforced by
William Thomson (Lord Kelvin), several decades later.
The correct explanation was actually proposed by German as-
tronomer Johann Heinrich von M adler in 1861. If one gives up the
notion of an eternal universe, then there would have only been a nite
time for light to reach our eyes. For example, if the universe were 10 bil-
lion years old, then only the light from stars within a shell of radius 10
billion light years in any direction could have reached us, no matter the
actual size of the entire universe. This limiting distance is called our
horizon. While this certainly contains a large number of stars, that num-
ber is not innite, therefore one can show mathematically that the night
sky would appear dark. Although von M adler had essentially solved the
problem, compared to the astronomical understanding of his day, his
proposal was nothing more than speculation. A theoretical basis for his
suggestion would not come for several decades.
THE PROBLEM OF THE NEBULAE
Ancient cultures stared at the hazy band of light arching fromhorizon
to horizon and explained it as the backbone of the night, or spilled milk
from a goddess breast. Modern sky watchers know it as the Milky Way,
our home galaxy, one of millions now known in the wider universe. The
road to our current perspective has not been a smooth one. The rst
detailed model of the Milky Way was proposed by English astronomer
Thomas Wright in his 1750 book An Original Theory of the Universe. He
pictured the Milky Way as a huge disk or sphere of stars, which includes
the sun. These stars orbitedthe center of the Milky Waywhat he termed
a center of creationlike planets orbiting the sun. Most importantly,
he claimed that many cloudy spots just perceivable by us, increasingly
seen in ever-more powerful telescopesthe nebulaemay be external
creations, bordering upon the known one, too remote for even our
telescopes to reach (Macpherson 1919, 329).
Immanuel Kant, the famed German philosopher and scientist, read
a review of Wrights book and was greatly inuenced by it. He laid out
his cosmological viewpoint in Universal Natural History and Theory of the
Heavens (1755). Gravity was the glue that held the universe together, as
the mutual gravitational attraction of the stars for each other perfectly
50 Cosmology 101
balanced their individual motions, preventing chaos from ensuing. Like
Wright, he pictured the Milky Way as a rotating disk of stars. He also held
that the nebulae were other Milky Ways, seen at large distances. Kant
also argued for an innite universe, largely on the theological grounds
that limiting space effectively limited the grandeur of God:
We come no nearer the innitude of the creative power of God, if we enclose
the space of its revelation within a sphere described with the radius of the
Milky Way, than if we were to limit it to a ball an inch in diameter. . . . Eternity
is not sufcient to embrace the manifestations of the Supreme Being, if it
is not combined with the innitude of space. (Kragh 1996, 3)
The word nebula comes from the Latin term for a cloud, and until
several decades into the twentieth century it was used as a generic term
for any object that had a fuzzy appearance when viewed through a
telescope at low power. This includes true clouds of gas and dust (what
we today still call a nebula), as well as clusters of stars andgalaxies, bothof
which can be resolved into their component stars with powerful enough
instruments. In the eighteenth and nineteenth centuries, increasing
numbers of these objects were discovered, classied, and cataloged. The
true nature of these bodies became a subject of considerable debate,
especially those which had a spiral structure, or spiral nebulae. Today we
know these objects as spiral galaxies, such as the Milky Way and the
Andromeda Galaxy.
The rst large catalog of nebulae was created by Charles Messier, the
famed French comet hunter and observing assistant at the Marine Ob-
servatory in Paris. In the course of sweeping the sky looking for comets
(named from the Greek term for fuzzy or hairy stars), he would occa-
sionally run across other fuzzy objects which one might confuse for a
comet. He kept a running record of these objects and their locations
in the sky, and published a list of 68 such objects in 1780. A later edi-
tion included 103 objects, which today comprise the Messier Catalog.
For example, the Orion Nebula is M42 and the Andromeda Galaxy
(then called the Andromeda Nebula) is M31. The catalog is sometimes
given as including 110 objects, because several objects which Messier
probably knew about were posthumously added to the catalog. Since
these objects are all visible with an 8-inch telescope from dark skies,
they comprise the foundation for most amateur astronomers observing
programs.
The number of known nebulae soon mushroomed due to the work of
WilliamHerschel. The GermanmusicianimmigratedtoEnglandin1757
The History of Cosmology 51
as a young man to escape the Seven Years War. He became interested in
astronomy and began building his own reecting telescopes, eventually
inventing a type of instrument which, unlike Newtons original reec-
tor, did not use a secondary mirror. His sister Caroline joined him in
England 15 years later and became his observing and telescope-making
assistant and a comet hunter in her own right. The largest instrument
they constructed was completed in 1788, a behemoth 48-inch aperture
reector with a 40-ft focal length, which remained the largest telescope
in the world for nearly 60 years. Herschel became well known in 1781
when he discovered the planet now called Uranus. He was appointed
court astronomer by King George III (after whom he originally tried to
name the planet) and his astronomical career blossomed.
Two of Herschels observing projects were observing nebulae, and
mapping the structure of the Milky Way through a technique known
as star gauging. This involved counting the number of stars visible in a
standard size region of the sky (such as that visible through a paper towel
tube held up to your eye) in different directions. Assuming (incorrectly)
that all stars have the same intrinsic brightness or absolute magnitude,
those that appeared dimmer were assumed to be more distant. The re-
sult was an inuential model of the Milky Way that resembled a disk or
grindstone, with the sun predictably located at the center. He origi-
nally proposed a diameter of 1,800 pc (1785), which he increased to over
6,000 pc by 1806 (Trimble 1995, 1139). One could say that the Coperni-
can Revolution took a slight step backward, because even though Earth
wasnt the center of the solar system, our solar system was believed to be
the center of the known universe.
In terms of the nebulae, Herschels observations also proved of lasting
importance. In a 1784 paper he reported the discovery of about 500
new nebulae, and also announced that he had been able to resolve
most of the objects in the Messier catalog into individual stars. This
suggested that these nebulae were actually star systems located at great
distances, and some could even outvie our Milky Way in grandeur
(Macpherson 1919, 329330). The term island universe was coined to
describe such large extragalactic systems. By 1786 he had published a
catalog of 1,000 nebulae, followed by a second catalog of 1,000 more
nebulae in 1789, and a nal catalog of 500 additional objects in 1802.
Herschels son, John, became an astronomer of note, and published a
cumulative catalog of over 5,000 nebulae and star clusters in 1863. As
William observed more nebulae, he found that there were some which
could not be resolved into individual stars, even with his large telescope.
He began to question the island universe hypothesis, and by 1811 seems
52 Cosmology 101
to have decided that most of the nebulae were in fact objects within the
Milky Way.
William Parsons, more commonly known as Lord Rosse, took tele-
scope making to another level in 1845 with the completion of a 72-inch
aperture reector with a 54-ft focal length. The so-called Leviathan
of Parsonstown was build on the grounds of Lord Rosses family cas-
tle in Ireland, and remained the largest telescope in the world until
1917, when the 100-inch Hooker telescope opened at Mount Wilson in
California. One of the rst discoveries made with the Leviathan was the
spiral structure of M51, the Whirlpool Nebula. Within ve years, Lord
Rosse reported the discovery of 14 other spiral nebulae, although some
of these are now known to be spurious. Around this time, the island
universe hypothesis enjoyed renewed support, partly through Kosmos,
the popular 1845 work by famed German geographer Alexander von
Humboldt. In addition, Lord Rosse was able to resolve even more neb-
ulae into individual stars, leading John Herschel to admit in 1849 that
it might be theoretically possible to resolve all nebulae.
Around this time, spiral nebulae became the focus of cosmological
debate and study. In 1852, Stephen Alexander, a Professor of Astron-
omy and Mathematics at what became Princeton University, published
a lengthy treatise on the classication of nebulae. He claimed that the
Milky Way is, in fact, a spiral nebula, with up to four spiral arms. If the
spiral nebulae were similar to the Milky Way in shape, then it seemed log-
ical for them to indeed be separate Milky Ways outside of our galaxy.
Another important fact which set spiral nebulae apart from other types
of nebulae and star clusters was their location in the sky. They tended
not to be found within the plane of the Milky Way (called the zone of
avoidance), but instead were numerous when looking out of the disk of
the galaxy, toward the galactic poles. The preponderance of evidence
seemed to point in the favor of the island universe hypothesis, and for a
time it gained increasing support in the astronomical community.
But dissenting voices remained. Inuential astronomer William
Huggins found that a nebula in Draco had an emission spectrum, char-
acteristic of a cloud of thin, hot gas, rather than stars. He announced
in 1864 that the riddle of the nebulae was solved. The answer, which
had come to us in the light itself, read: Not an aggregation of stars, but
a luminous gas (Harrison 1987, 115). He went further, proposing that
the spiral nebulae are also clouds of gas and not made of stars. The
pendulum thus swung back in favor of the Milky Way as the only large
star system in the universe. By the turn of the twentieth century, many
astronomers accepted that all nebulae were contained within the Milky
The History of Cosmology 53
Way, and that the sun occupied a central or near-central position within
the galaxy. In 1901, Dutch astronomer Jacobus Kapteyn used star counts,
distances, and motions, to develop an updated disk model for the Milky
Way (which he believed to be the only galaxy). The Milky Way was taken
to be some 9,000 pc wide and 3,000 pc thick, with the sun located near
the center. The inuence of this model can be seen in Agnes Clerkes
popular-level book, The System of the Stars (1905):
The questionwhether nebulae are external galaxies hardly any longer needs
discussion. It has been answered by the progress of research. No competent
thinker, with the whole of the available evidence before him, can now, it
is safe to say, maintain any single nebula to be a star system of co-ordinate
rank with the Milky Way. (Crowe 1994, 214)
Fortunately for the progress of astronomy, the self-correcting nature
of science was about to turn this smug but erroneous opinion on its ear.
THE GREAT DEBATE
Despite the pronouncements of Agnes Clerke, not all astronomers
had abandoned the island universe hypothesis. By the turn of the twen-
tieth century, it had spawned two versions: in the rst, the spiral nebulae
were thought to be objects outside the Milky Way but much smaller
than it (truly islands), while in the second, the spiral nebulae were
considered Milky Ways of their own (or comparable galaxies). In his
1919 popular-level reviewof the status of cosmology, Hector Macpherson
described it in this way:
In the present state of our knowledge we may compare our galactic systemto
a continent surrounded on all sides by the ocean of space, and the globular
clusters to small islands lying at varying distances from its shores; while the
spiral nebulae would appear to be either small islands, or else independent
continents shining dimly out of Immensity.(1919, 334)
Not all the evidence usedagainst the islanduniverse hypothesis turned
out to be good science. In 1916, Adriaan van Maanen of Mount Wilson
Observatory announced that he had photographic evidence of rotation
in M101, a spiral nebula seen face-on. Between 1921 and 1923 he pub-
lished similar evidence for six other spirals. The problem was not that
the spirals were presumed to rotate, but the size of the measured motion.
If the spiral nebulae were outside the Milky Way, the observed rotations
would correspond to unnatural velocities, some larger than the speed of
54 Cosmology 101
light! This was taken by some astronomers as irrefutable evidence that
the spiral nebulae could not be distant Milky Ways. However, as we shall
see, these measurements would later be proven to be erroneous.
One astronomer greatly inuenced by Van Maanens supposed ob-
servations was Harlow Shapley of Mount Wilson Observatory (later Di-
rector of the Harvard College Observatory). His rejection of the island
universe hypothesis was greatly inuenced by his loyalty to his friend Van
Maanen. Shapley had done his own calibration of the period-luminosity
relation for Cepheid variables, and had used it, and other methods, to
estimate the distances to the globular clusters visible in our galaxy. From
this he found that they occupy a spherical halo surrounding the galaxy,
centered around a point in the constellation of Sagittarius (where the
visible gas and dust are noticeably thicker than average). Shapley there-
fore removed humans from a point of centrality within the galaxy, as
the sun was now just another star orbiting at what Shapley estimated to
be 20,000 pc from the center. In addition, his observations suggested a
diameter for the Milky Way of about 60,00090,000 pc, 10 times larger
than previous estimates. If the Milky Way were such an immense ob-
ject, and the spiral nebulae were other Milky Ways, it would mean that
they, too, would be enormous, and the universe as a whole would be
of inconceivable size. This seemed to argue against the island universe
hypothesis.
On the other side of the argument was Heber Curtis of Lick Obser-
vatory (also in California). He and others had found novae in spiral
nebulae, and in 1917 attempted to use them to estimate the distances
to these objects. Using the fact that the average maximum apparent
magnitude of novae in spiral nebulae was ten magnitudes dimmer than
those in the Milky Way, he estimated that the spirals were approximately
100 times further away than the Milky Ways novae. A single, signicant
y in the ointment was known to the astronomical community. In 1885
an unusually bright nova had been seen in the Andromeda Nebula.
Its brightness suggested that either M31 was very close, or the object,
dubbed S Andromedae (S And), was an unusual type of nova. A resolu-
tion was not achieved until 1920. Despite the problem posed by S And,
Curtis was a vocal advocate for the island universe hypothesis. In 1917
he wrote Our present evidence, so far as it goes, leads to the belief that
the spirals are composed of great clouds of stars so innitely distant that
we cannot make out the individual stars. He also correctly explained
the zone of avoidance (where no spiral nebulae could be seen in the sky)
as being caused by a great ring of absorbing matter somewhat like those
which are found in certain edgewise spirals (van den Bergh 1988, 9).
The History of Cosmology 55
Curtis also realized that such absorbing material would dim the novae
seen in the spiral nebulae, making them seem more distant than they
would otherwise appear, but had no denite suggestion as to how to
account for this effect. He described the island universe hypothesis in
general as a wonderful, a brain-staggering conception, more tremen-
dous even than any other of the mighty ideas and facts of astronomy,
that our own stellar universe may be but one of hundreds of thousands
of similar universes (van den Bergh 1988, 10).
In 1920 a public debate was arranged between Harlow Shapley and
Heber Curtis on The Scale of the Universe. The two issues under
discussion were the size of the Milky Way and whether or not spiral
nebulae were other Milky Ways. On April 26, 1920, Shapley and Curtis
gave individual talks in the Baird Auditorium of the National Museum
of Natural History in Washington, D.C., the event being later known
as The Great Debate. Afterward, both men submitted their revised
comments for publication in the Bulletin of the National Research Council.
The thrust of Curtis argument was that the Milky Way was close in
size to Kapteyns estimates, and that star clusters and nebulae, with the
exception of spiral nebulae, are contained within the galaxy. The spiral
nebulae are Milky Ways in their own right, some 150,000 pc distant
or more. Shapley held to his estimates of a much larger size for the
Milky Way, and suggested that the spiral nebulae are much smaller
than our galaxy. In addition, they might have been created within our
galaxy but then ung outward some distance (which might explain
the zone of avoidance). In future years, it was discovered that neither
man was entirely correct. Although the spiral nebulae are indeed other
galaxies, Shapleys estimate for the dimensions of the Milky Way was
grossly oversized, and Curtis too small.
THE DEVELOPMENT OF HUBBLES LAW
Two impediments to the acceptance of the island universe hypothesis
were soon addressed by Knut Lundmark of Sweden. In 1920 he used
22 novae in M31 to estimate its distance to be about 200,000 pc. While
this was too small by a factor of four, it was still much bigger than even
Shapleys bloated estimate for the size of the Milky Way. This suggested
that M31 was located a signicant distance outside the Milky Way. As
for the mysterious S And, Lundmark recognized that it represented a
different type of object, what we today call a supernova, and there was no
problem reconciling it with the distance calculated using the normal
novae. Several years later, while visiting Mount Wilson, he utilized van
Maanens equipment and could nd no evidence for rotation in the
56 Cosmology 101
same spiral nebulae van Maanen had measured. To this day it is a mystery
as to what exactly van Maanen thought he saw. In addition, Shapleys
enlarged estimate for the size of the Milky Way was corrected in the
1930s to a more modern value, when the effects of interstellar dust were
nally taken into account.
The use of novae in estimating distances to M31 prompted the young
astronomer Edwin Hubble to do the same using Cepheid variables.
Hubble had received his Ph.D. in astronomy in 1917 fromthe University
of Chicago. His thesis had been a photographic study of spiral nebulae,
and his conclusion was that they were separate galaxies from our own.
In 1923 he discovered the rst Cepheid variable in M31 using the 100-
inch Hooker Telescope at Mount Wilson, and later found more, as well
as Cepheids in spiral nebulae M33 and NGC 6822. NGC refers to J.L.E.
Dreyers New General Catalog, an 1887 revision of John Herschels catalog.
With its later supplement catalog, it listed over 9300 objects. Using
Shapleys calibration of the period-luminosity relationship, Hubble was
able to demonstrate that these objects were clearly outside the Milky
Way, but was slow to publish his results, out of respect for Mount Wilson
colleague van Maanen. Nevertheless, his results were presented, in his
absence, on January 1, 1925, at a meeting of the American Astronomical
Society. It should be noted that his distances were far too small (by
about a factor of 2.5) due to the still-incorrect calibration of the period-
luminosity relationship.
Once he had determined that the spiral nebulae were external to the
Milky Way, Hubble became interested in their motions relative to each
other. In 1912 Vesto Slipher of Lowell Observatory began measuring
the Doppler shifts (and hence radial velocities) in the spectra of spiral
nebulae and found that they were moving much faster than individual
stars in our galaxy. He also found that the majority showed a redshift,
which meant that they were receding from the Milky Way. The work was
incredibly painstaking, and it took Slipher until 1925 to obtain results
from41 galactic spectra. During this time, Hubble was able to extend his
distance measurements to farther galaxies by using the Cepheids he had
observed in nearer galaxies. He was able to use the distances of these
closer galaxies to estimate the absolute magnitudes of the brightest stars
in these galaxies as well as the overall average absolute magnitudes of
brighter galaxies. From these values he could determine the distances
to galaxies out to the Virgo Cluster using the distance modulus formula.
In a 1929 paper he plotted his distances versus Sliphers radial velocity
data for several dozen galaxies and found a linear correlation between
distances and velocities (Hubble 1929, 170). In mathematical language,
The History of Cosmology 57
H
Figure 2.2 Hubbles Law
he found that velocity = distance some constant, where, as we see in
Figure 2.2, the constant is nothing more than the slope of the graph.
This implied that the farther away a galaxy was, the faster it seemed to
be receding from us. The result was surprising on several levels, but as
Hubble (1929, 170) himself noted, For such scanty material, so poorly
distributed, the results are fairly denite. By 1935 Hubble and Milton
Humason had extended the relationship to include a hundred more
galaxies, out to distances far beyond the Virgo Cluster, and there was no
doubt that the galaxies were receding away from each other.
The island universe debate seemed to be denitively closing. But
one lingering issue remained, especially at Mount Wilson, namely van
Maanens supposed rotation of the spiral nebulae. Despite Lundmarks
inability to reproduce van Maanens observations, there were those loyal
to their Mount Wilson colleague. Hubbles own patient respect for van
Maanen nally reached an end:
They asked me to give him time; well, I gave him time, I gave him ten years.
When a speaker at the R.A.S. [Royal Astronomical Society] announced if
it were not for van Maanens measurements, Hubbles results might be
accepted, I decided to make the measurements. (Hetherington 1993, 357)
He remeasured the galaxies, searching for evidence of rotation and
as he expected found nothing. This last impediment removed, Hubble
had solved one signicant problem in cosmology, but as is often the case
58 Cosmology 101
in science, his discovery now raised several more questions. What was
the physical cause of this motion? Did it suggest that the Milky Way was
the center of the universe? Was there some other possible explanation
for the redshift other than radial motion? The relationship eventually
became known as Hubbles law, the motion of the galaxies it represents
the Hubble ow, and the constant, the Hubble constant. In an interesting
twist of vocabulary, as we shall see, the Hubble constant is really not
constant, but changes as the universe evolves. What is calculated by
astronomers is the value of Hubbles constant now (H
0
), or at any other
particular era in the universes history. By measuring the change in the
Hubble constant over time, much information can be gleaned about the
past (and future) of the universe.
3
Warped Space-time:
General Relativity and
Modern Cosmology
GENERAL RELATIVITY
For two centuries, the Newtonian model of the cosmosstatic, eternal,
and ruled by gravity, a mysterious attractive force which derived from
some property of massserved as the basis for all cosmological research.
Space and time were considered different beasts altogether, and there
existed absolutes of both against which any measurements could be
compared. With the publication of his general theory of relativity in
1915, Albert Einstein forever changed our understanding of the very
nature of space and time, utterly shattering these absolutes in favor of
aninterconnectedspace-time inwhichobservers cananddo measure time
and distance dependent on their positions in a warped background.
There is an old joke that only three people in the world understand
general relativity, and one of them is Einstein himself. While general
relativity is based on rather complex mathematics, its basic concepts and
consequences can be understood with the help of analogies. As with any
analogies, we must take care not to take them too literally, for if we
do, we risk being led to absurdities. Although we experience space and
time very differently, Einsteins theory treats them as four dimensions
of something called space-time. This can be pictured as a exible fabric
made of three dimensions of space and one of time, interwoven in
a very specic way called the Einstein eld equations. The various ways
that this fabric can be stretched and warped correspond to different
mathematical solutions of this set of equations. Like any equations, there
are two sides of the eld equations that are set equal to each other. One
side contains information on the energy and mass in that region of
space-time. The other side of the equations contains information on the
geometry of that space-timehow it is deformed by the presence of that
60 Cosmology 101
mass and energy. Therefore these equations relate the shape of space-
time to the amount and distribution of matter and energy present in
the model. A helpful two-dimensional analogy is a rubber sheet. When a
heavy object, such as a bowling ball, is placed into the center of the sheet,
it bends or deforms in a particular way in response to the size, shape,
and mass of the ball. Similarly, the presence of the sun deforms the
fabric of the solar system, the resulting shape determining the orbits
of the planets. In the often-quoted (and seemingly Zen) description
by physicist J.A. Wheeler, Matter tells space-time how to curve, and
space-time tells matter how to move.
If your head is swimming, imagine how the scientic community of
1915 must have felt. What Einstein was proposing was nothing less than
mind-boggling. Gravity was no longer to be pictured simply as a force
like electricity, but as an entirely new type of phenomenon. Einstein un-
derstood the revolutionary nature of his theory, and as a good scientist,
proposed three famous experimental tests which could be used to sup-
port or negate his claims: the precession of the perihelion of Mercury,
the bending of starlight, and gravitational redshift.
Astronomy textbooks draw the orbits of planets as perfect ellipses,
always returning to exactly the same point in space nearest the sun (or
perihelion) at the end of one orbital period. In reality, the tugging of the
planets on each other makes the orbits wobble, or precess. The resulting
rosette-like pattern will be familiar to readers of the authors generation
as the classic spyrograph pattern. Mercury, the closest planet to the
sun, has the largest wobble of its perihelion. In 1859, Urbain Leverrier,
who predicted the existence of Neptune due to its gravitational effects
on the orbit of Uranus, found that after taking into account all the
known planets, there was still a leftover precession of 43 seconds of arc
per century inMercurys orbit. Astronomers were at a loss to explainthis,
some going so far as to predict the existence of another planet, dubbed
Vulcan, that supposedly orbited on the far side of Earths orbit and
remained hidden by the glare of the sun at all times. General relativity
predicts that the warping of space-time by the sun should contribute
precisely the observed amount of precession to Mercurys orbit.
According to Newtonian gravity, the path of a ray of starlight should
be deected by a tiny amount as it passes close by the sun. General
relativity predicts that the amount of deection should be twice the
amount predicted by Newton. This would be detectable as a slight shift
in the relative positions of stars appearing on the edge of the sun as
compared to the same location in the sky when the sun was elsewhere.
The practical problem is, of course, that one cant see stars appearing
Warped Space-time 61
near the sun in the sky because the sun is too bright! The one exception
is a total solar eclipse. British astronomer Arthur Eddington was an early
supporter of general relativity and urged that the eclipse experiment be
done. Astronomer Royal Sir Frank Dyson sent Eddington and colleagues
to viewthe May 29, 1919, eclipse onthe islandof Principe, off the western
coast of Africa, while another team led by Andrew Crommelin viewed
the eclipse from Sobral in northern Brazil. The measurements of the
two teams conrmed that starlight was bent by the amount Einstein had
predicted. Their success is astounding if one considers that the images
on the photographic plates were only deected by a few hundredths
of a millimeter! (Einstein 1961, 146147). The announcement of the
experimental results was reported in the London Times and other widely
read newspapers, soon making Einstein a household name.
The third classical test also involves starlight, but in a different way.
Since a star warps space-time, its light has to climb out of this gravity
well in order to reach our eye. In doing so, the light loses energy, which
means its wavelength becomes longer. This is called the gravitational
redshift. For normal stars like the sun, the effect is miniscule, but for
very dense stellar corpses, like white dwarfs and neutron stars, the effect
is more important. It was rst measured by Walter Adams in 1925, in the
spectrum of the white dwarf companion of Sirius (jokingly nicknamed
the Pup, as Sirius is known as the Dog Star). A black hole can be thought
of as an object whose gravitational eld is so intense that it creates an
innite gravitational redshift, as the light cannot escape at all!
Physicists and astronomers have since found numerous other tests of
general relativity. These include the LenseThirring effect, in which the
twisting of space-time by the earths rotation affects the orbits of satellites
in a specic way. This is currently in the process of being experimentally
veried. In a textbook case of the scientic method in action, general
relativity has withstood every experimental test thrown its way, and all of
its major predictions have come to pass, with one lingering exception
gravity waves. In 1916 Einstein noted that distinctive ripples in the fabric
of space-time should be created when masses move. Earthbound gravity
wave detectors have so-far failed to directly detect gravity waves, and
satellite detectors are still in the planning stages. In 1993 Joseph Taylor
and his former graduate student, Russell Hulse, won the Nobel Prize in
Physics for indirect evidence of gravity waves. Hulse had discovered a
binary system comprised of two neutron stars, and over the years their
orbit has become smaller in just the way predicted by general relativity.
As the motion of the dead stars generates gravity waves, the pair loses
orbital energy, resulting in a 3-ft per year shortening of their orbit.
62 Cosmology 101
COSMOLOGICAL MODELS
Modern cosmology is considered to have begun in 1917, when Ein-
stein applied general relativity to the universe at large. He began his
cosmological considerations with two basic assumptions, which he de-
scribed as follows:
1. There exists an average density of matter in the whole of space
which is everywhere the same and different from zero.
2. The magnitude (radius) of space is independent of time.
(Einstein 1961, 152)
His rst assumption, that matter and energy are distributed in an even
manner, predicts that the universe looks the same at all locations and in
all directions (it is homogenous andisotropic), whichis referredtoas the
cosmological principle. This greatly simplied his eld equations, but had
no observational evidence to support it at that time. The second assump-
tion, essentially that the universe was static, eternal, and unchanging,
appeared unavoidable to Einstein (1961, 153), who thought that one
would get into bottomless speculation if one departed from it. To his
surprise, Einstein found that his equations predicted that the universe
was nonstatic, either expanding or contracting, so he added a fudge
factor called the cosmological constant, which he admitted was not re-
quired by the theory as such nor did it seem natural from a theoretical
point of view, but instead was included to prop up the assumption of a
static universe (Einstein 1961, 152).
Einsteins model universe was closed in on itself into a sphere, like the
surface of the earth (but in a higher dimension), and was nite (limited)
insize but unbounded(hadnoedge). It alsofell prey toOlbers paradox.
Light rays would circulate around the universe, returning to their initial
starting point, and over time such a universe lled with stars would
become extremely bright.
In 1917 Dutch astronomer Willem de Sitter proposed another solu-
tion to the Einstein equations, a universe which also had a cosmological
constant, but was devoid of any matter and yet was still curved. Einstein
regarded the solution with suspicion, because he thought it was impos-
sible to curve space-time without matter. He eventually came to accept
the idea as possible, but because it was empty, probably not physically
relevant. In fact, because of its emptiness, the solution might have been
considered nothing more than a mathematical curiosity if it were not
for one peculiar feature. If test particles of matter are sprinkled in a de
Sitter space, they move apart at exponential speeds, producing a redshift
Warped Space-time 63
that increases with distance. Because de Sitter originally considered his
solution as static (because of the cosmological constant), this apparent
motion seemed contradictory, and although de Sitter acknowledged this
result, he did not investigate it sufciently to be given credit for predict-
ing the expansion of the universe discovered by Edwin Hubble. Hubble
was certainly aware of this prediction of the de Sitter model, and in
his 1929 paper, allowed for the speculation that the velocity-distance
relation may represent the de Sitter effect (173).
In 1922, Russian mathematician Alexander Friedmann found a math-
ematical error in Einsteins work. In trying to prove that the universe
must be static, Einstein had committed a well-known mathematical sin,
namely dividing by a constant which could, in certain cases, be zero.
Therefore the universe was not constrained to being constant. Fried-
mann published solutions which were nonstatic, or changing with time,
including one which expanded from an initial singularity (like that
found at the center of a black hole), and another which was periodic in
time, expanding for a while then contracting. This phoenix universe
was later studied in detail by Richard Tolman in the 1930s and became
known as the oscillating universe. Although Einstein came to support
this model, many questions remained as to whether it was physically
possible.
With the publication of several papers by Friedmann, Einstein began
to see the folly of his fudge factor. In a 1923 letter to mathematician
Hermann Weyl he admitted that if there is no quasi-static world, then
away with the cosmological term (Krauss 1999, 36). Einstein later called
it the biggest blunder of my life, and in a 1947 letter to cosmologist
Georges Lematre, admitted Since I introduced this term I had always
had a bad conscience. I am unable to believe that such an ugly thing is
actually realized in nature (Perivolaropoulos 2006, 16). Before we con-
demn Einstein for making such an error, recall the scientic method.
Scientists routinely develop hypotheses which are then tested against
observations and further theoretical models. Although one can argue
that Einstein himself knew that the cosmological constant was an arti-
cial construct from the rst moment of its creation, at the time of his
calculations (19161917) there was no solid observational evidence that
the universe was other than static.
Friedmann had demonstrated that both Einstein and de Sitters
models were special cases of cosmologies allowed by general relativity,
and had suggested wider classes of solutions that satised the cosmo-
logical principle. In 1935, H.P Robertson and A.G. Walker indepen-
dently generalized Friedmanns work to create a model known as the
64 Cosmology 101
FriedmannRobertsonWalker model. The central equation related the evo-
lution of the universe in size (including, for example, the current expan-
sion shown in Hubbles law) to the density of matter and energy in the
universe, and the curvature of the universe (as well as any cosmologi-
cal constant, in the most general case). In this model, whose analogies
are represented in Figure 3.1, the universe has three overall (global)
curvatures or geometries:
1. Euclidean or FlatThe universe can be pictured (in two dimen-
sions) as the surface of a at rubber sheet. As we learned in high
school geometry class, two lines (including light beams) which be-
gin parallel remain parallel no matter how far they travel. Euclid
also demonstrated that the sum of the interior angles of a trian-
gle equals 180 degrees. This geometry can be thought of as having
zero curvature. This is the simplest cosmological model that actually
has matter, and it is strangely referred to as the Einsteinde Sitter
model, despite the fact that it differs from the models proposed by
either cosmologist.
2. Spherical or ClosedThis is similar to Einsteins original cosmology,
in that the universe is pictured (in two dimensions) as the surface
of a sphere. Lines or light beams which begin parallel will eventu-
ally merge (just as lines of longitude on the earth are parallel at
the equator but meet at the poles). A triangle drawn on a sphere
is distorted (bulges), such that the sum of the interior angles is
greater than 180 degrees. This geometry has a positive curvature
(as it curves in on itself ).
3. Hyperbolical (Saddle) or OpenIn this case the universe ares
out like the center of a saddle, or a Pringles
TM
potato chip. Lines
or light beams which begin parallel are out as well, and a triangle
drawn in the center of the saddle is pinched, so the sum of the
angles is less than 180 degrees. This geometry is said to have a
negative curvature (the opposite of curving in on itself, or the
opposite of a sphere).
These three geometries correspond to three different fates for the
universe, depending on the precarious balance between the attraction
(pulling) of gravity and the repulsive (pushing) pressure of the hot early
universewhat we today call the big bang. If the density of matter in
the universe is below some critical value, nothing can counteract this
initial push, and the universe will expand forever, although slowing its
rate of expansion somewhat as time goes on. If the density of matter is
above the same critical value, then the collective self-gravitation of the
universe will counteract the expansion, and the universe will eventually
Warped Space-time 65
Figure 3.1 Three Possible Geometries of the Universe
collapse into what is called the big crunch. There is a possibility it could
then reexpand, as in the case of Tolmans oscillating universe model.
If the density of matter in the universe is precisely the critical value,
then gravity and the big bang would be balanced, and the expansion
of the universe would slow to a crawl over many billions of years, but
never actually stop. This should remind the reader of the old story of
Goldilocks and the Three Bearsone universe is too light (open), one
is too heavy (closed) and one is just right (at).
The special just right value for the density of matter inthe universe is
called the critical density, and cosmologists usually discuss it in terms of a
ratio of the current density of matter (and energy) to this critical value.
Therefore, in a at universe the actual density is exactly equal to the
critical density, so the ratio of the two (called omega, the last letter
of the Greek alphabet) is precisely one. In a closed universe, the actual
density is greater than the critical density, so is greater than one, and
in the open universe the actual density is less than the critical density,
66 Cosmology 101
Figure 3.2 The Expansion and Fate of the Universe
so is less than one. Therefore, if we could observationally determine
the density of matter and energy in the universe, we would theoretically
know its fate, assuming the value of the cosmological constant (if other
than zero) is also understood.
This last caveat is because the relationship between geometry and
destiny is more complex for the case where the cosmological constant is
not zero. The original use of the cosmological constant to create a static
universe dependedona perfect balance betweenthe attractionof gravity
and the repulsion created by the cosmological constant. But any hope at
balance would only be possible (albeit not probable) if the cosmological
constant is small and positive (greater than zero). As pictured in Figure
3.2, if it is large and positive it will not only counteract the tendency of
gravity to pull the universe together, but the universe will expand forever
and can actually accelerate as time increases, so long as the curvature is
at or negative. Aspherical curvature andpositive cosmological constant
is Einsteins original model (without the articial constraint that there
be static balance), and depending on how things are ne-tuned, can
expand forever or eventually recollapse. If the cosmological constant is
negative in value (less than zero), the opposite occursthe universe is
doomed to collapse in on itself, regardless of the curvature.
We now return to several unresolved questions, the rst of which
being the reason for the dark night sky. Recall that in 1861 von M adler
Warped Space-time 67
had argued that if the universe had a denite beginning in time, there
would be a natural resolution for Olbers paradox, because we would
only receive the light from a nite number of stars. In a modern formu-
lation of the problem, we must obviously replace stars with galaxies,
but the concept is identical. There has been some lingering confusion,
however, with some modern textbooks giving equal credit for the dark
sky to the nite age of the universe (and its galaxies) and the expansion
of the universe. The argument is that since the universe is expanding,
light is redshifted to wavelengths too long to see, thus making the sky
darker than it should be. In an attempt to resolve the issue once and for
all, in 1987, physicists Paul Wesson, K. Valle and R. Stabell did a detailed
analysis of the contributions of both the nite lifetime of the universe
and its expansion to solving Olbers Paradox. Their results showed that
the expansion only dims the light by a factor of about two, whereas the
fact that the light from a nite number of galaxies has had time to reach
us clearly plays the bulk of the role. In the words of the authors, there
can hardly be another astronomical subject that is so fundamental in
nature yet so widely misunderstood (Wesson, Valle, and Stabell 1987,
606).
This is also an appropriate time to clear up several common miscon-
ceptions about the expansion of the universe. Firstly, the galaxies are
not receding away from us, but away from each other. We are not the
center of the universe, nor does the concept of a geographic center have
any place in modern cosmology. The universe is not expanding into
something (such as more space). In the most literal sense, space itself
is expanding, and as it does it carries the galaxies along for the ride. If
this is the case, then what we have been calling the redshift really isnt
the Doppler effects redshift, but some effect that appears similar. This
is certainly true, but the difference involves more mathematics than you
really want to know, so we will pretend I havent revealed that little secret
to you. It is still true that the redshift is caused by the expansion of the
universe, and measuring the redshift of different galaxies tells us how
fast the universe is expanding at any given epoch during its life. It is also
true that although objects within the universe cannot move faster than
the speed of light, the fabric of the universe can stretch faster than light
can travel. This does not violate any of Einsteins calculations, but means
that there will be objects beyond our visible horizon that we will prob-
ably never be able to see. Some people consider these regions as other
universes, since they cannot communicate with us. The daring reader
is directed to the article by Max Tegmark listed in the bibliography for
more information.
68 Cosmology 101
With all these possible cosmological models allowed by general rel-
ativity, is it possible to know with certainty which model describes the
universe we currently inhabit? As part of that process of discovery, the
serious implications of cosmological models that have singular begin-
nings led to the development of perhaps the most famous cosmological
model of allnamely, the big bang.
THE BIG BANG AND STEADY-STATE THEORIES
Lem aitre and the Primeval Atom
Although Einstein began to regard the cosmological constant as noth-
ing more than a scientic regret, other cosmologists continued to ex-
plore cosmological models whichutilizedit. One important example was
published by Catholic priest and cosmologist Abb e Georges Lem aitre.
Unfortunately, it rst appeared in French, in a lesser-known journal,
and only caught the attention of the scientic community 4 years later
when an English translation appeared. In an interesting example of
scientic provincialism, Lem aitres paper both independently restated
and extended Friedmanns model (of which he had been unaware) and
provided the rst observational evidence of what we now call Hubbles
law, 2 years before Hubbles original paper on the subject.
Lematre described the birth of the universe, from the radioactive
decay of a large atomic nucleus, thusly:
We picture the primeval atom as lling space . . . being nearly an isotope of
a neutron. This atom is conceived as having existed for an instant only, in
fact, it was unstable and, as soon as it came into being, it was broken into
pieces which were again broken, in their turn. . . . An increase in volume
resulted, the disintegration of the atom was thus accompanied by a rapid
increase in the radius of space which the fragments of the primeval atom
lled. (Munitz 1957, 343)
His model divided the history of the universe into three eras, the rst
being a rapid expansion due to the initial disintegration of his primeval
atom. The second was a coasting epoch, in which the expansion rate
dramatically slowed down, during which there was sufcient time for
stars to form. During this phase gravitys attraction was more or less
balanced by the cosmological constant. This was followed by,
the third period of accelerated expansion. It is doubtless in this third period
that we nd ourselves today, and the acceleration of space . . . could well
be responsible for the separation of the stars into extragalactic nebulae.
(Kragh 1996, 52)
Warped Space-time 69
In this last stage, the cosmological constant dominated the evolution
of the universe, driving the accelerating expansion. The astute reader
may recognize this model as the rst genesis of what later became known
as the big bang theorythe idea that the universe had a dened begin-
ning in a small, dense, hot state and has been expanding ever since.
Lematres model drew interest and criticism from the scientic as
well as religious communities. Initial arguments against it included a
disbelief that the laws of physics could possibly be reliably pushed back
to the origin of the universe. Arthur Eddington found Lematres work
important but troubling:
Philosophically the notion of a beginning of the present order of Nature is
repugnant to me. I should like to nd a genuine loophole. . . . As a scientist
I simply do not believe the Universe began with a bang . . . it leaves me cold.
(Singh 2004, 280)
Some religious gures, including Pope Pius XII, embraced it as scien-
tic proof for the Judeo-Christian origin story. Lematre himself strug-
gled with the theological implications of his theory, but came down on
the side of separate magisteria: As far as I can see, such a theory remains
entirely outside of any metaphysical or religious question (Krauss 2005,
9). It is important to note that in terms of predicting the chemical make
up of the current universe, Lematres model failed miserably, in that it
suggested that most of the universe would be made of elements near the
middle of the periodic table, such as iron. The problemwas his use of the
ssion of a (very large) atomic nucleus as the initial state. Lematre was
on the right track, but it would take a better understanding of nuclear
reactions in the early universe to derive a more realistic model.
Gamows Evolutionary Model and the Steady-State
After World War II, Russian-born nuclear physicist George Gamow,
alone and collaborating with Ralph Alpher and Robert Herman, in-
vestigated an improved model of the early universe in an attempt to
discover the origin of the chemical elements. Rather than the decay of
a primordial atomic nucleus, their model involved a hot, dense gas
dubbed ylemthat expanded and cooled. One particularly important
paper was written by Alpher and Gamow, who added the name of col-
league Hans Bethe to the byline to satisfy Gamows well-known humor.
The paper by Alpher, Bethe, and Gamow appeared in the April 1, 1948
edition of the normally humorless Physical Review, and became known as
the paper. Gamow subsequently joked that he wished that Herman
would change his last name to Delter.
70 Cosmology 101
Several months after the publication of the paper, two papers by
Cambridge scientists explaining a rival cosmology appeared in the same
volume of the Monthly Notices of the Royal Astronomical Society. The rst, by
Hermann Bondi and Thomas Gold, laid out what they called the steady-
state theory, in which the universe obeyed the perfect cosmological principle.
This stated that the universe not only looks the same everywhere and
in all directions, but the same at each instant of time. In other words,
it is eternal and largely unchanging, although individual stars can, of
course, be bornanddie. However, since Hubbles lawmeant that galaxies
are moving farther apart from each other, the density of matter in the
universe would naturally decrease with time. In order to combat this,
new material would need to be spontaneously created in order to obey
the perfect cosmological principle. This aspect of the steady-state model
was explored in Fred Hoyles paper. This would clearly violate the long-
cherished concept of conservation of matter, but would occur at such a
slow rate of speed (imagine creating a single hydrogen atom in a one
liter soda bottle every trillion years) that it would be very difcult to see
directly.
The stage was now set for another fundamental debate in cosmology,
between what Gamow called the evolutionary theory and the steady-
state theory. The steady-state appealed to those who, like Bondi and
Gold, doubted that the laws of physics seen in the universe today could
really be expected to hold under the extreme conditions demanded
in the hot, dense state of Gamows ylem, or in general wished to side-
step knotty questions about the origin of the universe. There was also
an antireligious backlash against the evolutionary cosmological model,
which reportedly was part of the motivation for the rival steady-state.
The name by which Gamows model eventually became known, the big
bang, was actually rst spoken by Fred Hoyle during one of his popular
astronomy radio shows, and was meant as a serious put-down. On the
other side, Gamow and Lematre dismissed the steady-state out of hand,
with Gamow referring to it as articial and unreal (Kragh 1996, 297).
The Rise of the Big Bang
As with any scientic debate, the nal outcome depends not on a
popularity contest or democratic election, but on the scientic method.
Observations would ultimately determine which candidate theory de-
served to be taken seriously by the cosmology community. Hubbles law
was a natural outcome of the expanding universe of the big bang, but
could be accommodated, albeit only with the inclusion of the sponta-
neous creation of matter, in the steady-state. In fact, Hoyle and certain
Warped Space-time 71
colleagues spent many years unsuccessfully trying to nd an alternative
explanation for the redshift of the galaxies. The latter model did put a
serious constraint on Hubbles law. Since the universe had to obey the
perfect cosmological principle and could not evolve on large scales,
the value of Hubbles constant could not change over time. Conversely,
the big bang model suggested that speed of the expansion should
change over time, the exact rate of change depending on the over-
all geometry of the universe. When both models were developed, the
uncertainty in the graph of Hubbles law could safely incorporate either
possibility. But by the late 1950s the data suggested that Hubbles law
did not match the predictions of the steady-state model.
Another prediction of the perfect cosmological principle and the
spontaneous creation of matter was that, just as stars of different ages
could be found everywhere in the universe, individual galaxies had to
be created at different points in the universes history. However, all
the galaxies seen within our galactic neighborhood appeared to have
similar ages. Martin Ryle, one of the pioneers of radio astronomy, led
the radio astronomy group at Cambridge, and like Bondi, Hoyle, and
Gold had worked on radar during World War II. By 1962 he had found
another violation of the perfect cosmological principle. Extragalactic
radio sources, now known to be galaxies with voracious black holes
in their centers, seemed more common the farther out in space his
instruments probed, meaning they were more common the farther back
in time one looked. Martin Ryle and others continued to nd that the
distribution of these radio galaxies and their relatives, quasars, did not
match that predicted by a steady-state universe.
Since Gamows motivation for the big bang theory was the search for
an explanation for the formation of the chemical elements, the predic-
tions of both theories should be checked against the observed composi-
tion of the universe. The big bang largely failed in its grandiose mission
to build up all chemical elements by adding neutrons to successively
heavier atoms, because there are no stable atomic nuclei with an atomic
mass of ve or eight. This essentially stops the sequence at helium.
Despite this serious limitation, the big bang was exceedingly successful
in predicting that the universe would have an overall composition of
75 percent hydrogen to 25 percent helium (by mass). The steady-state
model had hydrogen being created through the spontaneous creation
of matter, with all other elements being cooked inside stars. Demon-
strating that this was possible was one of the main reasons behind the
famous B
2
FH paper (where the H stands for coauthor Fred Hoyle of
steady-state infamy). The problem was that stellar nucleosynthesis alone
72 Cosmology 101
could not account for the quarter of the universe that was helium. In a
rather strange twist of fate, the big bang theory, when coupled with the
B
2
FH paper which had been motivated by the steady-state theory, was
very successful in explaining the observed chemistry of the universe and
the origin of the chemical elements.
The strongest evidence for the big bang (and against the steady-state)
was the theorys originally least-discussed prediction. Gamow predicted
that as the universe cooled, there would come an important transition
at which radiation (light) would no longer strongly interact with matter,
but instead would be free to move about the universe. Alpher and Her-
man calculated that there would be a resulting background blackbody
radiation currently at a temperature of around 5 K. The prediction was
largely forgotten, possibly because such a weak signal was thought nearly
impossible to detect. Intrigued by the oscillating cosmological model,
which would have a series of big bangs and big crunches, Princeton
physicist Robert Dicke set out in 1964 to detect this feeble radiation,
which, according to Wiens law, would have its peak in the microwave
section of the electromagnetic spectrum. To Dickes chagrin, before he
could assemble his detector, Arno Penzias and Robert Wilson, two engi-
neers working for Bell Laboratorieslike Princeton, inNewJerseyhad
accidentally discovered this echo of the big bang (a discovery which
would earn them the 1978 Nobel Prize in physics), at a temperature
of approximately 3 K. Companion papers by both Dicke and his team
and Penzias and Wilson were published in May 1965, but to the endur-
ing annoyance of Gamow, neither mentioned that this cosmic microwave
background had previously been predicted. What is not widely known,
even among astronomers, is that the cosmic microwave background
narrowly missed discovery two decades before! In 1941, Australian physi-
cist Andrew McKellar found that a peculiar spectral line in the star
Ophiuchi was due to the excitation of molecules of cyanogen in the in-
terstellar space between the star and Earth, which had a temperature of
around 2 K.
By the summer of 1965, it was becoming nearly impossible to support
the steady-state model, and the big bang became the reigning model of
the evolution of the universe. At a 1982 meeting of the International
Astronomical Union, Russian cosmologist Yakov Zeldovich proclaimed
that the big bang was as certain as that the Earth goes round the Sun
(Rees 2000, 15). This has led to the scientically erroneous claim that
the big bang theory has been proven. There is also the equally wrong
idea that the big bang theory explains howthe universe exploded from
an initial singularity. It is more properly said that the expansion of the
Warped Space-time 73
universe from a hot, dense state of being as explained by the big bang
theory, especially its inationary revision in the 1980s, is the scientic
best t to the current accumulated observations of the universe. How
this initial state itself came into being, and whether it was ever truly a
singularity, is the regime of further speculation and other theories.
4
Measuring and
Mapping the Universe
NORMAL GALAXIES
We now return to the considerable contributions of one Edwin Hubble.
As part of his study of galaxies and their properties, he published a
classication scheme in 1926 which divided them into four classes:
ellipticals, spirals, barred spirals, and irregulars. Each class was again sub-
divided, creating a continuum of galaxy classes (minus the irregulars)
sometimes represented in a horizontal Y-shape called the Hubble tuning
fork diagram as represented in Figure 4.1. Galaxies seen face-on are far
easier to classify than those seen edge-on, and some ambiguities (and
arguments) are inevitable.
When most people think of a galaxy, spirals like the Milky Way come to
mind. This is certainly understandable, as we have seen that the spiral
nebulae were the rst objects that astronomers considered as possibly
existing beyond the Milky Way. Spirals are dened by two or more arms
that emanate from either a central nucleus or an elongated structure
called a bar. This led Hubble to differentiate between normal spirals
(sometimes just called spirals) and barred spirals. The ratio of barred
to unbarred is highly debated by astronomers, with estimates varying
from 30 to 60 percent of all spirals having central bars. Both types
of spirals were subdivided by Hubble according to three criteria: the
prominence of the nucleus (also called the central or nuclear bulge),
the tightness of the winding of the spiral arms, and the clumpiness of
the material within the spiral arms. Sa and SBa (SB denoting spiral-
barred) galaxies have prominent nuclear bulges, and tightly wound
arms that appear to have a smooth distribution of stars. Sb/SBb galaxies
have intermediate properties, and Sc/SBc galaxies have inconspicuous
bulges with open and fragmented spiral arms that are highly patchy
76 Cosmology 101
Figure 4.1 The Hubble Tuning Fork Diagram
(Waller and Hodge 2003, 11). There are also Sd galaxies (not part of
Hubbles original scheme) in which the nuclear bulge is unrecognizable
and the spiral arms hard to distinguish.
Several different populations of stars inhabit spiral galaxies, fromvery
old extreme population II stars in globular clusters to newborn popu-
lation I stars being created from the copious gas and dust in the spiral
arms. In early type barred spirals (SBa), the bars are composed of
stars, while in late type barred spirals (SBb and SBc) there is a signif-
icant proportion of gas and dust in the bar as well. Bar-like structures
are sometimes seen in irregular galaxies (such as the Large Magellanic
Cloud), which has led some authors to acknowledge another type of
spiral, the Magellanic type (Sm and SBm), which can have one arm.
Other sources classify these galaxies under the irregular category.
At the joint in the Hubble tuning fork diagram, where the two types of
spirals meet, we nd the lenticular (lens-shaped) galaxies. They resem-
ble spirals in that they have a central bulge (and sometimes a central bar
as well), as well as a truncated disk emanating from the bulge, but they
lack the spiral arms that dominate the disks of spiral galaxies. Lenticulars
without bars are called S0, while those with bars are termed SB0.
Passing through the lenticulars to the last arm of the tuning fork, we
nd the elliptical galaxies. Hubble subdivided them according to how
elliptical they appeared, with E0 galaxies being spherical and E7 highly
elongated (sometimes referred to as cigar-shaped). They are rather
Measuring and Mapping the Universe 77
featureless andtendtohave less gas anddust thanspirals, resulting infew
young stars or star-forming regions. Elliptical galaxies show the greatest
variety in size of any class of galaxies, from the humongous cD ellipticals
down to the puny dwarf ellipticals and dwarf spheroidals. The centers of
rich clusters of galaxies, where the galaxies are more densely packed, are
typically dominated by a huge cD elliptical. The name is shorthand for
supergiant-diffuse, which accurately describes their appearance. These
monsters can be dozens of times larger than our Milky Way and are
believed to be built up from numerous mergers of smaller galaxies.
Some contain evidence of multiple nuclei, the remnants of previous
galactic meals.
On the other end of the size scale exist dwarf galaxies, some no
brighter than a globular cluster. Dwarf ellipticals and their smaller and
dimmer cousins, the dwarf spheroidals, appear to be the most common
type of galaxies, making up at least 60 percent of all galaxies. Their con-
tribution might be as high as 85 percent, but it is easy to overlook small
and dim objects when doing surveys of the universe. Dwarf ellipticals
differ from the globular clusters they somewhat resemble in that they
contain a unique type of variable termed anomalous Cepheids. They
are approximately a billion years old, far younger than stars seen in
globular clusters. Dwarf spheroidals are several hundred parsecs across,
far larger than any globular cluster, but have a similar overall lumi-
nosity. Spreading the light over a much larger area makes these galax-
ies exceedingly diffuse and difcult to see. It is therefore no surprise
that dwarf spheroidal companions to the Milky Way continue to be dis-
covered.
Irregular galaxies generally lack spiral or elliptical structure, as their
name implies. They tend to be smaller than spirals, and also come
in dwarf varieties. They contain gas and dust in sufcient quantities
to have young stars and regions of ongoing star formation. They are
sometimes found close to other galaxies, leading to the speculation that
their disturbed shapes might be due to gravitational interactions among
galaxies.
Giventhat galaxies come ina variety of shapes and sizes, scientists have
searched for a connection between Hubble type and a galaxys birth and
evolution. The actual process(es) and timing(s) of galaxy formation in
our universe are areas of active research, but as strange as it may seem,
in some ways astronomers know less about the evolution of galaxies
than they do the evolution of stars or the universe at large. Fortunately,
astronomers can use look-back time to their advantage. By taking long-
exposure photographs, peering as deeply into space as possible, galaxies
can be studied at younger and younger times in their lives. For example,
78 Cosmology 101
the Hubble Ultra Deep Field (HUDF) photograph captured the images
of some 10,000 galaxies. It found that galaxies with redshifts larger than
four, corresponding to the rst few billion years of the universe, look
completely different from the Hubble classes seen in the universe today.
Clues as to how the various Hubble classes might have arisen can be
found by comparing spirals and ellipticals seen today. Ellipticals cur-
rently have very few young stars and star forming regions, suggesting
that the bulk of their stars were formed early on in the galaxys history.
By comparison, the arms of spirals contain active star forming regions,
with Sc/SBc subclasses having the most vigorous current star formation.
Galaxies with unusually high amounts of ongoing star formation are
called starburst galaxies, and are discovered by the enormous amounts
of infrared radiation this star formation spawns. In the 1980s, the IRAS
satellite discovered that starburst galaxies emit 100 times more energy
in the infrared alone than our galaxy does over all wavelengths. The
trigger for this abnormally high rate of star formation is possibly the
merger or gravitational interactions of two galaxies. If elliptical galaxies
formed through mergers early in their history, it would have triggered
such tremendous star formation that there would now be insufcient
free gas and dust for future stars.
The stars in the disk and spiral arms of spiral galaxies have regular,
orderly orbits around the center of the galaxy, assumed to be a relic from
the formation of the galaxy. Elliptical galaxies lack this orderly rotation,
and have been compared to star pilesswarms of stars that have long
since settled down into galactic orbits under one anothers gravitational
inuence (Jones and Lambourne 2004, 78). Computer models have
shown that the rate of initial rotation of the cloud presumed to form
the galaxy is key to whether a spiral or elliptical forms. If the cloud
spins quickly, most of the material will tend to form a disk before a large
number of stars form, leading to a spiral. If the cloud spins slowly, many
stars will begin to form while the cloud has a roughly spherical shape
and a discernable disk will not emerge, creating an elliptical galaxy.
Understanding how the attened disk of a spiral galaxy forms still
leaves us with one important questionhow do the spiral arms not
only form out of the disk, but remain intact billions of years after the
galaxys formation? Since spiral galaxies are not solid objects and the
inner regions can move faster than the outer portion (called differential
rotation), the spiral arms should wind up and dissipate over time. How-
ever, this is not observed to occur. The best explanation is the density
wave theory developed in the 1960s by C.C. Lin and his students at MIT.
The model is based on the premise that areas of slightly higher than
Measuring and Mapping the Universe 79
normal density will form in the disk of the galaxy, creating grooves
in the space-time fabric of the galaxy. As stars and gas clouds orbit the
center of the galaxy, they move into these gravitational troughs, slow
down, and linger there for some length of time before moving on. The
best analogy is a trafc jam, created because of rubbernecking near an
accident scene. Individual cars slow down and clump together in the
familiar bumper to bumper pattern dening the jam, but soon move
onward. The trafc jam is xed in space, but individual cars move in
and out of it. Likewise, individual stars and nebulae move in and out
of the spiral arms. While in the jam of the spiral arms, nebulae can
be compressed, triggering star formation. This explains why the spiral
arms are areas of ongoing star formation.
Computer models have not only shown that bar-like structures in
galaxies are relatively easy to make, but are in fact so easy that it is much
harder to explain spirals that lack them. Like spiral arms, it is believed
that individual stars move in and out of the bar, which itself stays xed
in its relative position within the galaxy. In keeping with these concepts,
the next topic to explore is the structure of the Milky Way itself and its
neighborhooddo we really live in a (normal) spiral galaxy?
THE MILKY WAY
As seen in our historical survey, it has been a serious challenge to
map the overall structure of the galaxy from our xed position within it,
made even more difcult by the gas and dust blocking our view in some
directions (the zone of avoidance). According to current observations,
our home galaxy is composed of about 10
11
stars, most of which inhabit
the attened disk. This disk has a radius of at least 15 kpc (15,000 pc) and
we are located about 8,500 pc from the galactic center. The stars of the
disk are divided into the thin disk component (closer to the equatorial
plane of the galaxy) and thick disk component (located just above and
below the thin disk). The majority of the gas and dust is found within
the thin disk. The disk has a slight warp near its edge, a property found
in roughly a quarter of spiral galaxies.
The sun orbits the center of the galaxy with a period of about 250 mil-
lion years. Its motion is rather complex, tracing out a rosette-like
pattern. For example, the sun moves slightly closer and farther away
from the center with an estimated period of 169 million years, and
bobs up and down through the plane of the galaxy with a period of
about 62 million years (Waller and Hodge 2003, 110). Using regions
associated with young stars and star formation, such as OB associations
and HII regions, astronomers have traced out three spiral arms in our
80 Cosmology 101
neighborhood: the SagittariusCarina arm (looking toward the center
of the galaxy), the OrionCygnus arm, and the Perseus arm. The sun
is located on the edge of the OrionCygnus arm, a structure that is
small enough to sometimes be considered merely a spur of a larger
arm.
As many as 95 percent of the disk stars are foundwithinthe thindisk, as
depicted in Figure 4.2, a region extending a few hundred parsecs above
and below the central plane of the galaxy. Stars specically residing
within the spiral arms are the youngest stars in the galaxy (population I),
with ages of 100 million years or less. The rest of the thin disk population
is older, withages between1 and10 billionyears (including the sun). The
older stars of this group are found at higher galactic latitudes (distances
above and below the plane of the galaxy), suggesting that stars tend to
be formed closer to the plane and later migrate outward, perhaps due to
gravitational interactions. Within a few thousand parsecs of the galactic
plane are found the older stars (around 10 billion years) of the thick
disk, and are called intermediate population I. It has been suggested
that these stars also formed closer to the central plane, but their orbits
were moved to higher latitudes by gravitational interactions as well. In
other words, over time the disk of the galaxy puffs out, forming several
populations.
Surrounding the galaxys disk is a halo, in the shape of a slightly
attened sphere. The exact size of the halo is difcult to measure, but
it extends out well past the disk. The most obvious citizens of the halo
are over 150 globular clusters, the majority of which are made of very
old, metal-poor stars (extreme population II). Isolated halo stars are
also found, including RR Lyrae stars, and are also old and metal poor.
The spherical shape of the halo and the swarming motion of its stars in
highly elliptical orbits well out of the plane of the galaxy, suggest that
the halo was the rst portion of the Milky Way to form, as the rapidly
rotating protogalactic cloud rst began collapsing into the disk.
Looking toward the center of the galaxy (in the general direction of
the constellation Sagittarius), the nuclear bulge comes into view. The
stars found here are old, nearly as old as those in the halo, but are
chemically distinct. Although they are normally classied as population
II stars, they have metallicities much higher than halo stars, and in some
cases have the same chemistry as the sun. The stars and HII regions in
the bulge have more highly elliptical orbits than the stars of the disk,
which led Gerard deVaucouleurs to suggest in 1964 that the Milky Way
might be a barred spiral. Further evidence has accumulated through
infrared studies, which are able to pierce through the intervening gas
Measuring and Mapping the Universe 81
Figure 4.2 The Structure of the Milky Way
and dust. The Galactic Legacy Infrared Mid-Plane Survey Extraordinaire
(GLIMPSE) found about 30 million IR sources between the sun and
the center of the galaxy. Based on their distribution, it appears that the
Milky Way has a central bar about 9 kpc long and a few kpc wide, tilted by
44 degrees tothe direct line betweenthe sunandthe galactic center. Now
that we have ascertained that the Milky Way is probably a barred spiral,
what about its Hubble subtype? Based on the size of the nuclear bulge
and the structure of the spiral arms, a Hubble class of SBbc (between an
SBb and SBc) is suggested.
The intervening gas and dust between our solar systemand the center
of the galaxy screen it fromview, at least at visible wavelengths. Whatever
lies hidden at the center of the galaxy generates a tremendous amount
of energy at infrared, x-ray, radio, and gamma ray wavelengths. The exact
center of the galaxy is believed to be marked by a strong radio source
named Sgr A

. Motions of stars near the center suggest that there is a

very compact mass of about two to three million solar masses contained
within an area of 0.001 pc. The only sensible astronomical explanation
appears to be a supermassive black hole (SMBH), with the energy being
generated by its messy eating habits. Such SMBH appear to be common
82 Cosmology 101
at the centers of galaxies, and the details of their energy generation will
be discussed in an upcoming section.
The structure of the galaxy, including the chemistry of its various sec-
tions, should be explained by any model of galactic formation. Since
the halo contains the oldest stars, it makes sense that it was formed rst.
As the protogalaxy began to collapse from a relatively spherical shape,
the rst stars formed. The bulk of the material continued to collapse
to form the disk, and star formation will occur later here, resulting in
younger stars. But what of the nuclear bulge? It contains stars nearly as
old as the halo, but with richer chemistry. Near the center of the orig-
inal protogalaxy, material collapsed and aggregated to form the bulge.
Star formation occurred very quickly here because of the high den-
sity, and the rst (massive) stars died relatively soon, rapidly enriching
the material incorporated in the next generations, leading to higher
metallicities.
THE LOCAL GROUP
Just as stars are typically found in groups and clusters, many galaxies
aggregate into a larger galactic neighborhood called a galaxy cluster.
Clusters of galaxies with fewer than 50 members are generally called
groups, as in the case of the Milky Way. Its extended family includes
some several dozen members, and was termed the Local Group by Edwin
Hubble. These 40 or so galaxies are contained within a diameter of
1.5 Mpc (million pc or megaparsec), and the intervening space between
it and the nearest groupsome three to four Mpccontains very few
galaxies. There are three large galaxies, all spirals: M31, the Milky Way,
and M33. Most of the remaining galaxies are dwarf irregulars, ellipticals,
and spheroidals. Most of these in turn cluster around either M31/M33
(which are relatively close together) or the Milky Way.
M31, the Andromeda Galaxy, is the largest and most luminous mem-
ber of the Local Group, and is considered an Sb galaxy. It is the most
distant object visible to the naked eye, and was rst recorded by al-Su
in the tenth century. It has at least 26 known satellite galaxies, as well
as ve objects which appear as extended globular clusters but in real-
ity are possibly the remains of dwarf galaxies that were gravitationally
cannibalized by their much larger family member. There is evidence of
a secondary core in M31, which could be the result of the past merger
with another galaxy. Our own Milky Way may merge with M31 in about
6 billion years.
The most massive satellite of M31 is M32, a small (but not dwarf)
elliptical. Although it has less than 1 percent of the mass of M31, it is
Measuring and Mapping the Universe 83
close enough for the two to interact gravitationally. The spiral arms on
M32s side of the Andromeda galaxy are irregular, and it is thought that
M32 was once larger, many of its stars having been ripped away by its
bigger brother. M33 is an Sc spiral and is about one-third the diameter
of M31.
Like M31, the Milky Way has a number of galactic companions (at
least 14, with new candidates continually being discovered) along with
several possible stripped core objects which are sometimes classied as
globular clusters. The zone of avoidance has made it difcult to get a full
census of the Milky Ways satellites, and low luminosity dwarf spheroidal
companions continue to be discovered. For example, SagDEG (Sagit-
tarius dwarf elliptical galaxy) was only discovered in 1994, hiding within
the zone of avoidance. A mere 15 kpc from the center of the galaxy,
it is doomed to be cannibalized by the Milky Way one day. Already
falling victim to the Milky Ways gravity is Sag DIG (Sagittarius dwarf
irregular galaxy), which lies on the far side of the galaxy. As it spills
its stellar guts across the sky in a 10 kpc stream, it adds its globular
clusters to our halo, their foreign identity revealed by their anomalous
motions. An increasing number of tidal streams, composed of material
stripped from satellite galaxies, have been found around the Milky Way
and M31, demonstrating the importance of gravitational interactions
between galaxies within clusters.
By far the most obvious members of the Local Group (outside of
the Milky Way) are the Large and Small Magellanic Clouds (LMC and
SMC). The LMC is less than half the size of the disk of the Milky Way,
and the SMCis half as large as the LMC. The LMCis alternately called an
irregular with a bar, or the prototype of the barred Magellanic spirals.
During its rst few billion years, it enjoyed an active period of star
formation, including around a dozen globular clusters and stars that
are now RR Lyrae variables. After a 7-billion-year hiatus, star formation
was violently renewed, and continues to this day. A stunning example
is the HII region called the Tarantula Nebula. The SMC is an irregular
dwarf galaxy, with more gas than its larger sibling and only one globular
cluster. These facts suggest that it is a more primitive and less evolved
galaxy, and star formation may have started off later, or more gradually,
than it did in the Large Cloud (van den Bergh 2000, 142).
The Magellanic Clouds interact gravitationally with each other, with
the larger cloud doing more damage to the smaller one. Both of these
interact with the Milky Way as well. Both Clouds have highly ellipti-
cal orbits around the Milky Way with 2 billion year periods. Although
they last passed close to the Milky Way hundreds of millions of years
84 Cosmology 101
ago, roughly 50 million years ago they were only 10 kpc apart (half the
distance they now keep from each other), and tidal disruptions were
probable. There is a bridge of gas peppered with young star clusters
connecting the clouds to each other, as well as an extended trail of neb-
ulae spread across nearly half of the SMCs orbit, called the Magellanic
Stream. Eventually both the Clouds will be devoured by the Milky Way,
as their orbits slowly spiral in toward our galaxy. It appears that violent
eating habits are the norm at the galactic level, and nowhere is this
more apparent than the cores of distant galaxies, dominated by their
voracious black holes.
ACTIVE GALACTIC NUCLEI
In 1943, Carl Seyfert noted that some spiral galaxies have unusually
bright, compact cores. Upon studying their spectra, he also found that
unlike normal galaxies, these galaxies had strong emission lines. These
Seyfert galaxies were later found to come in two varieties, Type 1, in which
the emission lines are broad, and Type 2, in which the emission lines are
narrower. The difference in the width of the spectral lines is caused by
Doppler broadening. Recall that when a light source approaches, its light
is blueshifted, and when a light source recedes its light is redshifted. If a
variety of light sources (such as stars or clouds of hot gas) are moving in
various directions, some of the light will be blueshifted and some will be
redshifted. If all the objects are emitting light due to the same element,
then the spectral line will appear widened as some of the light is shifted
one way and some the other. The wider (broader) the line, the faster
the stars or gas clouds are moving. The broad lines of Type 1 Seyferts
are caused by material moving faster than 1,000 km/sec.
Seyfert galaxies are also characterized by signicant and sometimes
swift variations in the amount of visible light and x-rays they emit. The
short time scale of these changesdoubling in brightness in only a
few hoursmeans that whatever engine is emitting the x-rays is only
the size that light can travel in a few hours (or similar to the size of
the solar system out through Neptune). A census of spirals suggests
that 10 percent of Sa/SbA and Sb/SBb spirals are Seyferts. This can be
interpreted in two ways: either only 10 percent of early-type spirals ever
exhibit Seyfert-like activity, or all such spirals can go through a Seyfert
phase for 10 percent of their current lifetime.
Elliptical galaxies can also exhibit strange behavior. In 1937, engineer
and amateur radio astronomer Grote Reber built a radio telescope in
his backyard and began mapping radio emissions from the Milky Way.
In 1944 he reported the discovery of radio sources in Cassiopeia and
Measuring and Mapping the Universe 85
Cygnus, later named Cassiopeia A and Cygnus A. The rst object was
found to be a supernova remnant, while the second was associated with
the center of an obscure galaxy, peculiar at visible wavelengths only in
that it was crossed by dark dust lanes. Cygnus A is a classic example of
a lobe-dominated radio source, where the radio emissions come from
two gargantuan regions which extend several hundred kiloparsecs into
space. Some radio galaxies show radio emission in more concentrated jets
that seem to connect the vast distended lobes to the core of the galaxy.
Radio galaxies with most of the emission coming from the jets are called
FanaroffRiley Type I (FR I) while those dominated by the radio lobes
are dubbed FanaroffRiley Type II (FR II). As in the case of pulsars, the
radio waves are synchrotron radiation generated by electrons spiraling
in powerful magnetic elds.
In 1960, astronomers set to work nding the visible counterparts to
the radio signals listed in the Third Cambridge radio catalog. It was
discovered that some sources, such as 3C 273 and 3C 48, did not seem
to coincide with a supernova remnant or galaxy. 3C 48 appeared to be
associated witha faint blue star, but stars normally do not emit signicant
amounts of radio waves. The stars spectrum was more puzzling still, as
it had very obvious and broad emission lines (also not characteristic
of stars) that did not correspond to any known elements. 3C 273 was
found to be a similar object 2 years later. Astronomer Maarten Schmidt
noticed that the unidentiable spectral lines had a familiar pattern, that
of the visible wavelengths of hydrogen, but were redshifted by such a
large amount that no one had realized their identity. From Hubbles
law, if these objects have very large redshifts, they must be located very
far away, and the fact that they could be seen as star-like objects must
mean they have high intrinsic luminosities. The term quasi-stellar radio
source, or quasar, was quickly coinedtodescribe them. The more generic
quasi-stellar object (QSO) became used because a large number of similar
objects do not have signicant radio emissions, but the term quasar is
still commonly used in popular level texts for all related objects.
For many QSOs, a faint host galaxy has been spotted, but due to their
huge distances, it is extremely difcult to see these host galaxies clearly
enough to determine their Hubble classes. All QSOs emit an unusually
large amount of energy at infrared, ultraviolet, and x-ray wavelengths,
as well as have high visual luminosities. Although QSOs were rst iden-
tied through their radio emissions, about 90 percent are radio quiet.
These radio-quiet QSOs are found in both spiral and elliptical hosts.
The broadening of the emission lines showed that material within the
QSO is moving at up to 10,000 km/sec. In this and other ways, the
86 Cosmology 101
spectra look similar to Type 1 Seyferts, leading to the comparison of
radio-quiet quasars with Seyferts on steroids (Wiita 2006, 3). The fact
that radio-loud QSOs are generally associated with elliptical and inter-
acting galaxies leads to a similar conclusion in their relation to radio
galaxies.
The extreme redshifts of QSOs and their corresponding vast distances
leads us to the conclusion that we are looking at objects as they were long
ago. Most have redshifts of two to three, corresponding to an era several
billion years after the origin of the universe, and QSOs appear to have
become more rare as the universe evolved. The most distant QSOs have
conrmed redshifts of just under seven, corresponding to an era only
800 million years into the life of the universe. Whatever engine powers
the QSOs must generate less energy with time, and possibly shuts off at
some late date. Another aspect of QSOs with cosmological importance
is how their light interacts with intergalactic space. Intervening clouds
of cool hydrogen will absorb the Lyman line as electrons jump from
the ground state to the rst excited state. But each of these clouds
is at a different distance, both from each other and from the QSO
itself, resulting in a series of closely spaced lines at slightly different
redshifts called the Lyman forest. In studying the spectra of distant
QSOs, we get free information about material spread throughout the
universe.
In 1926, German variable star astronomer Cuno Hoffmeister discov-
ered a nondescript variable star which was given the ordinary name BL
Lacertae (BL Lac for short). In 1968 it was discovered to be a strong ra-
dio source, and reclassied as a QSO. It was afterward reclassied once
more, as the prototype of a new class of objects, the BL Lac objects, which
are most often found associated with the cores of distant elliptical galax-
ies. Although they appear stellar like QSOs, these objects lack the strong
emission lines found in QSOs and are generally strong and rapidly vari-
able emitters of rays, the most energetic form of light. Their emissions
in the visible and radio sections of the electromagnetic spectrum are
also highly polarized. Some of these properties are shared by a subclass
of QSOs called optically violent variables (OVV), and they and BL Lac
objects are sometimes grouped together under the term blazar.
By the 1970s it appeared that there were a number of strange mem-
bers of the cosmological zoo who shared certain similarities, including
extreme energy output over many segments of the electromagnetic spec-
trumoriginating froma rather small area, large distance, and an associa-
tion with the cores of galaxies. Could it be that the relationship was more
Measuring and Mapping the Universe 87
than skin-deep? In 1985, R.R.J. Antonucci and J.S. Miller demonstrated
that Type 2 Seyferts are nothing more than Type 1 Seyferts viewed side-
ways through a donut, or torus, of dust. This was the beginning of the
unied model of active galactic nuclei (AGN) as pictured in Figure 4.3. At
the heart of an AGN is a supermassive black hole millions of times the
mass of the sun, similar to that found in the core of the Milky Way. This
is the surprising engine that drives the energy source, despite the fact
that nothing can escape from a black hole (at least above the quantum
level). The key is the black holes messy eating habits.
Recall that as material is drawn further into the intense gravitational
eld of the black hole, it forms a swirling mass called an accretion disk.
Gas and dust particles move so fast and collide with such energy that
the temperature rises to 1 million Kelvin, and x-rays and ultraviolet rays
are emitted. These high energy wavelengths will heat any clouds of gas
orbiting above the plane of the accretiondisk, ionizing the hydrogenand
causing them to give off broad and narrow emission lines, depending
on how fast the clouds move. If the accretion disk has an associated
magnetic eld, charged particles such as electrons will be accelerated
and produce radio waves of synchrotron radiation. The radio waves will
be emitted in two beams or jets, 180 degrees apart. A torus of dust
surrounding the accretion disk will normally shield it from direct view.
The high temperature will heat the dust and cause it to give off infrared
radiation.
In this unied model, the type of AGN we see depends on the angle
of view, the strength of the magnetic eld (which may be related to
the rotation of the central supermassive black hole), and how much
material is being cannibalized at any given time. For example, if the
AGN is viewed directly down one jet, a blazar would be seen, whereas
if it were viewed at a slight angle, the object would appear as a QSO,
and at a greater angle, it would be labeled a Seyfert. As the black hole
cleans its plate (swallows most of the material orbiting close by), its
source of energy dries up, and the AGN fades. Future feasts (perhaps
caused by a small galaxy being eaten by its larger companion) can cause
temporary are-ups, but increasingly as time marches on, the black
holes are put on meager diets. This is the current status of the SMBH
at the center of the Milky Way. This suggests that our home galaxy was
once a Seyfert galaxy or QSO, and observers on some distant planet
billions of light years away may, at this very moment, be looking in our
direction and seeing the Milky Way as it once wasa wondrous and
violent AGN.
88 Cosmology 101
Figure 4.3 The Unied Model of Active Galactic Nuclei
DARK MATTER
Black holes are not the only case in cosmology where existence can
only be veried through gravitational inuence. In 1933, eccentric and
brilliant Swiss astronomer Fritz Zwicky found that the mass of the Coma
cluster, estimated from the luminosity of the galaxies it contained, was
signicantly less than the amount of mass needed to generate enough
gravity to hold the cluster together. He posited the existence of some
dunkle materie, or dark matter. Three years later, Sinclair Smith found
a similar result for the Virgo cluster, that the total mass of the cluster
far exceeds the amount of mass visible in the individual galaxies. He
suggested that the cluster might contain internebular material . . . in
the formof great clouds of low luminosity (1936, 30). Both studies were
largely ignored by the scientists of the time, perhaps due to their focus
on single galaxy clusters.
Evidence for dark matter in individual galaxies began to slowly ac-
cumulate several years later through the study of the rotation of spiral
galaxies. In1914, Vesto Slipher found that M31 and the Sombrero galaxy
had a distinctive tilt to their spectral lines caused by the rotation of the
galaxies (seen nearly edge on). This is very different from the rotation
Measuring and Mapping the Universe 89
Adriaan van Maanen had claimed to see in face-on spirals. As an edge-
on spiral rotates, some of its stars are approaching us at varying speeds,
whereas some are receding at similar speeds. This makes half of each
spectral line blueshifted and the other half redshifted, causing it to ap-
pear tilted. The degree of tilt of different sections of each spectra line
tells astronomers how fast stars in different parts of the galaxy are mov-
ing. A graph which plots rotational velocity against the distance from
the center of the galaxy is called the galaxys rotation curve.
An edge-on spiral resembles an over-easy fried egg, with the nuclear
bulge being the yolk and the disk represented by the at white of the
egg. The halo of globular clusters is insignicant in this model. Observ-
ing the galaxy, it is clear that the majority of the mass shouldbe contained
inthe nuclear bulge, andas we move farther out inthe white, we expect
to encounter much less mass. This is similar to the solar system, where
the vast majority of the mass is contained in the center, in the sun,
and the planets form a relatively at plane containing far less mass. If
this model of the mass distribution for the spiral galaxy is correct, then
spirals should rotate like the solar system, with the inner stars orbiting
faster (as Mercury orbits the sun with a higher velocity) and the outer
stars orbiting at much slower speeds (similar to Neptune). A graph of
this type of motion, seen in Figure 4.4, is called a Keplerian rotation curve
(named for Keplers laws of planetary motion).
Horace Babcocks 1939 study of the rotation of M31 demonstrated a
serious problem with this assumption, as he found that the stars in the
outer parts of the galaxy orbited at nearly the same speed as those closer
to the nucleus. Jan Oort obtained similar results for the spiral galaxy
NGC 3115 the following year, and suggested that the galaxy contained
either a large number of dim dwarf stars or some other unseen material.
Babcock and Oort are given joint credit for recognizing the dark matter
problem in individual galaxies.
Rotation curves for other galaxies, using visible spectra as well as the
21-cmradio line, continued to show that spiral galaxies do not follow Ke-
plerian motion, but instead have at rotation curves. The fact that stars
orbiting near the visible edges of the galaxies orbit at nearly the same
velocities as stars much closer to the nuclear bulge suggested that either
our understanding of gravity is seriously suspect, or there is far more to
a galaxy than meets the eye. Vera Rubin and her colleagues found that
our own Milky Way has a at rotation curve, a pattern that she would nd
in hundreds of spirals over several decades. Despite the accumulation
of observational data suggesting some dark material dwelled in galax-
ies and galaxy clusters, the idea was not given wide credence until two
90 Cosmology 101
Figure 4.4 Flat Rotation Curve Observed for a Spiral Galaxy Versus a Kep-
lerian Rotation Curve
theoreticians, Jeremiah Ostriker and James Peebles of Princeton Univer-
sity, found evidence in 1973 that spiral galaxies might become distorted
or even y apart unless they possess massive halos that extend beyond
the visible galaxy.
The amount and distribution of dark matter within a galaxy can be
foundby modeling its observedrotationcurve, andsuggests that galaxies
have roughly spherical halos of dark matter that extend beyond the
visible edges of the galaxies and outweigh the visible stars, gas, and
dust by several times. Determining the amount of dark matter in a
galaxy cluster depends on accurately estimating the total mass of the
cluster and subtracting off the mass contributed by stars and other visible
constituencies. The rst of three methods of cluster mass determination
is that used by Zwicky, a study of the velocities of galaxies within a cluster.
The second involves haloes of intracluster mediumhot gas discovered
through its x-ray emissions. The hot gas would dissipate into space if
it were not gravitationally bound to the cluster. The total mass of the
cluster cantherefore be estimated by calculating howmuchgravitational
attraction would be needed to hold on to the observed gas, given its
temperature.
The nal method for determining the mass of a galaxy cluster involves
another prediction of general relativity. In 1936 Einstein noted that if
Measuring and Mapping the Universe 91
one star passed exactly in front of a more distant star, the image of the
distant star would be gravitationally distorted or lensed into a ring of
light. A less than perfect alignment would result in a more complex
series of multiple images. Given the slight chance that two stars would
align in this way, the prediction of gravitational lensing was largely ig-
nored for several decades. That changed in 1979, when Dennis Walsh,
R.F. Carswell, and R.J. Weymann discovered what seemed to be nearly
identical twin quasars appearing very close to the image of a galaxy.
Rather than claim that the two objects were instead a pair of unusually
similar quasars, they suggested the radical solution that they were actu-
ally two images of a single quasar caused by gravitational lensing by the
intervening galaxy. Since this original discovery, numerous examples of
gravitational lensing have been found, including the so-called Einstein
rings produced by the perfect alignment of galaxies and quasars.
Entire clusters of galaxies can also act as complex lenses, generating
images of more distant objects as strange blue arcs of light. The amount
of lensing can be used to construct a map of the mass distribution in
the lensing object, whether it be an individual galaxy or a galaxy cluster,
and has been a successful means of measuring dark matter distributions
in clusters. The results of numerous studies are that galaxies only make
up 10 percent of the total mass of a cluster, the hot intracluster gas
contributes 1025 percent, and the majority of the mass is found as dark
matter. In August 2006, this method was used to nd direct evidence
of dark matter. The Bullet cluster is a small piece of a galaxy cluster
which is ramming into another galaxy cluster. When galaxy clusters
collide, the galaxies are too spread out to smack into each other, but
the intracluster medium in the individual clusters collides, slows down,
and heats up to millions of degrees, emitting x-rays. The Chandra x-ray
satellite mapped the location of the hot gas in these merging clusters
through its x-rays. Dark matter is rather aloof and doesnt play well with
normal matter, so while the intracluster gas was busy colliding, the dark
matter kept passing through, and separated out from the normal gas.
The gravitational lensing tells us where all the mass is, the x-rays tell us
where the normal matter, so by default the rest of the material is the
dark matter.
LARGE-SCALE STRUCTURE
The existence of galaxy clusters led to the possibility of larger struc-
tures, now called superclusters. For example, the Local Group is a part
of the Local Supercluster, a loose aggregation of galaxy clusters about
30 Mpc wide, centered on the Virgo cluster. Superclusters and larger
92 Cosmology 101
structures are generally termed large-scale structure, and do not ap-
pear to represent features that are gravitationally bound to each other,
in contrast to galaxy clusters. One possible exception is the infamous
Great Attractor. In 1988, Sandra Faber, Alan Dressler, Donald Lynden-
Bell, Roberto Terlevich, Roger Davies, Gary Wegner, and David Burstein
unexpectedly found that our Local Supercluster is being pulled in a
direction in space in which there did not appear to be anything large
enough to so obviously disrupt the normal Hubble ow. The Seven
Samurai, as the astronomers were later nicknamed, dubbed the unseen
object the Great Attractor, and estimated that its mass must be that
of a very large supercluster. The exact identity and properties of the
supercluster (or superclusters) responsible for this motion are still un-
certain, hampered by the fact that it appears to be located in the zone
of avoidance and is therefore difcult to study.
The rst hints of the mammoth scale of the largest structures came
in 1981, when Robert Kirshner, Augustus Oemler, Paul Schechter, and
Stephen Shectman discovered a huge void in the constellation Bootes.
This zone of the sky, some 50 Mpc wide, was found to have many fewer
galaxies than normal. It was expected that voids were not very com-
mon, and that matter was distributed rather smoothly, as predicted by
the cosmological principle. The true extent of the voids and their in-
tervening structures was found in 1985 with a redshift survey headed by
Margaret Geller and John Huchra of the Harvard-Smithsonian Center
for Astrophysics (CfA). In a redshift survey, the spectra for all galax-
ies in a small slice of the sky are electronically captured by a CCD.
This slice would look similar to taking a piece of pizza and holding
the pointed end up to your eye. The slice of the universe observed is
very thin and stretches a larger distance side to side, causing the slice
of space measured to be wedge-shaped. Additional wedges of the sky
can be taken, either to the right or left or above and below in the sky,
and by stacking slices a three-dimensional picture of the local universe
emerges.
Their rst slice measured galaxies out to about 100 Mpc, and the data
was given to French graduate student Val erie de Lapparent to plot ap-
parent location in the sky (right ascension or celestial longitude) versus
redshift. To everyones surprise, the galaxies were not evenly distributed,
but were clumped together in a pattern similar to taking a slice through
Swiss cheese or a sponge, with the galaxies being found on the surfaces
of bubble-like structures some 3060 Mpc wide. Several extended strings
of galaxies, called walls, have been found in the CfA and other redshift
surveys, and can stretch for several hundred Mpc. Note that they were
plotting redshift, rather than distance, in this survey. FromHubbles law,
Measuring and Mapping the Universe 93
Figure 4.5 Redshift Survey Slice
velocity = distance H, so knowing the redshift leads directly to the
velocity, and therefore should give the distance. However, as we shall
soon see, one must rst separate out any peculiar velocities caused by
the motion of galaxies in a cluster. The CfA plot (and most redshift
surveys) fail to do this, resulting in some articial features appearing
in their maps. These structures are easily identied, as they look like
ngers pointing directly at the origin of the map, as in Figure 4.5, and
are called Fingers of God.
This lamentary structure of matter in the universe seems to threaten
the assumptionof homogeneity presumedinthe cosmological principle.
However, it has been found that on larger scales, the universe does not
become clumpier, but instead looks more uniform. This is similar to a
beach. As seen from the ground, the dunes, rocks, and even individual
sand grains give the beach a lumpy appearance. But as seen from an
airplane, the beach looks smooth and featureless. As we have seen in
this part of our journey through the cosmos, it may be homogeneous,
but the universe is far from dull.
THE COSMIC DISTANCE LADDER
There is no single method which by itself allows astronomers to mea-
sure the distance from Earth to the farthest visible galaxies in a single
94 Cosmology 101
step. Instead, there are a host of different methods, each of which is use-
ful for a specic range of distances, sometimes collectively referred to
as the cosmic distance ladder. By cross-checking different methods which
work for overlapping ranges of distances, errors are minimized. Despite
these good intentions, it should be noted that measuring the distances
to the farthest objects is extremely difcult, and uncertainties (a po-
lite way of saying errors) of 1020 percent (or more) are sometimes
encountered.
We begin literally in our own backyard, with the orbit of Earth. As
described previously, the astronomical unit is a standard unit of mea-
sure in astronomy. The rst (albeit imprecise) measurement of Earths
orbit was organized by Giovanni (Jean) Cassini in 1672. Mars appeared
close to a fairly bright star, and simultaneous measurements were made
from French Guyana and Paris. From the apparent difference in the
positions measured (i.e. parallax), a distance of 86,000,000 miles was
suggested. A more precise method was suggested by Edmund Halley,
which utilized transits of Venus across the face of the sun, events which
happen at a rate of about two per century. The transits of 1761, 1769,
1874, and 1882 were observed with these calculations in mind. Further
renements were possible in the 1960s, when radar waves were bounced
off Mercury, Venus, and Mars. The speed of light multiplied by the
time required for the round trip of the signals is twice the distance
between Earth and each planet, and through geometry, the au can be
derived
Once the au was known with good precision, the distances to nearby
stars can be determined through trigonometric parallax, often abbre-
viated to parallax. This is the same effect that Aristotle could not see,
and which he used as evidence against the motion of Earth. As we orbit
the sun, we view stars from different perspectives. If observations are
made of stars six months apart (on opposite sides of our orbit), as in
Figure 4.6, nearer stars should be seen to shift their positions relative
to farther (apparently xed) stars. The only information needed is the
size of Earths orbit and the amount of the shift (an angle). In prac-
tice, however, the measurements are quite difcult. For even the nearest
stars, the shift is a tiny fraction of a degree, measured in seconds of arc
(where there are 3,600 sec of arc per degree). These small angles are
generally smaller than the resolution of Earth-bound telescopes, as the
atmosphere smears out the star images. It is therefore not surprising that
successful parallax measurements were not made until the mid-1800s.
Friedrich Georg von Struve published a parallax of 0.26 sec of arc for
Vega (1837), Thomas Henderson found a parallax of Alpha Centauri
Measuring and Mapping the Universe 95
Figure 4.6 Parallax
of 1.26 sec of arc (1838), and Friedrich Wilhelm Bessel estimated a
parallax of 0.31 sec of arc for 61 Cygni. All three values were too large,
by factors ranging from 10 to100 percent (Webb 1999, 72). The parsec
was dened as the distance of a star having a parallax of 1 sec of arc (be-
ing the contraction of parallax-second). In reality, there are no known
stars that close to our own. The nearest star system, Alpha Centauri, has
the largest possible parallax, and the modern value is only 0.74 sec of
arc!
96 Cosmology 101
Given the difculty in doing these measurements, it is not surpris-
ing that parallaxes for only seventeen stars were known by 1878 and a
hundred by 1908. The seminal work on stellar properties, the Yale Bright
Star Catalog (1952), listed fewer than 6000 parallaxes. In 1989, the Euro-
pean Space Agency launched the Hipparcos satellite (named after the
ancient Greek astronomer), the rst dedicated to the measurement of
stellar parallaxes, positions, and motions. Following the completion of
its data collection in 1993, an impressive catalog of 22,396 parallaxes
(out to 90 pc) were measured to within 10 percent accuracy. Thousands
of other parallaxes were measured with less certainty, and positional data
on hundreds of thousands of stars were collected. Included in these stars
were a number of Cepheid variables, which led to a further renement
of the period-luminosity relationship.
Stars move through space in three dimensions (their space velocity),
but measurements are generally done in pieces. The radial velocity is
calculated from the Doppler shift, while the motion across the plane
of the sky is termed the tangential or transverse velocity. By combin-
ing the tangential and radial velocities through Pythagoras theorem
(a
2
+ b
2
= c
2
), the space velocity can be calculated. Although the vari-
ous velocities are normally reported in km/sec, the apparent motion of
a star across the sky is sometimes reported in seconds of arc per year,
called the proper motion. The closer stars tend to have larger proper
motions, just as a low ying plane seems to cover a large area of the
sky faster than a higher ying plane. Proper motion was discovered by
Edmund Halley, when he compared the current positions of the bright
stars Sirius, Procyon, and Arcturus to those on ancient Greek star maps.
For nearby groups or clusters of stars, proper motion can be used as
a method to determine distance. This moving cluster method is based
on the principal that since the stars in the cluster are all moving like a
ock of birds toward a common location, their proper motions will be
in the same direction and if the arrows of their motion are extended,
will all converge at some point in space. An example is ve of the stars
in the Big Dipper. Using trigonometry, the proper motion can be used
to determine the distance to the cluster. Like parallax, this method is
constrained to nearby groupings of stars. The most important moving
cluster is the Hyades, a naked-eye group of stars that makes up most
of the face of Taurus the Bull. At a distance of approximately 46 pc,
Hipparcos was able to determine an accurate parallax distance for this
cluster, which can be compared to that derived from the moving cluster
method.
Measuring and Mapping the Universe 97
STANDARD CANDLES
At this point trigonometry and geometry have to be abandoned for
less reliable distance measuring tools. At higher rungs of the cosmic
distance ladder, the concept of standard candles becomes increasingly
important. A standard candle is any astronomical object whose absolute
magnitude can be estimated with some degree of condence. The dis-
tance modulus formula is then used to calculate the distance. The most
important example weve encountered so far is the period-luminosity re-
lationship for Cepheid variables. The absolute magnitudes for standard
candles should derive or calibrate their values from reliable methods
(such as parallax), and wherever possible, a number of different stan-
dard candles should be used to estimate the distance to a given star
cluster or galaxy in order to reduce error.
Besides problems with calibrating the absolute magnitudes of stan-
dard candles, extinctionthe effect of intervening dust in making dis-
tant objects appear dimmer than they really areadds to the error.
Although astronomers have conscientiously tried to take extinction into
account since the 1920s, there still remains some uncertainty. For very
distant samples of objects, a third source of error creeps in. In 1920,
Karl Gunnar Malmquist warned that a bias toward brighter members
of any population of distant objects can lead to signicant error. The
Malmquist bias can be understood as follows. Consider a series of light
bulbs of different wattages (10, 25, 60, and 75). The average brightness
of the sample is 42.5W. If you observed the same sample from some dis-
tance away, you would not be able to see the 10W bulb because it is too
dim. Therefore, based on the three bulbs you can see you would estimate
the average wattage of the sample to be 53W. The farther and farther
you are from a population of light sources, be they light bulbs, stars, or
galaxies, the more dim members of that population you will overlook,
and the brighter you will believe the average to be. This has signicant
implications for standard candles, where we need to know the absolute
magnitude of a population of objects with certainty. The net result of all
these issues is that with each higher rung of the cosmic distance ladder
we (unwillingly) introduce greater error and uncertainty.
Let us return to the Hyades, whose stars distances are known reliably
through parallax and the moving cluster method. We can plot all 200+
stars onanHRdiagramif we knowtheir absolute magnitude andspectral
class. Since we know the distance, we can turn the distance modulus
formula around and calculate the absolute magnitude. The spectral
98 Cosmology 101
class of the star is determined from the strength of various spectral
lines. Recall that the width of the spectral lines places the star in its
proper luminosity class (such as main sequence or giant). Therefore
once we plot the positions of the Hyades stars, we can calibrate the
absolute magnitude for stars of a given spectral class and luminosity
class. For example, the absolute magnitude of a main sequence F0 star
is +2.6. while the absolute magnitude of a supergiant F0 star is 4.5. Of
course, we have assumed that all supergiant F0 stars are exactly alike,
which is not the case. But they are similar enough to use in this way. We
can also use any other stars whose parallaxes are known to augment our
calibrated HR diagram. The resulting diagram is then used to nd the
absolute magnitudes of other stars. If we know the stars spectral class
and luminosity class from its spectrum, we can use the HR diagram to
estimate its absolute magnitude. This is known as spectroscopic parallax.
Of course, a method is only as good as our calibration of the HR dia-
gram. The concept of stellar populations becomes important here. Re-
call that stars of lower metallicity (population II stars) are dimmer than
similar stars of population I. This was the origin of the subdwarfs, for
example, as the population I equivalent of the main sequence (dwarfs).
The problem is that of the stars close enough to have their parallaxes
measured, only about 0.15 percent are population II stars. Therefore
one of the purposes of Hipparcos was to measure enough subdwarfs to
have a proper calibration for objects made of population II stars, such
as globular clusters.
Of all standard candles, the most trusted are Cepheid variables. They
are fairly bright and can be easily seen in galaxies of small and moderate
distances. Since an individual galaxy can have many visible Cepheids,
the average of many distance measurements can be taken and checked
for consistency. Their distinctive variability makes them easy to pick out
of a series of photographs and identied as Cepheids, rather than other
kinds of variables. They remain in this variable stage for a long time (rel-
ative to the life of the average astronomer), so as technology improves,
well-known Cepheids can be remeasured. Also comforting is the fact
that the theoretical reasons for their variability are understood and can
be modeled via computer. Therefore, once the period-luminosity rela-
tionship is calibrated (preferably using parallax), Cepheids are utilized
with a high degree of condence.
Cepheids, however, are not perfect tools. Since they are population I
(younger) stars, they tend to be found in association with gas and dust,
which leads to problems with extinction making them appear dimmer
than they truly are. The ne-tuning effects of chemical composition
Measuring and Mapping the Universe 99
(metallicity) on the period-luminosity relationship have not been re-
solved. As more distant galaxies are observed, the stars appear closer
together, and it becomes more difcult to pick out the Cepheids.
The current limit of useful observations of Cepheids is about 30 Mpc
(30 million pc).
RR Lyrae stars are cousins to the Cepheids, and are also used as stan-
dard candles. Their absolute magnitude appears to be fairly constant,
between 0.5 and 1.0, and the limited metallicity effects are believed to
be fairly well understood. Because they are intrinsically much dimmer
than Cepheids, they can only be used for nearby galaxies, within Lo-
cal Group. Since RR Lyrae stars are population II (older) objects and
Cepheids are population I objects, unless we take care to use both kinds
of objects in the same galaxies and cross-calibrate, we run the risk of
having two parallel distance ladders rather than one reinforced ladder.
Recall that both RRLyrae and Cepheid variables occupy the instability
strip on the HRdiagramand represent stages in the old age and death of
a star. Other stages in the late evolution of a star can be used as distance
measuring tools. For example, the absolute magnitude of the tip of the
red giant branch (TRGB) on the HR diagram can be used. If one plots
an HR diagram for a globular cluster, the TRGB will show up as a sharp
cusp, representing stars which are beginning to use helium as fuel. This
is the brightest stage in the red giant evolution of a star. The absolute
magnitude of the TRBG appears to be fairly consistent, with small (if
any) metallicity effects, and can therefore be used as a standard candle.
When a low mass star completes its helium burning stage, it swells up
to the asymptotic giant branch and eventually puffs off shells of material
to form a planetary nebula. It has been found that although not all plan-
etary nebulae have the same absolute magnitude, there is a well-dened
distribution for the number of planetary nebulae at each absolute mag-
nitude within a given galaxy. Since this planetary nebula luminosity
function has a specic shape, it can be used to estimate the distance to
a galaxy assuming enough planetary nebulae can be seen. Since they
are found in both spiral and elliptical galaxies (unlike Cepheids), plan-
etary nebulae can tie together distance-measuring techniques that are
designed for a specic type of galaxy.
We have seen that the maximum brightness of novae has been used
as a distance indicator since in the early twentieth century. However it
was found that there was some dispersion in this value. More consistent
indicators seemed to be the magnitude fteen days after maximum and
the rate of decline in brightness right after maximum. Because of the
amount of data required to determine these values, novae have fallen
100 Cosmology 101
out of favor as a standard candle. Fortunately when Knut Lundmark dis-
tinguished between novae and supernovae in the 1920s, a new standard
candle was born.
As described previously, Type Ia supernovae (SNIa) are exploding
white dwarfs, while Type II supernovae are exploding red supergiants.
SNIa have become a well-trusted standard candle for several reasons.
Because of their high luminosity (approximately 19 maximum abso-
lute magnitude), they are easily visible over large distances (around
400 Mpc). This peak brightness seems fairly consistent from supernova
to supernova. What dispersion there is can be adjusted for by looking
at the shape of the light curve. Finally, the mechanism of the explosion
is fairly well understood. However, searching for a SNIa is like looking
for a needle in a haystack, and one never knows when or where one of
these relatively rare events will occur. Another lingering problem, which
will become of vital importance much later in our journey, is whether or
not supernovae in the earlier ages of the universe are the same as those
seen now. Type II supernovae are fainter and vary in their maximum
brightness. While it is possible to estimate their distances through de-
tailed calculations, they are not considered one of the favorite standard
candles.
For galaxies too distant for individual stars to be seen, no matter how
bright, the properties of the galaxy as a whole can be used to estimate dis-
tance. For example, in 1977 Brent Tully and Richard Fisher developed
a relationship between the rotation of a spiral galaxy and its overall lu-
minosity. Recall that unlike van Maanens supposed observations, spiral
galaxies are known to rotate, through their Doppler shifts. HI regions
in the spiral arms of the galaxy orbit the center of the galaxies, as do the
stars. Recall that HI regions emit distinctive 21-cmradio waves. However,
as the galaxy rotates, some of the HI regions will be moving toward Earth
(and have a slight blueshift relative to the center of the galaxy) while HI
regions on the other side of the galaxy will be moving away from Earth
(and have a similar slight redshift). The spread in wavelengths smears
out the 21-cm line, and the amount of smearing (or width of the line)
depends on the mass of the galaxythe greater the mass, the greater the
gravitational eld, and the faster the rotation. Assuming that the more
massive the galaxy, the more stars it has (and hence the more luminous
it is), the width of the 21-cm line therefore allows one to estimate the
overall absolute magnitude of the galaxy. It was found that because of
the effect of dust in the spiral arms of the galaxies, it was more reliable
to use the luminosity in IR rather than visible light, which is called the
IRTF or Infrared TullyFisher relationship.
Measuring and Mapping the Universe 101
What about elliptical galaxies? Their stars do not neatly orbit the
center of the galaxy, but instead have a more random motion. How-
ever, the average speed of the stars is also related to the total mass of
the galaxy. In 1976 Sandra Faber (of the Seven Samurai) and Robert
Jackson suggested that the dispersion in velocitythe spread in ve-
locities from averagefor stars near center of an elliptical galaxy can
be used to estimate the mass and therefore the luminosity. It was un-
fortunately found that this FaberJackson relationship has a larger in-
herent uncertainty than the TullyFisher relationship. Marc Davis and
Stanislav Djorgovski subsequently proposed that the velocity disper-
sion was better correlated to the diameter of a central region of the
galaxy within which the apparent brightness was 20.75 magnitudes per
square arc second. This fundamental plane relationship yields an un-
certainty of 1020 percent in distance for galaxies within a single galaxy
cluster.
For galaxies so distant that not even these increasingly complex meth-
ods are effective, only one method remains. Recall that Hubble found
that velocity =distance H, where His the Hubble constant. If we know
H, thenby measuring the velocity of a galaxy throughits redshift, it seems
simple enough to calculate its distance. That is, of course, assuming that
the value of His known with sufcient precision. That daunting task will
be tackled after one intermediate topicthe question of the age of the
universe.
THE AGE OF THE UNIVERSE
Although the age of the universe appears to be one of its most basic
properties, arriving at a consistent and veriable number has proven to
be a challenge. There are two basic routes used to attack this problem
directly measuring the age of the universe (howlong it has been expand-
ing), or estimating the ages of the oldest objects visible in the universe
and using this as a lower limit on the age. The former will be discussed in
the next section, while the latter will be tackled here. The three types of
objects whose ages have been used to set limits on the overall age of the
universe are radioactive isotopes, white dwarfs, and globular clusters.
In 1929 Lord Rutherford introduced the use of radioactive atomic
nuclei as a method to estimate the ages of rocks. When applied to
the oldest earth rocks, moon rocks, and meteorites, this technique has
been used to derive the currently accepted age of Earth and the solar
system of 4.6 billion years. In the 1950s, seminal research by Margaret
Burbidge, Geoffrey Burbidge, William Fowler, and Fred Hoyle (B
2
FH)
studiedthe productionof elements inthe universe, andopenedthe door
102 Cosmology 101
for extending radiometric dating to the universes oldest stars. The basic
concept is as follows. An unstable atomic nucleus decays into another,
lighter atomic nucleus at a specic rate. The original nucleus is called
the parent, and the resulting nucleus, the daughter. The time required
for half of a sample of radioactive nuclei to decay is called the half-life,
and it is a measurable and constant property of that species of nucleus
(or isotope). By comparing the relative amounts of parent and daughter
in a sample, the age of the sample can be determined. For example, A
decays into B with a half-life of 10 million years. If we nd a rock with
50 percent A and 50 percent B we know that the rock is one half-life,
or 10 million years, old. If we nd a rock that has 25 percent A and
75 percent B, we know the rock is 2 half-lives, or 20 million years, old.
The assumptions used here are that there was originally no B in the
rock, and that B is stable.
Recall that the rst generation of stars (population III) contained
essentially only hydrogen and helium, and that the rst atoms of heavier
elements, including radioactive ones, were created in the old ages and
deaths of these massive, short-lived stars. Therefore, by measuring the
amounts of certain radioactive isotopes remaining in very old (but not
rst generation) stars, the ages of these stars can be estimated. One
would then add to this age the lifespan of the rst generation of stars
(assumed to be of the order of a few million years) plus the time it took
for that rst generation of stars to form (a few hundred million years)
to derive the age of the universe. This technique is sometimes termed
nuclear cosmochronometry.
The two most useful isotopes are thorium-232 (which decays into lead-
208 with a half-life of 14.1 billion years) and uranium-238 (which decays
into lead-206 witha half-life of 4.5 billionyears). Althoughthe procedure
seems straightforward, it presents serious difculties to the astronomer.
First of all, the spectral lines of thoriumand uraniumare weak, so getting
an accurate measure of the relative abundance of these elements in a
star is problematic. Once the current abundances are measured, one
must know how much of these elements was originally present in the
star, which means one must understand how they were produced in the
rst place. Despite these problems, there have been a few successes.
Several very metal-poor stars have had their ages calculated as between
12 and 14 billion years.
White dwarfs, the corpses of low mass stars, can be used to estimate
the age of the universe through their slow march toward oblivion. After
a star reaches the white dwarf stage, it is no longer generating energy,
so it slowly cools and dims at what is believed to be a predictable rate. By
Measuring and Mapping the Universe 103
measuring the surface temperature of a white dwarf (through its spec-
trum), the time since its formation can be calculated, with the dimmest
andcoolest objects being the oldest. As withnuclear cosmochronometry,
there are complications with this method. The rate of cooling depends
on the composition of the white dwarf, particularly the relative amounts
of carbon and oxygen. It is also important to note that the white dwarfs
currently studied are in the disk of the Milky Way, which represents
a younger sample of objects than the globular clusters. It is therefore
expected that white dwarf ages will be signicantly lower than those of
the metal-poor stars used in nuclear cosmochronology. For example, a
number of disk white dwarfs have been found with estimated ages of
around 9 billion years.
By far the most widely trusted method for estimating the age of the
universe using astronomical bodies involves globular clusters. Consid-
ered the oldest visible objects in our galaxy, they are relatively plentiful
(over 150 known). The key is to understandstellar evolution, particularly
how the HR diagram of a star cluster changes over time as its compo-
nent stars evolve (with the more massive stars evolving faster than the
less massive stars). The basic underlying assumptions are that the stars
within the cluster were born at approximately the same time and have
essentially the same chemical composition.
Let us examine the stars of a globular cluster soon after their birth.
All of the stars will lie along the main sequence, with those of higher
mass lying higher up (at hotter spectral classes and greater luminosities)
along this narrowstrip, and stars of lower mass occupying lower positions
(cooler spectral classes and lower luminosities) on that same strip. Two
billion years later, if we return to visit the same cluster, stars heavier than
twice the mass of the sun (two solar masses) will have exhausted the
hydrogen in their cores and left the main sequence for the red giant
branch (or even death, in the case of the heaviest stars). Therefore the
main sequence will no longer have any really hot and bright stars. The
top of this clipped main sequence, seen in Figure 4.7 as the point where
stars are just moving toward the red giant branch, is called the main
sequence turnoff point, and is occupied by stars of exactly two solar masses.
Avisit 8 billion years later will show that the main sequence has retreated
even more, and only stars with the same mass of the sun or less remain.
The rest have become red giants or died. The main sequence turnoff
point is occupied by these one solar mass stars.
The method of dating globular clusters is then as follows. An HR
diagram of the stars in the cluster is plotted, and based on detailed
models of stellar evolution, the position of the main sequence turnoff
104 Cosmology 101
Figure 4.7 HR Diagram of a Globular Cluster
point is used to estimate the age. Implicit in this is the fact that the age
determination is only as good as the understanding of stellar evolution.
As computer models of stellar evolution have improved, the predicted
ages for globular clusters have signicantly decreased, from 16 billion
years or more in the 1980s to less than 13 billion years in 2003. This
decrease in estimates for the age of globular clusters occurred at a time
of serious debate over the age of the universe, as will be described in the
next section.
MEASURING THE HUBBLE CONSTANT
Since H, the Hubble constant, is the slope of the graph which plots the
velocity and distance of galaxies, both those properties must be deter-
mined with some precision in order to have a well-dened and accurate
value for H. The difculties in determining distance have already been
recounted. Although one might expect velocities to be more straightfor-
ward to measure, since they derive from the Doppler shift of a galaxys
Measuring and Mapping the Universe 105
spectrum, this is not the entire story. The Doppler shift is a measure
of the galaxys overall motion, toward or away from us. If a galaxy is
isolated, with no nearby galaxies exerting a signicant pull on it, then
it is safe to assume that the entire motion of this eld galaxy is due
to the expansion of the universe, or Hubble ow, carrying the galaxy
along. However, many galaxies occur in groups or clusters, where the
individual members exert a signicant gravitational pull on each other,
and clusters exert a gravitational pull on other relatively close clusters.
This portion of the motion, termed peculiar motion or non-Hubble ow,
is responsible for creating the Finger of God effect seen in redshift
surveys, and must be subtracted from the overall motion in order to
determine the Hubble ow. Doing this accurately is much harder than
it seems. For example, the galaxies in the nearest large cluster, the Virgo
Cluster, exhibit a complex motion relative to each other, plus the overall
motion of the cluster is being signicantly affected by other neighboring
clusters.
To avoid this problem, astronomers might only use eld galaxies to
plot a Hubbles law graph, or use the motion of a galaxy positioned
at the center of gravity of a cluster (if one can really determine the
center of gravity). The usual procedure is to use galaxies so far away
that their peculiar motion is negligible compared with the expansion
of the universe. But then we run into the problem that it is increasingly
difcult to measure distances accurately the farther out we look. Another
issue to keep in mind is that the value of H changes over time as the
universe evolves. Because of this, the shape of the graph of Hubbles
law is expected to deviate from a straight line when distant galaxies
are plotted. The exact change predicted depends on the cosmological
model used(for example, open, closed, or at FRWmodels). Conversely,
if a precise graph can be derived, the deviations from a straight line can
be used to determine the geometry and evolution of the universe. We
will revisit this important topic. Measurements of H using what might
be considered our cosmic neighborhood determine the current value
of H, more specically called H
0
by cosmologists.
Since H
0
measures the rate of expansion of the universe, we can
use it to derive the time since the universe began expanding. Roughly
speaking, the age of the universe is the inverse of H
0
times a trillion
(1/H
0
10
12
). This is a rough estimate, since H changes with time,
and is actually most accurate for an open universe with no cosmological
constant. If one assumes a at universe with no cosmological constant,
the age is two-thirds of this value (2/3H
0
10
12
). If one includes a
cosmological constant, the universe can be much older than the value
106 Cosmology 101
predicted from 1/H
0
. In 1936, Hubble himself was advocating a value of
530 km/sec/Mpc for the constant, which translates to a maximum age
of the universe of 1.9 billion years. This is clearly younger than the age
of Earth (as derived from rocks). As might be expected, this was deeply
troubling to scientists, and was one of the motivations for Lematre
including a cosmological constant in his model of the universe. One
could adjust the value of the cosmological constant such that there was
good agreement between the age of the universe predicted by H
0
and
the age of Earth.
Models with a cosmological constant fell out of favor, but a major
breakthrough occurred in the 1950s, when Walter Baade discovered
that there were two different types of Cepheids with different period-
luminosity relationships. Distances to galaxies were recalculated and
found to be greater, leading to a lower value of H
0
. In 1956, Milton
Humason, Nicholas Mayall, and Allan Sandage used over 800 galaxies
to calculate H
0
, which they found to be 180 km/sec/Mpc. This translates
to a maximum age of the universe of 5.6 billion years, older than the age
of the earth. Allan Sandage of Mount Wilson Observatory has devoted
much of his career to rening measurements of H
0
, and in 1958 pub-
lished a further downward estimate of 75 km/sec/Mpc. He based this
on a correction of several errors that had crept into previous estimates,
including the misidentication of HII regions as bright stars, which had
been erroneously used as standard candles by Hubble and others.
During the next four decades, there developedtwocamps incosmol-
ogy, one led by Allan Sandage and Gustav Tammann of the University of
Basel, advocating for a lower value of H
0
(around 50 km/sec/Mpc), and
another spearheaded by Gerard de Vaucouleurs (University of Texas)
which argued for higher values around 100 km/sec/Mpc. The 100 per-
cent difference in their numbers was based on what methods for mea-
suring distances each group favored. Many observational astronomers
leanedtowardthe higher values, while many theoreticians supportedthe
lower ones, mainly due to the ages of the universe each predicted. For
an open universe (with no cosmological constant), H
0
=50 predicts the
universe is 20 billion years old, while H
0
= 100 leads to an age of 10 bil-
lion years. As discussed earlier, the predicted ages for globular clusters
stood at 16 billion or more years for several decades, in open contradic-
tion to the H
0
=100 value. The contradiction became signicantly more
serious in the 1980s with the development of the inationary model of
the early universe, which predicted that the universe was at. This meant
that the ages predicted by both H
0
= 50 and H
0
= 100 (13.3 billion and
6.7 billion, respectively) were too young to house the globular clusters.
Measuring and Mapping the Universe 107
Clearly a denitive study of H
0
was required. In 1986, a team of
astronomers led by Marc Aaronson was awarded the Key Project on
the Extragalactic Distance Scale on the then-unlaunched Hubble Space
Telescope. The goal of the project was to observe Cepheids in targeted
spiral galaxies (including those in the Virgo Cluster), and to use these
measurements to estimate the distances to farther objects such as the
Coma cluster in order to determine H
0
to within 10 percent uncertainty.
Following the tragic death of Aaronson in a freak accident at Kitt Peak
Observatory in 1987, Wendy Freedman took the reins of the project. In
the wake of the Challenger shuttle tragedy, the launch of the HST was
delayed from1986 to 1990. Due to the optical problems of the telescope,
major data collection was delayed until after special optics were installed
on the cameras in December 1993.
In 1994, a paper by Freedman and her colleagues describing their
initial results set the astronomical world spinning. Based on the mea-
surements of Cepheids in M100, a galaxy in the heart of the Virgo
Cluster, Freedmans team derived a value of 80 17 km/sec/Mpc for
H
0
, and a value of 77 16 km/sec/Mpc using the Coma Cluster. Un-
derstanding that these high values for H
0
were not consistent with the
accepted ages of globular clusters in a at (inationary) universe, they
openly predicted that this age conict suggests that either the stan-
dard cosmological model needs to be revised, or present theories (or
observations) bearing on stellar and galactic evolution may need to be
reexamined (Freedman, et al., 1994, 761).
The popular press was quick to smell the virtual blood. The March
1995 cover of Discover magazine was blazoned with Crisis in the Cos-
mos, with the crowing pronouncement that these days cosmology
seems to be collapsing in on itself (Flamsteed 1995, 66). The story
quoted astronomer David Weinberg as saying It would be premature to
panic . . . but if these results are conrmed, we theorists will be in trou-
ble (Flamsteed 1995, 68). Unraveling Universe was the cover story
of the March 6, 1995 issue of Time, where the young whippersnapper
Wendy Freedman was set in dramatic opposition to grumpy colleague
Allan Sandage. The article sensationalized the scientic debate, writing
that
Nobody can say what the turmoil meanswhether the intellectual edice
of modern cosmology is tottering on the edge of collapse or merely feeling
growing pains as it works out a few kinks. If you ask me, says astrophysicist
Michael Turner . . . either were close to a breakthrough, or were at our
wits end. (Lemonick and Nash 1995, 77)
108 Cosmology 101
As it turned out, the former proved to be true. Within several years, ev-
idence for a cosmic repulsion akin to Einsteins cosmological constant
began accumulating, which meant that the universe could be much
older than the value predicted by H
0
. At the same time, the ages of glob-
ular clusters were also revised, downward to about 12.6 billion years,
which was within the timeframe predicted by an updated cosmological
model including the cosmological constant. By the time Freedmans
team published their nal value of H
0
in 2001, 72 8 km/sec/Mpc,
the crisis had been resolved. Viewed through the detached lens of time,
historians of science will dub the entire cosmic age crisis a textbook
illustration of the self-correcting mechanism of science. It is also a re-
minder that scientists are human beings, and as such, can sometimes be
extremely stubborn when faced with the demise of models, observations,
or calculations they hold dear.
5
The Evolution of
the Universe
THE STANDARD MODEL OF PARTICLE PHYSICS
The Fundamental Forces and Particles
It may seem strange that a book whose subject is the largest thing that
existsnamely the universeshould devote time to the study of the
tiniest forms of matter. But consider the fact that the universe has not
always been the unfathomable immensity it is today. The foundation of
the big bang model is the concept that the universe emerged froma hot,
dense state of being, in which the universe had more in common with
high-energy particles than galaxies. Particle physics, the scientic study
of the fundamental building blocks of nature, derives from quantum
mechanics. Particle physicist Sidney Coleman once quipped that if
thousands of philosophers spent thousands of years searching for the
strangest possible thing, they would never nd anything as weird as
quantum mechanics (Randall 2005, 117). So far we have encountered
several aspects of quantum mechanicsthe waveparticle duality, Pauli
exclusion principle and degeneracy pressure, and blackbody radiation.
Another outgrowth is quantum eld theory.
A eld is some physical quantity which can be dened over a region
of space-time. The best-known examples of elds are electric elds and
magnetic elds (spoken of under the generic title of electromagnetic
elds), and the gravitational eld. Fields are invisible, but they affect
matter and therefore can be measured and studied. For example, a
charged particle such as an electron will move in a particular way in
a magnetic eld. The interactions of elds (with each other and with
matter) are governed by eld equations, such as the Einstein eld equa-
tions of general relativity. In quantumeld theory the uctuations of the
110 Cosmology 101
eld are manifested as particles. For example, uctuations in electric
and magnetic elds produce photons (particles of light).
One of the strangest predictions of quantum mechanics involves the
Heisenberg uncertainty principle. In classical physics it is possible to mea-
sure both the position of a car and its momentum at any instant as
accurately as ones tools will allow. But consider an electron instead of
a car, and the rules of quantum mechanics now apply. The very act of
measuring the position of an electron (for example, by shining a light
on it) changes its motion, preventing us from knowing both its position
and momentum at the same time to absolute certainty. Another pair of
variables so connected is energy and time. This means that it is possible
for the universe to violate conservation of energy for a brief period of
time, creating energy out of what most people consider the ultimate
nothingnessthe vacuum. This borrowed energy is used to create a
particle-antiparticle virtual pair, called virtual because it only enjoys an
ephemeral existence. In the blink of an eye, the pair annihilates, and in
the process pays back the energy debt the universe accumulated from
their creation. As strange as this process may seem, it is one of the most
fundamental aspects of nature at the quantumlevel, and has measurable
effects we can directly observe. For example, two parallel metal plates
set in a vacuum will be attracted to each other via the Casimir Effect, not
by the gravitational or electric forces, but because of the virtual pairs
continually being created and annihilating between them. Virtual pairs
also create a tiny shift in the energy levels of atoms, observable as the
Lamb effect.
If quantum mechanics affects small particles of nature, it would be
helpful to know what kinds of particles (and their associated antiparti-
cles) exist. Molecules are made of atoms, and atoms are made of elec-
trons, protons, and neutrons. Are these particles truly fundamental, or
are they made of even smaller particles? Like a set of wooden Ukrainian
nesting dolls, every time we peel off one layer of structure we nd
another, smaller level underneath. Have we found the most fundamen-
tal particles of nature? The search for this answer began in the 1950s
and 1960s, when physicists collided atomic nuclei and other particles at
higher and higher energies. They found that they could create an entire
zoo of particles that resembled protons and neutrons but were much
heavier. These hadrons (from the Greek for thick) were suggested by
physicist Murray Gell-Mann to be made of even smaller particles still,
which he named quarks. This suggested that an entirely new language
and classication system of particles was about to be born.
The Evolution of the Universe 111
Particles in nature can be divided into two categories, fermions and
bosons. Fermions obey the Pauli exclusion principle, while bosons, in the
words of physicist Lisa Randall, are like crocodilesthey prefer to pile
on top of one another (2005, 147). The particles that make up matter,
such as electrons, protons, and neutrons, are fermions, while bosons,
like the photon, are responsible for mediating the four fundamental
forces of nature. These forces are gravity, electromagnetism, the weak
nuclear force, and strong nuclear force. Hadrons are particles that interact
with each other via the strong nuclear force (sometimes abbreviated as
the strong force) because they are made of quarks. The strong force
itself holds the quarks together to make the hadrons. Since this force is
so strong that it overcomes the electric repulsion of quarks of the same
charge, it earns its name. The strong force is 137 times stronger than
the electromagnetic force, its nearest competitor in strength.
Hadrons can be divided into baryons (from the Greek for heavy),
which include protons and neutrons, and mesons, which do not enter
into our story. Since the vast majority of the mass of an atom is due to its
neutrons andprotons, what we consider normal matter is referredto as
baryonic matter. Each baryon is composed of three quarks, held together
by bosons called gluons. Therefore gluons are said to mediate the strong
force. Quarks come in six varieties, called avors, but these are not your
normal chocolate, vanilla, and strawberry. Instead, particle physicists
somewhat humorously dubbed them up, down, strange, charm, top,
and bottom. Each quark has a fraction of the charge of the electron
or proton, with the up, charm, and top quarks each having a charge of
+2/3, while the others have a charge of 1/3. If you suddenly doubt
your high school chemistry teacher, who taught you that particles could
only have whole number charges, dont. Isolated quarks are not directly
observable in nature. The trios of quarks in baryons have a total whole
number charge. For example, a proton is a combination of two up
quarks and one down quark (uud), which has a total charge of 2/3 +
2/3 1/3 =+1. Aneutron is one up and two downs (udd), with total
charge 2/3 1/3 1/3 = 0. The other avors of quarks are believed
to make up a miniscule amount of matter in the universe and will not
bother us further.
Take a closer look at the protonfor a moment. The twoupquarks seem
identical to us. But doesnt that violate the Pauli exclusion principle?
There must be some property that makes one of the ups different from
the other so they occupy a different quantumstate inthe proton. Particle
physicists gave this property the name color, although it has nothing to
112 Cosmology 101
do with how we normally think of the word. Of the three quarks in
the proton, one is said to have blue color, another red and the
third green. The total proton is color neutral, as it has one quark
of each color. Because of all the different possible color interactions,
there are eight different types of gluons which mediate the strong force.
The theory that describes how quarks interact to form hadrons is called
quantum chromodynamics, or QCD. This theory has been very successful
in explaining many aspects of the subatomic world, and in 2004 David
J. Gross, H. David Politzer, and Frank Wilczek shared the Nobel Prize in
Physics for their contributions to the development of QCD.
The weak nuclear force (or weak force for short) is about 10,000 times
weaker than the strong force, and describes how quarks change avor, as
well as interactions involving a different class of fermions, called leptons
(from the Greek for light). Just as there are six avors of quarks, there
are six different types of leptons, the most important being the electron
and the electron neutrino (commonly called the neutrino, although
in actuality there are three different types of neutrinos). Unlike the
hadrons, the leptons are truly fundamental particles. To the best of our
knowledge they cannot be subdivided into smaller pieces. Because its
interactions involve both quarks and leptons, the weak force mediates
the radioactive decay of atomic nuclei, andeventhe decay of the neutron
itself. Afree neutron(one not bound ina nucleus) falls apart, witha half-
life of about 10 min, creating a proton, an electron, and the antiparticle
of the electron neutrino (an antineutrino). At the quark level, one of
the neutrons down quarks is converted into an up quark, with the
release of an electron and the antineutrino. Since quarks have positive
and negative charges, a weak force interaction can change charge (from
positive to negative or negative to positive) or keep it the same. Because
of this, there are three different bosons that mediate the weak force
W
+
, W

, and Z. This last particle has no charge and is similar to a

photon, the particle that mediates the electromagnetic force, but unlike
the photon, the Z has a nonzero mass. In 1984, Simon van der Meer and
Carlo Rubbia shared the Nobel Prize in Physics for their experimental
discovery of the W and Z particles.
Unifying the Forces
Four fundamental forces might seem a reasonable number, but physi-
cists were intrigued by the possibility that this number could be an arti-
fact of the state of the universe today. For example, before the mid-1800s,
scientists believed that the electric and magnetic forces were separate
entities. By 1868 James Clerk Maxwell had demonstrated that they are
The Evolution of the Universe 113
really two aspects of a more fundamental entity, the electromagnetic
interaction. The quantum eld theory description of the electromag-
netic force is called quantumelectrodynamics, or QED. Sin-Itiro Tomonaga,
Julian Schwinger, and Richard P. Feynman shared the 1965 Nobel Prize
in Physics for their work in developing QED. The similarities between
the W and Z particles that mediate the weak force and the photon,
which mediates the electromagnetic force, suggested that in high energy
experiments (at a temperature of around10
15
K) these similarities would
become even more overwhelming and the two forces would act as one
dubbed the electroweak force. Sheldon Glashow, Abdus Salam, and Steven
Weinberg shared the 1979 Nobel Prize in Physics for developing the
theoretical underpinnings of the electroweak theory. Experiments have
veried the predictions of the electroweak theory (such as the proper-
ties of the W and Z particles), and the electroweak theory combined
with quantum chromodynamics forms what is called the Standard Model.
Although this model has been very successful in terms of experimental
verication, there remain questions that suggest it is a work in progress.
It relies on 19 arbitrary constants (such as the masses of the funda-
mental particles) whose values are not predicted by the theory itself,
but rather the proper values are put in by hand. This arbitrariness of
what should be fundamental constants of nature has led some to com-
ment that the Standard Model may be successful, but it is certainly not
elegant.
If the electromagnetic and weak forces unify at higher energies, could
the same be possible with the strong force? It appears so. Although there
is no one unique unication scheme, the class of solutions are called
grand unied theories (or GUTs). Part of the problem is the unimaginable
energy necessary to test the predictions of GUTs directly. In order to
experimentally generate the necessary energy and temperature (around
10
28
K), a particle accelerator (atom smasher) over 20 pc long would
be required (Guth 1997, 31). An advanced degree in particle physics
isnt required to understand that this experiment will never be possible.
How, then, canphysicists test the various versions of GUTs and see which,
if any, most properly describes the universe we live in? The key is the
universe itself. According to the big bang model, the universe is only cold
today (at an ambient temperature of around 3 K). When the universe
was much younger, it was also much hotter, and the extreme energy
necessary to unite the electroweak and strong forces was generated. It
is no surprise that beginning in the late 1970s, particle physicists and
cosmologists began joining forces to understand not only the tiniest
particles of nature, but the infancy of the universe itself.
114 Cosmology 101
Although the nal properties of GUTs are still being studied, there
are three important predictions that could be experimentally veried.
The rst is the creation of particles called magnetic monopoles. These are
particles that act like the north or south pole of a magnet, but isolated
without the other pole. These particles have not yet been discovered.
The second prediction is that the proton will very slowly decay, with a
half-life much, much longer than the current age of the universe. Recall
that the half-life is the time required for half of a sample to decay, so
given a large enough sample, several protons should be seen decaying.
The fact that protons have not yet been observed to decay means that
their lifetime is very long indeed, and has actually been used to rule out
some of the competing versions of GUTs.
The third prediction is the existence of particles called axions. They
arise through a rather complex system of conservation laws that have
been found to be slightly broken by the universe. Particle physicists have
found that particles and antiparticles have slightly different properties
(called the violation of charge invariance), that the laws of physics can
differ when the spins of particles are reversed (violation of parity invari-
ance) and that the laws of physics are even slightly different in some
cases when reactions go forward versus backward (violation of time in-
variance). These violations are important, because they are believed
to have allowed matter to dominate over antimatter in the very early
evolution of the universe, a process dubbed baryogenesis. Axions are an
outcome of some models of baryogenesis, and although they individu-
ally have small masses, they could exist in such large numbers that they
could affect the evolution of the universe.
If three of the four fundamental forces can be unied in the high
temperatures and energies of the early universe, why not all four? The
problemis that the three forces combined inGUTs all play by the rules of
quantum mechanics. The fourth fundamental force, gravity, is currently
described by Einsteins general theory of relativity, and to say that quan-
tum mechanics and general relativity dont play well together would be
an understatement. Einstein failed in his attempt to unify the two forces
best understood in his time, gravity and electromagnetism, for precisely
this reason. A marriage between quantum mechanics and general rela-
tivity would result in a completely unied theory of everything (TOE),
sometimes called quantum gravity. A number of different possible ver-
sions of quantum gravity are currently being explored by physicists,
including supergravity, an outgrowth of the theory of supersymmetry. This
is an attempt to achieve the ultimate unication by connecting the par-
ticles that mediate the forces (bosons) with those that make up matter
The Evolution of the Universe 115
(fermions). In supersymmetry (called SUSY for short), every boson is
associated with a fermion, and vice versa. The still-unseen associated
particles are called supersymmetric partners. For example, the electron,
a fermion, would have a supersymmetric partner called the selectron,
which would be a boson. Versions of SUSY that include gravity are called
supergravity (or SUGRA), where gravity is thought to be mediated by a
boson called the graviton. Its SUSY partner is an as yet unseen fermion
called the gravitino. The temperature necessary to unify all four forces is
phenomenal, estimated to be about 10
32
K (a hundred thousand billion
billion billion degrees).
Since the four forces of nature were unied at the high temperatures
of the very early universe, they must have separated into the distinctive
personalities we see as the universe cooled. The mechanism is termed
spontaneous symmetry breaking. An everyday example of symmetry break-
ing can be found at a round dinner table. Typically the eight or so
guests will sit down, look around, and be uncertain as to which bread
plate or drinking glass is supposed to be theirs. There is a therefore
perfect symmetry of the dishware on the table. However after an awk-
ward moment or two, one bold person will reach out and grab a glass,
or subtly drag one of the bread plates closer to his main plate, thus
breaking the symmetry for the entire table. If the person chose the glass
to his left, so must everyone else at the table. In quantum eld theory,
symmetry preserving states are unstable, as if the symmetry wanted to be
broken.
The actual breaking of the symmetry is done by a theoretical particle
called the Higgs particle, proposed by Peter Higgs in 1964 to explain why
particles such as electrons have mass. It is an example of a scalar par-
ticle, which has no spin. Since particles and elds are interchangeable
through the wave-particle duality, there also exists a Higgs eld, which
is a scalar eld. This eld is simpler than the electric or magnetic eld,
because only the value of the scalar eld at a point, not its direction, is
measured. For example, a weather map that plots temperatures repre-
sents a scalar eld, whereas a weather map that plots wind speed and
direction does not. In grand unied theories there are multiple Higgs
elds, the actual number varying from model to model. The symmetry
of the model is preserved when all Higgs elds are zero, but as soon as
least one of them takes a value other than zero, the symmetry is bro-
ken. The electroweak theory uses a different Higgs eld in its symmetry
breaking. Although the Higgs particle (and its related eld) has so far
eluded detection, most physicists expect it to be only a matter of time
(and higher energy experiments). Although the Standard Model does
116 Cosmology 101
not require the existence of the Higgs particle from rst principle, with-
out it the model is only mathematically consistent if all the fundamental
particles are massless and travel at the speed of light (like the photon).
Since this is clearly not the case in our universe, the Higgs particle (or
something very similar to it) is needed to make most of the fundamental
particles massive rather than massless.
THE FIRST FIFTEEN MINUTES
If we follow the expansion of the universe backward in time, we seem
to be driven to a singularity, where the entire universe is compacted into
a mathematical point of innite density and zero volume, and where the
equations of general relativity break down. This seemingly ridiculous
prediction is a symptom of our lack of a unied theory of quantum
mechanics and general relativity. Once such a theory of everything
is developed, it is expected that the singular birth of the universe will
be avoided. This also means that we currently lack the information
to trace the history of the universe all the way back to the absolute
beginning. Just how far back can we reliably go? This is determined
by the scale of nature at which quantum effects cannot be ignored in
gravitational calculations, which in turn is determined by the values of
three fundamental constants of natureNewtons gravitational constant
(which sets the relative strength of gravity), the speed of light (which
sets the maximum speed limit), and the Planck constant (which sets
the minimum uncertainty in measurements). Combining these three
constants, it is found that the scale of quantum gravity is measured
using Planck unitsthe Planck length (10
35
m), the Planck time (10
43
seconds), and the Planck mass (10
5
g). The Planck length can be
thought of as the size of the quantum mechanical texture of space-
time, the Planck time as the earliest time in the universe we can discuss
without quantum gravity, and the Planck mass is the smallest possible
mass of a black hole.
Our tour of the history of the universe begins at the Planck time,
10
43
sec after the event that spawned the big bang. Before this time, in
the Planck era, all four fundamental forces were one unied superforce,
whose details await a successful theory of quantum gravity. At the end of
the Planck era, the universe underwent a spontaneous symmetry break-
ing, and the gravitational force became separate and distinct, destined
to remain so for the rest of time. The next period is called the GUT
era, which can be described by grand unied theories. The universe is
a soup of quarks, leptons, and photons interacting in madly energetic
The Evolution of the Universe 117
reactions that we cannot replicate in our laboratories. It is in this era
that baryogenesis takes place, and matter gets the upper hand over an-
timatter. Pioneering theoretical work on baryogenesis was done in 1967
by Russian physicist Andrei Sakharov. An advocate of human rights and
nuclear weapon opponent, he received the Nobel Peace Prize in 1975.
Ironically, he was one of the scientists who helped the Soviet government
develop the hydrogen bomb. At 10
35
sec (10
28
K) another symmetry
breaking event takes place, and the strong force takes on a separate life
of its own, leaving the electroweak force behind. Leptons do not feel
this newly distinct strong force, so they and quarks can no longer in-
teract as equal partners as they had before. Another important event is
theorized to have occurred at this important transition, namely ination,
which will be a major theme in a later section.
During this Electroweak era, the temperature cools to 10
15
K (a thou-
sand trillion), and when this temperature is reached (10
11
sec), the
electromagnetic and weak forces separate from each other, establishing
the four fundamental forces we experience today. It is amazing to con-
sider that no new forces have established themselves after the rst ten
trillionths of a second of the universes history. The next epoch is called
the Hadron era, in which quarks form hadrons and just as quickly the
hadrons are broken back up into quarks. Only at the end of this era
(10
5
sec), when the universe has cooled to a mere 10
10
K (ten billion
Kelvin), can the quarks remain conned into hadrons. This results in
the birth of protons and neutrons.
With this phase completed, the stage is nally set for the formation of
the rst elements. During this Leptonera, as inthe previous epochs, pho-
tons (radiation) and matter were in thermal equilibrium. This means
that because of the ferocious high-energy interactions, the temperature
of the radiation was the same as the temperature of the matter. At tem-
peratures higher thanabout 10
10
K, the slight difference inmass between
the heavier neutrons and the lighter protons is ignored by weak force
interactions, and protons and neutrons turn into each other in equal
numbers. For example, a proton and an electron can combine to create
a neutron and an electron neutrino, or vice versa, with equal probability.
When the universe cools below this critical temperature, the neutrinos
decouple, meaning they fall out of equilibrium with the other particles
and become free. This background of relic neutrinos lls the universe.
With the neutrinos essentially out of the game by a few seconds after the
beginning of the universe, neutrons were doomed to decay, with their
characteristic half-life of about 10 min. But fortunately long before all
118 Cosmology 101
the neutrons in the universe could decay, another critical temperature,
10
9
K, was reached, at about 3 min, and the Lepton era was completed.
In order to form atomic nuclei more complex than simple hydrogen
(which is, after all, merely a single proton), the rst stage is to combine
a proton and a neutron to form a deuteron, a nucleus of deuterium
(heavy hydrogen). However, at temperatures above a billion degrees,
deuterons are torn apart by collisions with high-energy photons as soon
as they are formed. This deuterium bottleneck holds up the creation of
atomic nuclei until about 3 min into the history of the universe. Once
the bottleneck is breeched, neutrons and protons rapidly combine to
form in succession deuterons, nuclei of helium-3 (one proton and two
neutrons), helium-3 (two protons and one neutron), and nally helium-
4 or alpha particles (two protons and two neutrons). This continues
until all the free neutrons are used up, and once the neutrons are
bound into nuclei, they do not decay but are stable. This primordial
or big bang nucleosynthesis xes the overall chemistry of the universe at
roughly 75 percent hydrogen and 25 percent helium, with tiny amounts
of deuterium and helium-3.
Recall that there are no stable nuclei with atomic mass of ve, so
the few nuclei that jump the gap end up producing a small amount
of lithium-7. The second mass gap, at eight, is much harder to jump,
and primordial nucleosynthesis ends about 15 min after it begins. The
universe must wait for the creation of the rst generation of stars for
further modications in its chemistry.
THE DARK AGE
After the end of this Nucleosynthesis era, the universe entered a long
stretch of relative quiet. The temperature of the universe, as determined
by the sea of photons it contained, cooled as the universe expanded
further. It is at this period of time that the observed ratio of photons
to matter particles, a whopping two billion to one, was xed. Thus far,
the evolution of the universe had been dominated by the effects of
radiation, leading to the generic term of Radiation-dominated universe
for this portion of the universes life.
As the universe expanded, the energy density of photons eventually
became less than the energy density of matter, and the universe became
Matter-dominated. Shortly afterward, about 300,000 years after the rst
instant of time, the universe underwent another vital transition, that of
recombination. A rather confusing choice of words, this refers to the fact
that the temperature had nally cooled enough (to about 4,500 K) to
allow electrons to be bound to the hydrogen and helium nuclei for the
The Evolution of the Universe 119
rst time and the rst stable atoms were formed. With the electrons
safely bound into atoms, there was nothing to scatter the photons, and
they ew through the universe unimpeded. Previously the universe had
beenshroudedinanopaque fog of scatteredphotons, but it nowbecame
transparent. These photons continue to ood the universe today as the
cosmic microwave background. Over the intervening billions of years,
their wavelengthhas beendramatically redshiftedandtheir temperature
has dropped to its current value of about 3 K.
While radiation evolved on its own after recombination, matter began
the rst steps toward accumulating into the structures we see today.
In keeping with the cosmological principle, matter was more or less
uniformly distributed throughout the universe, but not perfectly so.
Regions of matter with slightly higher than average density tended to
clump together under the inuence of gravity, eventually leading to the
rst generation of stars. Before this occurs, the universe is plunged into
a temporary but utter darkness, as the CMB is redshifted from visible to
infrared wavelengths. This dark age lasted from a half million to about
200 million years after the beginning of the universe. When the rst
generation of stars, called population III, turned on, the universe was
again lit.
Discerning the details of the birth of this rst generation of stars has
been a challenge for astrophysicists, because they were made from the
raw materials produced by primordial nucleosynthesisnearly pure hy-
drogen and helium with virtually no metals. One of the problems is
that in the formation of normal stars (those which contain metals), met-
als and dust grains cool the nebula out of which the star forms, allowing
gravity to win out over the outward pressure of radiation and collapse
the material into a protostar. Molecular hydrogen (H
2
) is believed to
have acted as the cooling (heat-dissipating) agent in population III star
formation and is very inefcient.
Another important difference between population III and population
I and II stars is that they tend to form with unusually large masses100
times the mass of the sun or more. These supermassive stars consume
their fuel at an astounding rate, shining with a luminosity over a mil-
lion times that of the sun. Because they lack heavy elements such as
carbon, they cannot use the carbon cycle to fuse hydrogen but are
limited to the protonproton cycle despite their high core temperature.
They also lack the necessary materials to make carbon-based life-forms
or even solid planets like the earth. They could have had gas giants like
Jupiter, but these unfortunate planets would have been roasted by their
peculiar stars unusually high surface temperaturenearly 110,000 K.
120 Cosmology 101
After a brief lifespan of only a million years, these rst stars would have
consumed all their fuel and begun to die. But their deaths may possi-
bly have differed greatly from those of future generations. Stars with a
nal mass less than 140 solar masses or greater than 260 solar masses
would have died as black holes, trapping most of their chemically en-
riched mass in an unusable form. Population III stars with masses in
the 140260 solar mass window are theorized to die by a process called
pair-instability supernovae (PISN). In these stars, when oxygen fusion be-
gins the temperature is high enough for photons to create electron
positron pairs. A positron is the antiparticle to the electron, and has
a positive charge, hence its name. This reduces the outward pressure
which normally counteracts gravity, and the star suffers a precipitous
implosion which triggers the explosive burning of the remaining nu-
clear fuel in the star, leading to its complete and utter destruction
(Bromm 2003, 30). No corpse, either neutron star or black hole, is
left behind. This means that all the chemically enriched material con-
tained within the star is blown off into space, enriching the interstellar
medium.
Once the interstellar medium reaches a critical metallicity, normal
star formation begins, and stars of only a few solar masses (or lighter)
are possible. The Wilkinson Microwave Anisotropy Probe (WMAP), a
space telescope that makes very precise measurements in slight varia-
tions in the cosmic microwave background in different directions of the
sky, found evidence of this rst generation of stars in 2006. These hot,
massive stars ionized the interstellar medium, ripping electrons from
their host atoms. The photons of the cosmic microwave background in-
teracted with these electrons and were polarized in a unique and observ-
able way. This reionization of the ISM and resulting radiation pressure
made it harder for gravity to collapse clumps to form structures. Star
formation probably occurred at a slower rate, but more importantly this
might have affected the formation of larger structures such as globular
clusters and whole galaxies.
Althoughthe only evidence for these rst stars has so far beenindirect,
in their deaths they may have provided direct signs of their existence.
It has been suggested that those population III stars that had masses
either too low or too high to die as PISN would emit tremendous bursts
of gamma-rays before forming black holes. The NASA Swift satellite
is currently monitoring gamma-ray bursts, and it is hoped that many
candidate signals from the distant past will be observed over the next
few years.
The Evolution of the Universe 121
THE FATE OF THE UNIVERSE
The ignition of the rst population III stars heralded the beginning
of what astrophysicists Fred Adams and Greg Laughlin have dubbed the
Stelliferous (Star-lled) era. It will end when the last stars burn out.
The most frugal of all stars, low mass red dwarfs, have an estimated
lifespan of at least 10
13
years, some 1000 times the total lifespan of the
sun. It is also estimated that when the universe is between 10
12
to 10
14
years old, star formation will stop, as all available hydrogen gas dense
enough to collapse to form stars will have been used up. Therefore not
much longer than 10
14
years after the rst stars were born, the last ones
will die. The fate of galaxies near the end of the Stelliferous era is no
less gloomy. Individual galaxies within clusters will merge, as our Milky
Way may do with the Andromeda Galaxy in an estimated 6 billion years,
creating one large mass of dying stars.
The death of the last red dwarfs brings us to the start of the Degen-
erate era, in which the universe is dominated by stellar corpseswhite
dwarfs, neutron stars, and black holes. The name comes from the fact
that white dwarfs and neutron stars are held up from further collapse
by the degenerate pressure created by the Pauli exclusion principle.
The universe is locked in a dim twilight, occasionally punctuated by
the collision of two brown dwarfs, igniting a temporary period of nu-
clear fusion, or the supernova explosion of coalescing white dwarfs. The
Degenerate era comes to an end about 10
39
years after the universe be-
gins, the actual timing determined by the decay half-life of the proton.
Once all the protons contained within white dwarfs and neutron stars
have decayed (into positrons, neutrinos and other particles), the only
remaining stellar corpses are black holes.
In the Black Hole era, the only physical process of importance is the
slow evaporation of black holes predicted by Stephen Hawking. The
amount of time it takes for a black hole to decay into elementary par-
ticles increases as the mass increases. The largest theorized black holes
would contain the mass of an entire galaxy, and would take approxi-
mately 10
100
years to self-destruct. Note that the number 10
100
is called
a googol, not to be confused with the search engine of a similar name.
The demise of the black holes marks the end of the Black Hole era. Time
marches onward, as the universe enters the Dark era, with space-time
utterly empty with the exception of widely scattered photons, neutrinos,
electrons and positrons. As it was before the rst generation of stars, the
universe is a dark and lonely place, certainly not a comforting thought.
122 Cosmology 101
When poet Robert Frost pondered whether the universe would succumb
to re or ice, even his creative mind could not have imagined a fate so
bereft of hope. A more hopeful future appears in the classic short story,
The Last Question, where Isaac Asimov invokes a somewhat religious
solution to this scientic inevitability.
INFLATIONARY COSMOLOGY
Problems with the Big Bang
Although the big bang model was a success in tting with the cosmo-
logical observations of the 1960s, a small number of nagging problems
subsequently began to emerge. Four of the most important are the at-
ness problem, the horizon problem, the density perturbation problem,
and the magnetic monopole problem. The rst to be discussed was the
atness problem, rst posed in a lecture by Robert Dicke of Princeton
in 1969. It asks why the current density of material in the universe is
so close to the critical density needed for a at FriedmannRobertson
Walker model. In other words, why is so close to one? Based on
simple counts of the visible matter in the universe, one gets an ulti-
mate lower limit of 0.015, and taking into account the dark matter
contained in galaxies and galaxy clusters, this limit climbs to 0.20.3.
While these numbers may seem vastly different from 1, consider the
followingwhy are they not 0.000001 or 50,000? In terms of order of
magnitude, 0.015, and especially 0.2, are rather close to the special value
of one.
The importance of this seeming coincidence becomes clearer when
considering what itself determines. If >1, the universe is closed
because it is too dense and is doomed to recollapse, assuming there is
no cosmological constant. If <1, there is insufcient density to close
the universe, so it is open, and it will expand forever at an ever-slowing
rate. The case of =1, a at universe, is a nely tuned value, and is often
compared to balancing a pencil on its point. In the case of a balancing
pencil, any slight deviation or uctuation to one side or the other will
send it toppling over. Likewise, any slight wiggle from the absolute value
=1 will mushroom over time, driving farther away from one as the
universe evolves. In order for to be as close to 1 as we observe today,
it would have originally had to have been precisely equal to 1 to within
1 part in 10
49
or better. While it is certainly possible that the universe was
created with exactly this special value, there is no reason in the classic
big bang model why it should have chosen such an inherently unstable
The Evolution of the Universe 123
value, and some scientists regard the reason because it just happened
to be that way to be articial at best.
Another articial-appearing observation of the universe involves the
cosmological principle. On average, the universe should be homoge-
neous and isotopic, not only in terms of its matter, but also the cosmic
microwave background. The key words are on average. Measurements
of the CMB, the energy ngerprint of the hot early universe, have con-
sistently shown that it is extraordinarily smooth, to within a few parts in
100,000. It is possible that the initial state of the big bang was absolutely
uniform in temperature to this amount (i.e., it was created in remark-
able thermal equilibrium), but exceedingly unlikely. By comparison,
consider the water coming out of the tap in your bathtub. The water has
natural uctuations in temperature, sometimes running hotter, some-
times cooler. Before stepping into the tub, the wise bather either stirs the
water to help it equalize in temperature, or waits a few minutes for equi-
librium to be reached naturally. Now imagine how difcult this would
be if the bathtub were expanding! In order for the different parts of the
universe to reach equilibrium, they would have needed to communicate
with each other far faster than the speed of light.
This is also called the horizon problem, for the following reason. As
you look up into the night sky, you could (with a telescope of the greatest
theoretical power), see as far as 13 billion light years in each direction,
if the universe is 13 billion years old. This is because you can only see
light which has had sufcient time to reach your eye. A year from now,
you could see 13 billion + 1 light years in each direction, and so on.
This means that observable patches of the universe on opposite parts of
your sky are separated by 26 billion light years, or it would take 26 billion
years for a light signal to travel from one side of your horizon to the
other. How, then, can the CMB on opposite sides of the sky be at the
same temperature to within a few parts in 100,000? One might think it
would be possible for information to have been freely exchanged when
the universe was much smaller, but it was initially expanding so fast that
it contains at least 10
83
causally disconnected regions that have still not
had sufcient time to communicate.
Despite the remarkable smoothness of the CMB, it is not completely
smooth, otherwise we would not exist today. The minute uctuations
in temperature now observed are a reection of the fact that, while
homogeneous on the largest scales, matter clumps at smaller scales, in
the form of galaxies and clusters of galaxies. There must have existed
seeds in the early universe from which these structures grewnamely
124 Cosmology 101
uctuations or perturbations in the density of matter. The original big
bang model is silent on the source of these all-important deviations
from perfect homogeneity, sometimes called the density perturbation
problem.
Theories that seek to describe the early eras of the universe, namely
GUTs, are the source of the nal problem. Recall that they predict the
existence of massive particles namedmagnetic monopoles. The problem
is the high number predicted to exista trillion times more (by mass)
than all the observed matter in the universe! Increasing the density of
the universe by such a mammoth amount would not only make it closed
(dooming it to a ery big crunch), but it would have slowed the expan-
sion of the universe to the rate currently observed in about 30,000 years.
Given that the age of the universe is now estimated to be over 13 billion
years, there is clearly a serious problembetween theory and observation.
Not only is the predicted density of magnetic monopoles catastrophic,
but not a single magnetic monopole has yet been observed. In the words
of MIT physicist Alan Guth, like the unicorn, the monopole has con-
tinued to fascinate the human mind despite the absence of conrmed
observations. The monopole, however, has a much better chance of ac-
tually existing (1997, 30). It was the search to nd a solution to this
magnetic monopole problem that led Guth to develop a model which was
actually capable of solving all four problems.
Old Ination
In 1981, Guth was researching the details of spontaneous symmetry
breaking in the early universe, specically at the end of the GUT era
(10
35
seconds into the history of the universe) when the strong force
separated from the electroweak force. Specically, he investigated the
possibility that when the transition temperature at which the symmetry
breaking should occur was reached, the universe failed to immediately
go through the transition. A similar process can occur when water is
cooled below its freezing point yet remains liquid, as in the case of
freezing rain. This is termed supercooling. To understand what happens
if the universe is in a supercooled state, we need to revisit the concept
of Higgs elds.
It is simplest to picture a toy model where there are two Higgs elds,
which can be represented pictorially like a chessboard. The center of the
chessboard represents both Higgs elds having a value of zero. A third
dimensionheightcan nowbe added to our picture, representing the
energy density of space-time. The lowest possible value of this energy
density (remaining at on the chessboard) is called the true vacuum.
The Evolution of the Universe 125
If this is called the true vacuum, it suggests that there must be a false
vacuum. Picture a Mexican sombrero placed on our chessboard. It has
a slight dimple or dent at its top. For a bug crawling around the top of
the hat, the bottom of the dimple looks like the lowest possible point.
Scientists say that it represents a local minimum in the height of the
hat. However, a wider view shows that this is not the true lowest point
in the hatthe round trough in the brim of the hat (where it touches
the chessboard) has that honor. This is the global minimum of the hat.
The dent in the top of the hat is the false minimum or false vacuum,
whereas the trough in the brim is the true minimum or true vacuum.
Note that the dent at the center is a state of symmetry, as all the Higgs
elds are zero, whereas the brimof the hat is a state of broken symmetry,
as the Higgs elds are not zero. The surface of the hat represents how
the energy density depends on the values of the Higgs elds.
Now consider a small ball rolling on the surface of the hat. The act of
rolling from the top to the bottom represents a phase transition, from a
state of preserved symmetry to a state of broken symmetry. It is possible
that the ball could get stuck in the false vacuum in the top of the hat
and be unable to complete the transition to the broken symmetry state.
But recall that nature wants to break the symmetry, so there is a price to
be paid for being caught in the false vacuum. If the ball were a bubble of
the early universe it would be in a supercooled state. While stuck in the
false vacuum, the bubble would be under the inuence of a repulsive
force, andit wouldinate exponentially, growing 10
25
times bigger inthe
blink of an eye. Such an exponentially expanding universe is described
by the de Sitter solution to the Einstein eld equations. The universe
inates so much that the density of matter declines to essentially zero,
making the matterless de Sitter model a good approximation. However,
we obviously do not currently live in a de Sitter universe, so the bubble
must somehow get out of the false vacuum state and reached the broken
symmetry state of the true vacuum.
In Guths original inationary model, the bubbles cannot get out of
the false vacuumdimple by rolling upandover the hump, because they
dont have enough energy to do so. Instead, as shown in Figure 5.1, they
tunnel out through the hump. This is clearly impossible by the classical
laws of physics, but it is common enough in the strange world of quan-
tum mechanics. Quantum tunneling is the basis of such technological
advances as the tunnel diode in electronics and the scanning tunneling
microscope. Inthis so-called old ination, the universe tunneled out one
bubble at a time, and somehow the bubbles collided and recombined on
the other sidein the broken symmetry state. However, as Stephen
126 Cosmology 101
Figure 5.1 Old vs. New Ination
Hawking and others later showed, the bubbles would never actually
merge. Thus old ination suffered from what is called the graceful exit
problemthere is no realistic way to end the inationary phase.
New Ination
The new ination model, independently proposed by Andreas Al-
brecht and Paul Steinhardt of the University of Pennsylvania, and Rus-
sian physicist Andrei Linde, avoided this problem by not having bubbles
try and reconnect. Instead, the entire visible universe was spawned from
a single bubble of false vacuum. The mechanism for leaving the false
vacuum state was different as well. Instead of an energy barrier which
had to be tunneled through, the shape of the hat included a long,
at region, giving this model its sometime name of the slow rollover.
Quantum uctuations would start the ball rolling, in a literal sense,
slowly driving the bubble off of the false vacuum plateau, inating as it
rolled. When the ball reaches the bottom of the hill (the brim of the
sombrero), it would oscillate back and forth, reheating the supercooled
universe to the level expected by the big bang model. Some of the
energy would be converted into particles, repopulating the now-empty
universe.
Ination was developed in response to problems with the original big
bang model. How successful was it in solving those problems? At the end
of the inationary period, the density of any particles created before-
hand is diluted to essentially zero by the huge increase in the volume
of the universe, thus solving the magnetic monopole problem. This also
explains why there has as of yet been no experimental observation of
magnetic monopolesthe nearest one could nowbe many parsecs away!
The incredible expansion of the universe also solves the smoothness
problem in much the same way. Any inhomogeneities in the CMB which
The Evolution of the Universe 127
existed prior to ination would be smoothed out, just as wrinkles in fabric
(or skin) disappear when it is stretched. Another way of looking at the
problem is that it is no longer assumed that the entire visible universe
arose froma state of thermal equilibrium, only that what we see was born
from a tiny patch of the initial big bang which, in itself, had achieved
a constant temperature over its small volume. A solution to the atness
problem follows close behind. Any large-scale curvature to space-time
which was created by the initial big bang would be attened out, as
the surface of a sphere (like Earth) looks at when it becomes large
enough.
The density perturbation problem was the most complicated to solve,
and it was nally realized that the Higgs eld itself is not the eld respon-
sible for ination. Instead, there must be some other weakly coupled
eld, dubbed the inaton eld, which drives ination. The exponential
expansion of the universe ends when the phase transition is made from
false vacuum to true vacuum, or, in pictorial terms, the inaton eld
ball reaches the steep part of the energy hill. Due to quantum uc-
tuations, these uctuations would not be the same everywhere. These
uctuations would be the source of small uctuations in the density of
matter in the early universe, which would in turn act as the seeds for the
formation of larger structures such as galaxies.
A common misconception about ination is that it replaces the big
bang. It is actually a modication of the classic big bang model, devel-
oped to answer several problems and nagging questions. Although the
exponential expansion which gives the theory its name only lasted for
an innitesimally small fraction of a second, its effects are visible for
the entire history of the universe. After ination, the universe evolves
as predicted by the big bang model. Cosmologist David Schramm called
ination pretty, noting that It really gave a boost to this interface of
cosmology and particle physics, capturing some of the most exciting
aspects of both (Lightman and Brawer 1990, 445).
Eternal Ination
Ination is also not a monolithic theory, but a paradigm or class of
theories that can derive from different sources and make measurably
different predictions for the universe we inhabit. For example, the uni-
verse could have entered an inationary state by supercooling fromhigh
temperatures (as recounted above), or, as in the research of Alexander
Vilenkin of Tufts University, the result from tunneling from nothing
directly into a false vacuum state. Here nothing literally refers to a
state of no classical space or time. A third possibility, developed by
128 Cosmology 101
Andrei Linde, is called chaotic ination. In this version of ination,
the inaton eld can nd itself in a variety of different energy levels at
different locations due to random uctuations. In some places, it hap-
pens to be in a state of higher energy, which acts like a false vacuum
state and that region of space-time would inate. In chaotic ination,
different universes are born from this initial state at different times and
individually evolve into their own separate, observable universes.
This aspect of chaotic ination was later generalized to include all
versions of ination. It is now believed that once ination begins, it will
never end, a concept called eternal ination. The inationary era was orig-
inally expected to end because the false vacuum is inherently unstable
and decays, similar to a radioactive isotope. Like the decay of uranium
or plutonium, the rate of decay of the false vacuum is mathematically
exponential, but recall that the rate of expansion during the ination-
ary era is also exponential. It turns out that in most models the rate
of inationary expansion is signicantly faster than the decay of the
false vacuum, with the overall effect that the false vacuum actually gets
bigger, not smaller, with time! Individual pockets of the false vacuum
will decay at different times, each leading to a separate universe, dubbed
pocket universes. The collection of all pocket universes that derive from
the original false vacuumis sometimes called the multiverse. According to
eternal ination, at all moments, some pocket universe is being created
by undergoing ination, and once formed it evolves independently of
all its sibling universes. For example, our pocket universe began about
13 billion years ago in an inationary era we generically call our big
bang, and as far as we know our universe will continue to exist (albeit
becoming a darker and lonelier place) for an innite period of time.
Does this mean that ination has always occurred somewhere in our
multiverse? Has a discussion of the ultimate origin of the universe really
been avoided? It appears that, at least in simple ination models, there
is still a nite beginning somewhere in the distant past, and ination is
only eternal in the future direction.
No matter how pretty (or mindblowing) the theory, the scientic
method demands that the predictions of any theory be tested to see if
it is a consistent and useful description of the observable world. Among
the testable predictions of most inationary models are the follows:
1. = 1
2. The spectrum of density perturbations produced should be nearly
(but not exactly) scale invariant. This means that although the
density perturbations would come in different sizes, smaller ones
The Evolution of the Universe 129
would contribute in nearly the same way as the large ones. This is
the same form assumed in most models of structure formation in
the early universe.
3. There should exist a distinctive background of gravity waves left
over from the inationary epoch.
In the early 1990s, ination was left hanging in terms of observations.
was considered to be closer to 0.3 than 1, there was no precise mea-
surement of the density perturbations, and gravity waves had not been
detected. Yet many cosmologists felt condent that it was only a matter
of time before observations caught up to theory.
DARK MATTER REDUX
The increasing evidence for dark matter has already been laid out
for the reader, with the result that by the 1990s it appeared that the
universe was dominated by dark matter over visible matter by a factor
of 20:1. The thought that 95 percent of the universe is made of some
invisible substance only detectable through its gravitational inuence
is certainly unsettling, and leads to an important questionwhat is the
nature of this mysterious material?
A natural suggestion was that it was made of some small, dark nuggets
of normal matter that emit little to no light, such as planets, brown
dwarfs, and stellar corpses. These candidates for dark matter were
lumped together under the term MACHOS, short for massive astro-
physical compact halo objects. Several studies (with rather tongue-in-
cheek names such as EROS, OGLE, and DUO) utilized microlensing to
search for MACHOS. In microlensing, a small object like a planet or
star passes directly in front of another small object, causing a distinctive
brightening of the image through a gravitational focusing of the light.
Unfortunately, although these studies did discover MACHOS, it was in
disappointingly small numbers. It was clear that MACHOS could not
contribute signicantly to the dark matter.
There is also a theoretical reason for discounting MACHOS as the
main contributor to dark matter. Normal (or baryonic) matter is com-
prised of neutrons and protons. The percent of deuterium generated
during big bang nucleosynthesis is exceedingly sensitive to the density
of baryonic matter. Observations of the current amount of deuterium
in the universe (and other light isotopes created in big bang nucleosyn-
thesis) limit the amount of baryonic matter in the universe to about

B
= 0.04 to 0.05. This means that while a small amount of dark matter
130 Cosmology 101
can be in the form of normal material made of protons and neutrons,
the vast majority must be made of some exotic material.
Nonbaryonic dark matter candidates can be broken into two classes
hot dark matter (HDM) and cold dark matter (CDM). The former is called
hot because its particles move very fast, near the speed of light, and
the latter cold because they are more sluggish. The main candidate for
HDM is the neutrino, which has such a small mass that it travels close
to the speed of light. Unfortunately, because HDM travels so quickly
it resists clumping together into structures unless there are a lot of
particles. If HDM dominates the universe, it would tend to smooth out
any clumping at small scales in the early universe, with the result that
huge, attened structures several hundred megaparsecs in diameter
would form rst in the universe, only to slowly fragment into smaller
structures such as galaxy clusters and individual galaxies over time. This
top-down or pancake model of structure formation was rst proposed
by Russian theorist Yakov Zeldovich in 1970.
Neutrinos have one very specic advantage as a dark matter
candidatethey are denitely known to exist! The leading candidates
for cold dark matterweakly interacting massive particles (WIMPS)do
not enjoy this luxury. The favored candidates for WIMPS are the su-
persymmetric partners predicted by supersymmetry. For example, each
neutralino, the SUSY partner of the neutrino, would have a mass 100
1000 times that of the proton. The problem is that SUSY is still a spec-
ulative theory and no SUSY partners have been discovered. Another
candidate for WIMPS is the axion, the as-yet unseen particles predicted
by theories that seek to explain how matter won out over antimatter.
Despite this rather sizeable (but not insurmountable) problem, cold
dark matter is the reigning explanation for dark matter for one very
important reasonit correctly predicts the structures seen in the uni-
verse today. Since CDM moves slowly (relatively speaking), it clumps
together much more readily than HDM, predicting that smaller struc-
tures such as galaxies should form rst, and later aggregate into clusters
and larger structures. More specically, galaxies formwithin dark matter
haloes, condensing out of gas and later fragmenting into stars. The dark
matter halo remains surrounding the galaxy, as in the case of our own
Milky Way. The dark matter haloes of individual galaxies later gather
together to form galaxy clusters, and so on, resulting in the pattern of
laments seen in redshift surveys. In addition, surveys of galaxies and
galaxy clusters show that most galaxies were formed at redshifts 1 to
4, with the earliest formed within the rst billion years of the universe
(z >6.5), while galaxy clusters formed much more recently, around z 1
The Evolution of the Universe 131
or less, and superclusters are still coming together, consistent with a
CDM-dominated universe.
DARK ENERGY
The Accelerating Universe
Despite the successes of the CDM model of the universe, there were
still serious problems in the late 1990s. Firstly, recall that the ination
predicted that = 1 (the universe is at) yet even with dark matter the
universe appeared to be open, with 0.3. In addition, the ages of
globular clusters seemed older than the age of the universe predicted
from the Hubble constant, which was settling down at a value around
75 km/sec/Mpc. All these issues could be addressed if Einsteins long-
maligned cosmological constant was not zero, but most cosmologists
were averse to involving this fudge factor without signicant obser-
vational evidence. Such evidence was discovered by two independent
teams of astronomers in 1998.
Recall that the value of Hubbles constant is expected to change over
the evolution of the universe, as the expansion of the universe changes
speed. It was expected that the expansion of the universe should be
slowing down, or decelerating, hence the variable that measures the
rate of change of the expansion was dubbed the deceleration parameter
(q). This parameter itself can change values as the universe ages. In
order to measure q, cosmologists need to plot a Hubbles law diagram
looking far into space (literally back in time) for redshifts greater than
z =0.1, where Cepheids are too distant to be seenand type Ia supernovae
are the most trusted standard candles. What the two teams found was
astoundingdistant supernovae were fainter than they should be given
their distance, meaning the universe had expanded faster, not slower,
during the intervening time. In other words, in the recent past (and still
today), the expansion of the universe is accelerating, not decelerating,
and the value of q is therefore negative. More curious still, supernovae
farther away (z 1.7) looked brighter than would be expected for an
accelerating universe, whichmeant that back that far intime the universe
was still decelerating!
Stop, take a breath, and close your eyes for a moment, because the
universe is about to get a whole lot weirder in the next sentence. The
observations described in the previous paragraph seem best explained
if the universe really is at (=1), but nearly 30 percent of the universe
is composed of dark matter and most of the other 70 percent is made of
some other invisible component which acts as a repulsive force, called
132 Cosmology 101
dark energy. Based on the supernova observations, the expansion history
of the universe can be divided into three broad epochs. During the rst
few hundred thousand years of its history, the evolution of the universe
was dominated by the radiation (photons) it contained, and the expan-
sion of the universe slowed down over time. After this, the evolution
of the universe was matter-dominated until z 0.5, corresponding to
about 4 billion years ago. During the matter-dominated epoch the ex-
pansion rate of the universe continued to slow down, but at a slightly
different rate than occurred in the radiation-dominated era. About 4
billion years ago, dark energy began to dominate the evolution of the
universe, and the expansion rate began to accelerate. The reason for this
is that unlike radiation and matter, whose density thins out as the uni-
verse expands, dark energy is a quality of space itself, so its density stays
the same. At some point, the universe had expanded so much that the
density of the dark energy was nally greater than the density of matter
and radiation, and its repulsive nature began to dominate the evolution
of the universe. Because of this accelerated expansion, the universe is ac-
tually older than one would predict from the current value of Hubbles
constant, and there is no longer a conict between the age of the uni-
verse and the ages of globular clusters. This is reminiscent of Georges
Lematres use of a cosmological constant to try and relate the high
values of H
0
derived in the 1920s and 1930s to the accepted age of the
Earth.
Several questions immediately come to mind. Firstly, how reliable are
these supernova measurements? Supernovae seen at large distances are
seen as they were much earlier in the universe. Are supernovae created
by previous generations of stars the same as supernovae that occurred
more recently? Is interstellar or intergalactic dust skewing the brightness
measurements? While astronomers investigate these important points,
alternate ways of afrming the acceleration of the universe continue to
be explored.
Candidates for Dark Energy
Perhaps more immediate is the problem of guring out the nature of
this dark energy. Many cosmologists simply use the terms dark energy
and cosmological constant interchangeably, and even use the Greek
letter lambda (), which was used in equations to denote the cosmolog-
ical constant, in describing dark energy. But care needs to be taken. The
cosmological constant is generally pictured in a very specic wayas
the vacuum energy (the inherent energy density of the vacuum itself).
Recall that the vacuum is not really empty, but is instead a seething sea
The Evolution of the Universe 133
of virtual particle-antiparticle pairs being created and annihilated due
to quantum effects. However, particle physics currently predicts that
when adding up all the contributions of all possible particle-antiparticle
pairs, the energy density of the vacuum should be more than 10
120
times
greater than the energy contained in all the normal matter in the uni-
verse! If the cosmological constant/dark energy were really that large, it
would rip apart all matter in the universe, down to the level of protons
and neutrons. The fact that the predicted and observed limits of the vac-
uum energy density differ by such a catastrophic amount is sometimes
called the cosmological constant problem. As theorist Leonard Susskind of
Stanford University quipped,
Being wrong by one order of magnitude is bad; two orders a disaster; three,
a disgrace. Well the best efforts of the best physicists, using our best theo-
ries, predict Einsteins cosmological constant incorrectly to 120 orders of
magnitude! Thats so bad that its funny. (2006, 66)
If dark energy isnt the vacuum energy, then what is it? One popular
suggestion is called quintessence. This is a proposed new type of parti-
cle or eld whose properties change in time as the universe evolves.
Quintessence acts like the inaton eld, in that it has a negative pres-
sure and is repulsive, accelerating the expansion of the universe. It is
much weaker than the inaton eld, so it does not cause as great an ac-
celeration as during ination. The fact that the quintessence eld varies
also differentiates it from being just another version of ination. Since
quintessence changes, it is possible that if it were the correct explanation
for dark energy that the acceleration currently observed could cease in
the future. A third contender is actually a unied explanation for dark
energy and dark matter called quartessence, based on theories of ex-
otic uids. Finally, the strangest candidate is called phantom energy, an
extremely unstable quantum eld that would cause so much acceler-
ation that it could cause the Big Ripthe shredding of all structures
down to the atomic level. Some comfort can be gained from the fact
that if such a runaway dark energy does exist, we would not have to fear
the evisceration of the universe for another 20 or so billion years. The
reader should take from this discussion the idea that dark energy is an
active playground for theorists, and until observations can denitively
discriminate between the predictions of these different candidates, the
theoretical gymnastics will continue.
Putting aside the rather obvious demise predicted by the Big Rip,
how does the existence of dark energy change the future evolution of
134 Cosmology 101
the universe from that described previously? The key difference is that
a continued accelerated expansion will create a horizon around every
observer in the universe, a distance beyond which one cannot observe
anything because light from those objects has not had time to reach
himor her. One by one galaxies will disappear behind this impenetrable
curtain. Computer models show that if the universe is dominated by a
cosmological constant, by the time it is about 100 billion years old, the
only galaxies we will be able to observe fromEarth are those of the Local
Group, or more accurately, whatever exists in our neighborhood after
the merger of the Milky Way and M31.
THE CONCORDANCE MODEL
Ination, cold dark matter, dark energythree extraordinary con-
cepts which combined have given cosmologists the current best descrip-
tion of the universe we inhabit. Called the inationary CDM model, it
is based on a number of measurable parameters, including the Hub-
ble constant, the deceleration parameter, the age of the universe, the
temperature of the cosmic microwave background, and the densities
of baryonic matter, dark matter, neutrinos, and dark energy. As with
any scientic model, it must be vigorously tested, and as we shall see,
it certainly has. Because of its success in meeting every major observa-
tional challenge thrown its way, the inationary CDM model has been
dubbed the concordance model. At this moment, its weakest link is ac-
tually the assumption that dark energy is described by a cosmological
constant modeled on the vacuum energy. Much of the observational
support for the concordance model has come from exquisite details in
the cosmic microwave background, caused by sound waves in the early
universe, a curious modern twist on the ancient idea of the music or
harmony of the spheres.
Acoustic Waves and the Early Universe
Aristotle wrote in his Metaphysics how the Pythagoreans ascribed nu-
merical values to the musical scales, which could then be related to the
universe at large, especially the motions of the planets. This music of
the spheres could not be heard by any but the especially gifted (report-
edly including Pythagoras himself). Aristotle himself believed the idea to
be beautiful but illogical. Ptolemy, as well, wrote in the third book of his
Harmonica that the motions of the stars matched up with musical scales.
Boethius, the Roman philosopher, wrote about three types of music:
that of the universe, human music, and music created by instruments.
The Evolution of the Universe 135
He believed that the rst was the most important to study in order to
understand the motions of the heavens.
Johannes Kepler gave the idea serious consideration (one might say
obsessively so), in his Harmonices Mundi. He discovered that the ratio of
certain properties of planetary and lunar motions were approximately
the same numerical value as that between the notes in chords. On the ba-
sis of the observational relationships he found, he constructed musical
scales composed of notes representing each of the planets. Interestingly,
in order to do so, he had to ignore the octave of the notes (e.g., lowering
the sound of Mercury by six octaves in order to t it into the scale). De-
spite the ideas poetic qualities, the concept of the music of the spheres
eventually became little more than a historical footnote in the minds of
scientists until modern day. With the rise of the concordance model,
the existence of sounds generated in the early universe became an
important means of testing the predictions of modern cosmology.
Sound is created by the propagation of pressure waves through a
medium, such as air. Since the word sound is usually thought of in
connection with the sense of hearing, a more generic term is acoustic
wave, which encompasses frequencies both too high and too low for
the human ear to register. As the trailers for the lm Alien correctly
warned, in space, no one can hear you scream. In general, interstellar
space is currently so close to a perfect vacuum that sound waves cannot
propagate. However, in astronomical locations where the gas density
is high enough, acoustic waves can and do occur. For example, sound
waves can travel through the atmospheres of Venus, Mars, and Saturns
moon Titan, as well as Earth. Stars, as hot spheres of dense gas, are also
laboratories for the detection of acoustic waves. In 1962 it was discovered
that the visible surface of the sunits photosphereis bubbling with a
period of about 5 min. This ve minute oscillation was found to be
a global rather than local phenomenon caused by sound waves (called
p-modes) traveling through the sun. Helioseismologists have monitored
these oscillations for a number of years through the Global Oscillation
Network Group of ground-based telescopes (GONG) as well as the Solar
and Heliospheric Observatory (SOHO). Such studies allow astronomers
to map the suns interior.
But how could acoustic waves form in the vacuum of space, and what
do they have to do with radiation? The key point is that although in-
terstellar space is unbelievably cold and relatively empty at this point in
the universes evolution, during the rst few hundred thousand years or
so it was nothing of the sort. In contrast to the popular misconception
136 Cosmology 101
fostered by the term big bang, the universe began in silence, because
there were no initial acoustic waves. However, inherent inhomogeneities
in the density of matter and energy are predicted by inationary mod-
els. These small density perturbations created a series of gravitational
valleys. Light and matter (bound together in the opaque fog of the
universe before recombination) fell into the valleys and compressed.
The compressed radiation exerted pressure and the matter and radia-
tion shot out of the valley, only to settle into another valley. Dark matter,
which does not interact with matter and radiation, also settled into the
valleys, gradually making them bigger by its gravity. Because matter was
compressed and expanded, it generated pressure waves, and because
the radiation was coupled to the matter, the behavior of the radiation
was affected by these pressure waves.
Results from the Wilkinson Microwave Anisotropy Probe (WMAP)
Our story now returns for the moment to more familiar ground,
the cosmic microwave background. Detailed measurements of the CMB
were made by the Cosmic Background Explorer (COBE) in 1992, which
veried the effective temperature of the radiation (2.725 K) and discov-
ered slight variations in temperature across the sky to about one part
in 100,000, a reection of slight density perturbations in the early uni-
verse, like those predicted by ination. It also found that over angular
scales of 1090 degrees there was nearly equal power on each scale,
agreeing with the nearly scale-invariant spectrum predicted by ination.
Scientists who supported ination were understandably ecstatic. George
Smoot, the lead investigator of the project, told journalists, if youre
religious, its like seeing the face of God. Stephen Hawking called it
the discovery of the century, if not of all time (Singh 2004, 462463).
Not surprisingly, COBE scientists Smoot and John Mather were awarded
the 2006 Nobel Prize in Physics for these discoveries. Angular variations
or anisotropies smaller than about 7 degrees unfortunately could not
be seen because of the satellites limited resolution. It was these smaller
angular anisotropies which would later provide some of the strongest
evidence for ination and the concordance model.
It has become standardprocedure todescribe variations inthe CMBin
terms of mathematical functions called spherical harmonics. An analogy
can be found (in a single dimension) in a guitar string. The fundamental
tone of the string is such that the entire string moves back and forth
except the two points where the string is connected to the guitar. The
rst harmonic tone occurs when there is one stationary point on the
string other thanthe ends, and so forth. Interms of spherical harmonics,
The Evolution of the Universe 137
Figure 5.2 Angular Power Spectrum of the Cosmic Microwave Background
these are described by the angular frequency l, where l = 0 represents
the fundamental tone, the rst harmonic, l =1, etc. In terms of the
CMB, cosmologists plot the intensity of temperature variations versus l,
creating a graph, as seen in Figure 5.2, called the angular power spectrum.
Since l = 180/, the larger the value of l (or multipole, as it is called),
the smaller the angular size of the variations. The most powerful signal
in the CMB is the dipole radiation (l = 1, =180

). In one direction of
the sky the CMB is 3.36 mK (thousandths of a Kelvin) warmer than in
the other, due to the fact that the solar system is moving in this direction
at 370 km/sec. This signal is more than 100 times stronger than other
angular anisotropies and does not give us information about the early
universe, so it is subtracted out before the angular power spectrum is
graphed. Since COBE could not see angles less than 7 degrees, it could
not pick up multipoles larger than about 25 or 26. The ability to see
higher multipoles is vitally important, as theory predicts acoustic waves
in the early universe led to several pronounced peaks at multipoles
higher than l = 50 (or angles smaller than a few degrees).
The sizes of the primordial acoustic pressure waves varied, from the
longest or fundamental to smaller harmonics, and depended on the
speed the waves traveled and how much time had elapsed from the start
138 Cosmology 101
of the universe. Since the speed of sound in this dense soup was
approximately 60 percent the speed of light, in the estimated 400,000
year spanof this acoustic era the longest waves (anddeepest tones) that
could be generated were approximately 220,000 light years in length.
This should appear in the CMB as a noticeable peak at around 1 degree
(or l 220). Smaller peaks should also appear at multiples of this
multipole (or fractions of a degree). These are collectively known as the
acoustic peaks. At recombination when electrons formed stable atoms
and radiation ran free, the speed of sound slowed dramatically, and
eventually stopped, and the waves froze into place as the pressure
ceased. Matter then fell into the gravitational wells, which had been
enlarged by dark matter, and entered the road to the formation of
astronomical structures. Multipoles over l 1000 are smeared out and
blend together into a smooth curve. The acoustic waves responsible for
these multipoles are sosmall that their effects are overrunby interactions
between photons and electrons. This is called Silk damping.
Thanks to the Wilkinson Microwave Anisotropy Probe (WMAP),
launched in 2001, the existence of the rst three acoustic peaks has
been conrmed and the angular location and relative strengths of the
rst two peaks are precisely known. This is vitally important, because the
exact angular location of the acoustic peaks as well as the relative heights
of the peaks depend very dramatically on cosmological parameters such
as the densities of matter and radiation and the geometry of the uni-
verse. The details of the WMAP angular power spectrum are not only
strong evidence for the concordance model, but help to further hone
the values of some of its important parameters. For example, the WMAP
data ts exceedingly well with a at ( = 1) universe where normal
matter made of baryons contributes 4 percent, dark matter contributes
26 percent, and dark energy makes up the majority of the density of
the universe, at 70 percent. The exact identity of dark energy cannot be
conrmed based solely on the WMAP observations, but various models
under consideration are somewhat constrained by the observed data. It
also supports a value of Hubbles constant around 73 km/sec/Mpc, in
agreement with the Hubble Space Telescope studies. The best estimate
for the age of the universe in 13.7 billion years, in agreement with the
most recent studies of the ages of globular clusters. WMAP gives addi-
tional support for ination (although it has not reached the sensitivity
necessary to distinguish between the predictions of the various models
of ination). Recall that in general ination predicts a nearly scale in-
variant spectrum of density perturbations, which should be reected in
the cosmic microwave background. WMAP has conrmed that this is
The Evolution of the Universe 139
the spectrum observed, providing some of the strongest evidence yet for
ination.
Not only have the WMAP observations provided evidence for the
inationary CDM model, but they have also largely ruled out other
competing models. For example, Israeli physicist Mordehai Milgrom
proposed an alternative to dark matter in the 1980s. Known as Modied
Newtonian Dynamics (MOND), the theory postulates that, on the scales
of galaxies, nature does not exactly follow Newtons law of gravity, with
these modications giving rise to observations such as the at rotation
curves of spiral galaxies that most cosmologists attribute to dark matter.
Not only does MOND fail to account for observations supporting dark
matter in galaxy clusters and has difculty reproducing the currently
observed large-scale structure, but it also predicts a much lower third
[acoustic] peak than is observed by WMAP (Spergel, et al. 2006, 14).
Other Observations
So far we have focused on the relic anisotropies in the cosmic back-
ground radiation caused by the acoustic waves generated by matter be-
fore recombination. When recombination occurred and the radiation
decoupled, clumps of matter were expanding and contracting indepen-
dently and were frozen in various stages of this pattern. This is best
understood by tossing a pebble into a pond, and after the ripples have
expanded some distance, freezing them into place. There should there-
fore be similar acoustic peaks in the distribution of galaxies. This was
veried by the Sloan Digital Sky Survey (SDSS) in 2005. In agreement
with the predictions of the inationary CDM model, there is a 1 per-
cent overdensity of galaxies with separations of 150 Mpc. The SDSS
and other large-scale galaxy surveys have also veried two other cosmo-
logical predictions involving observations of distant galaxies. The rst,
called cosmic shear, is a weak kind of gravitational lensing. Fluctuations
in the distribution of dark matter throughout the universe distort the
shapes of distant galaxies but do not cause enough lensing to create
multiple images, as in so-called strong gravitational lensing. Observa-
tions of cosmic shear support the predictions of the inationary CDM
model. The second observation, related to cosmic shear, is called cosmic
magnication. Light from distant quasars is focused by intervening dark
matter in such a way as to make the quasars appear slightly brighter
than they should otherwise. The magnication of 200,000 quasars was
discovered by SDSS in 2005. The results were also in agreement with the
concordance model.
140 Cosmology 101
Has cosmology been solved? Is the concordance model the ultimate
description of the universe? The scientic community has certainly
learned a valuable lesson from the rather conceited comments of Lord
Kelvin, who in1900 declared that There is nothing newto be discovered
in physics now. All that remains is more and more precise measurement
(Davies and Brown 1988, 34). It is generally accepted that the predic-
tions of the inationary CDM model have been quite successful at
tting the observed universe, but there are still observations to be con-
ducted with improved technology. For example, the direct detection of
gravity waves will provide a serious test of the model. It should also be
noted that the WMAP observations taken in isolation do not demand
that dark energy exists, but rather conrm that the data ts very well
(some might argue best ts) with a at universe composed of 70 percent
dark energywhatever form it might take. Further observations of the
CMB and distant quasars and galaxies should eventually be able to dif-
ferentiate between different versions of ination as well as the different
explanations for dark energy. So even if the inationary CDM model
emerges as the undisputable champion of cosmological theories, there
is enough work to do to keep researchers and graduate students busy
for years to come.
But what of dissenting voices, and others who constantly push the
boundaries of current knowledge? In science, multiple models emerge
at any given time and are subjected to scrutiny and peer review. No
model is considered nal, but always opens the door for further avenues
of exploration. Our journey now turns to these speculations.
6
The Road Goes Ever
On: Continuing
Speculations
MULTIPLE DIMENSIONS
According to Einsteins general theory of relativity, space and time are
interwoven into the exible four-dimensional fabric of space-time. But
couldthere be more dimensions of space that we donot experience inev-
eryday life? In the 1921, Polish mathematician Theodor Kaluza utilized
an extra dimension of space to try and unify gravity and electromag-
netism. In 1926 Swedish mathematician Oskar Klein demonstrated that
this extra dimension was unobservable because it was rolled up, or com-
pactied, into a circle the size of the Planck length (10
35
m). Unfortu-
nately their unication scheme did not succeed, but it opened the door
for further investigations of the physics of higher dimensions. Princeton
theorist Edward Witten warns that the idea of extra dimensions might
sound a little bit strange to anyone who hasnt studied physics. Anyone
who has gone into physics professionally, will know that there are
many things that are a lot stranger than extra dimensions (Davies
and Brown 1988, 101). We will now put his observations to the test.
String Theory
In 1968, Italian physicist Gabrielle Veneziano proposed a theory that
involved extra dimensions as an explanation for the every-growing num-
ber of hadrons (before the acceptance of the quark model). His model
became known as string theory, and visualized the interactions of the
strong force in terms of bosonic strings. In 1970 John Schwarz and
Andr e Neveu developed a fermionic string theory, opening the door
for further study. With the success of QCD, string theory was no longer
necessary to describe the strong force, and seemed doomed to the
dustbin of the scientic method. John Schwarz and Joel Scherk rescued
142 Cosmology 101
it in 1974, recognizing that string theory was actually a theory that in-
cluded gravity, leading to its revision as a candidate for a unied theory
of particles. Each unique frequency of vibration of a tiny bosonic or
fermion string (each only about 1035 m long) was postulated to be
what we observe as a different elementary particle. But since bosons me-
diate forces and fermions make up matter, having two separate theories
seemed contradictory to a successful model of unication.
String theory fell out of favor until 1984, when Michael Green and
John Schwarz combined string theory with supersymmetry, creating a
theory that unied bosonic strings and fermionic strings into a theory of
superstrings. Superstrings could be closed (have their ends join together
to form a tiny loop) or open (the ends remain unattached). Closed
strings appeared to give rise to gravitons, the supposed particle which
mediates gravity in models that unify general relativity and quantum
mechanics. The price to be paid for this seeming unication was that
superstring theory was mathematically consistent in ten dimensions,
while the real universe seems to exist only infour (three of space and one
of time). Following the work of Kaluza and Klein, the extra dimensions
were saidtobe curledupintotiny little knots far toosmall tobe observed,
called Calabi-Yau manifolds.
Superstrings soon became the new fashion in theoretical physics, but
not without making vocal enemies. Proponents hailed its elegant mathe-
matics, but beauty alone is not a reason for accepting a highly speculative
theory. Indeed, RalphAlpher andRobert Hermanof big bang fame warn
that
Science brings some problems upon itself. For example, scientists are fre-
quently attracted by a theory because they nd it beautiful, when in fact
it may be wrong. . . . All this reects the fact that the doing of science is a
human and creative process. (2001, 173)
Leading theoretical physicist Stephen Hawking cut to the chase and
declared string theory pretty pathetic for its lack of ability to make
predictions observable with current technology (Hawking and Penrose
1996, 123). Others charged that superstring researchers were receiving
an unfairly large amount of available grant money (much of it received
from the U.S. Departments of Defense and Energy) and were snatching
up some of the best graduate students for research that was speculative
at best. String theorist Gary Horowitz noted that although many of the
remarks by colleagues have been disparaging, the large number of
The Road Goes Ever On 143
them at least seems to indicate that string theory is quite important!
(Hawking and Penrose, 1996, 135).
Despite the strong opinions of physicists on both sides of the string
controversy, in the end only the scientic method truly matters. But
string theorists have openly admitted that superstring theory does not
neatly t into the hypothesis-predict-test model of science. MITs Lisa
Randall explains that
An enormous theoretical gulf separates string theory, as it is currently
understood, from predictions that describe our world. String theorys
equations describe objects that are so incredibly tiny and possess such ex-
traordinarily high energy that any detectors we could imagine making with
conceivable technologies would be unlikely ever to see them. Not only is
it mathematically tremendously challenging to derive string theorys conse-
quences and predictions, it is not even always clear how to organize string
theorys ingredients and determine which mathematical problem to solve.
(2005, 69)
It is therefore possible that string theory may have little to do with
our observed universe, and it would be very difcult to know one way or
the other. Despite the serious challenges in directly testing superstrings,
there are several predictions hopefully within the grasp of reasonable
technology to test. Superstring theory predicts a specic spectrum of
gravity waves, which may be veried or refuted once gravity waves are
directly detected. Superstring theory also predicts small imprints on the
cosmic microwave background that may be detectable with improved
technology. On the negative side, if the concordance model of cosmol-
ogy continues to gain further observational support, string theory will
be thrown into question, because superstrings are generally not thought
to be consistent with the existence of a cosmological constant.
M-theory and Braneworlds
For several reasons, superstring theory was once more out of fashion
in the early 1990s, not the least of which being that there were ve com-
peting and distinct models. If superstrings were purporting to unify the
forces of nature, shouldnt they rst unify themselves? A possible answer
sooncame inthe formof the holographic principle, discoveredby Dutch
physicist Gerard t Hooft. This principle asserts that for some region of
space, all the information contained inside it can be represented by
the regions boundary. This is similar to a hologram, a two-dimensional
144 Cosmology 101
projection of a three-dimensional object made with lasers. One of
the most important special cases of the holographic principle is the
AdS/CFT correspondence or duality. Also named the Maldacena con-
jecture, after its Argentine creator, Juan Maldacena, the acronym stands
for Anti-de Sitter/Conformal Field Theory.
Anti-de Sitter space is a variation of the de Sitter space used in in-
ationary models, and has a negative cosmological constant. It has a
relatively simple geometry, which means that its properties have been ex-
tensively studied, and even though it is not an especially realistic model
of the observed universe, calculations done using this space are more
simple than in other geometries. A conformal eld theory is a theory of
elds in which the equations have a particular mathematical symmetry
(i.e., the theory does not vary under certain mathematical transforma-
tions, such as multiplying all the factors by the same constant or rotating
them by the same amount). A duality connects two apparently different
physical theories and shows that they are equivalent. This is extremely
useful when one theory is easier to calculate than the other. The most
widely known example in physics is electromagnetism. The electric and
magnetic elds (and the corresponding charges) can be interchanged
in Maxwells equations and the equations will remain the same.
Using these mathematical principles, Princeton string theorist Ed-
ward Witten demonstrated in 1995 that the ve superstring theories
as well as supergravity theory itself were nothing more than special
cases of a much more fundamental, underlying theory in eleven di-
mensions which he dubbed M-theory. The reason for the name of the
theory, as well as its precise characteristics, are not well understood.
The M is alternately said to stand for magic, mystery, or membrane,
according to taste although critics might prefer mythical or mys-
tical (Naeye 2003, 40). Others have suggested it stands for mother
of all theories. The key is that M-theory predicts that besides point-
like (zero-dimensional) particles and one-dimensional strings, there are
two-dimensional membranes. There are also higher dimensional objects
generically called branes.
The AdS/CFT correspondence states that a unied theory of quan-
tumgravity (e.g., a superstring or supergravity theory) dened inananti-
de Sitter space-time of four spatial dimensions is equivalent to another
quantum eld theory living on the three spatial dimensional boundary
of the AdS space. In various braneworlds, the boundaries are depicted
as branes (one of which represents the universe we experience) and the
intervening space of four spatial dimensions is called the bulk. Other
higher dimensions are either compactied or otherwise distracted in
The Road Goes Ever On 145
such a way as to not be directly obvious to us. In some sense, our universe
is merely a shadow being projected by the bulk dimension. Our universe
is the boundary and the bulk is the bounded region. Therefore, studying
physics on the boundary (in our universe) tells us about the physics of
the higher dimension, and vice versa. It is as if we are shadow puppets,
and by studying the details of the shadows we can gure out the way the
hands are arranged in higher dimensions.
One of the most important braneworld models was developed by
Lisa Randall and Raman Sundrum in 1999, and was used in an attempt
to tackle one of the fundamental lingering roadblocks in the path of
unifying the forces, namely why gravity is so much weaker than the
other forces. Most people have the mistaken impression that gravity is
strong, especially when trying to climb several ights of stairs. But the
fact that it is done so routinely clearly shows the weakness of gravity. The
simple action of hanging a note on the refrigerator using a magnet also
demonstrates gravitys inherently wimpy nature. The electromagnetic
force that holds our atoms together is 10
43
times more powerful than
gravity. This relative weakness of gravity is called the hierarchy problem,
and has no obvious solution in the standard model of particle physics.
The key is the way the different forces are described in string theory.
Recall that closed strings represent gravity (through the graviton), while
open strings represent the particles of matter as well as the particles
that mediate the other three forces. The open ends of these strings
attach to the brane that represents our universe and prevents these
particles or their corresponding forces from leaking into the bulk. The
closed strings are not conned to our brane and leak into the bulk,
and can actually travel to any other branes contained in that particular
model. The rate of leaking can be carefully balanced such that it solves
the hierarchy problem yet is not so drastic that current experiments
would notice the difference in gravitys behavior. It is possible that future
particle accelerators withmuchhigher energies may be able to ndsmall
variations caused by gravitys proposed leaky behavior.
Braneworlds have also been used as alternatives to the inationary
cosmology. In2001 Paul Steinhardt, one of the fathers of newination,
and colleagues created what they felt was a direct competitor to the
earlier theory. In their model, our universe is represented as a visible
brane which is located some nite distance in the bulk away from
a hidden brane. The universe was initially empty and cold, which
changed when the two branes collided, bouncing or passing through
each other. The collision converts some of the kinetic energy of the
hidden brane into particles and radiation on our brane, and from there
146 Cosmology 101
the model evolves similarly to the hot big bang. This theory was dubbed
the ekpyrotic model, named after the Greek word for conagration, a nod
to the model of the ancient Greek school of the Stoics who claimed the
universe was born from re. One selling point of the theory is that it
avoids an original singularity, as the laws of physics do not break down
at any point in the process.
A year later, two of the ekpyrotic universes creators, Steinhardt
and Cambridge physicist Neil Turok, expanded the idea to encom-
pass a cyclic model of ery collisions and expansions which includes
a quintessence model of dark energy as a central component. In this
cyclic model, the general homogeneity and isotropy of the universe as
well as the density perturbations are explained as relics of the transition
between cycles. Both the cyclic model and ination predict density per-
turbations that are so similar that current technology cannot distinguish
between the two models. When gravity waves are nally detected, cos-
mologists will be able to compare the relative success of the two models
more carefully, as ination predicts distinctive gravity waves while the
cyclic model predicts no primordial gravity waves.
COSMIC STRINGS
Ination and its ekpyrotic rival are not the only theories proposed to
explain the primordial density perturbations that seeded structure in
the universe. In grand unied theories, topological defects or snags
would form in the fabric of space-time as various symmetries were bro-
ken in an imperfect way. These would be areas of high energy density,
very massive, and come in a variety of dimensions. For example, point-
like defects are called monopoles (such as the magnetic monopoles
previous discussed), one-dimensional defects are called cosmic strings
(to prevent confusion with string theory) and two-dimensional defects
are termed domain walls. In the seminal paper on the subject in 1976,
T.W.B. Kibble demonstrated that the existence of domain walls can be
ruled out because of their gravitational effects. They have an inherent
repulsive gravitational eld and produce large-scale structures differ-
ent from what is observed in the universe today. While theorists Kolb
and Turner called domain walls cosmological bad news, they acknowl-
edged that cosmic strings are much more palatable to a cosmologist
(1990, 220).
Cosmic strings would have less drastic effects on the universe, and it
was suggested that they could act as the density perturbations or seeds
for the formation of galaxies, a problem which was actively being re-
searched in the 1980s. Because of this, a urry of theoretical papers
The Road Goes Ever On 147
soon appeared describing both the properties of cosmic strings and
possible observational tests of their existence. Cosmic strings were the-
orized to come in two varieties: innitely long, open strings, and loops
of closed strings. The latter were envisioned to have been created when
open strings intersected and chunks broke off and closed. Both ver-
sions were pictured as very thin (approximately 10
30
cm) and have an
unbelievably high density10
15
tons per inch! Cosmic strings of both
types were envisioned as possible sources of structure in the universe.
The closed loops were thought to be able to attract matter to them as
they oscillated close to the speed of light, while the open strings were
pictured as moving through the universe, creating a gravitational wake
in the process. Matter would tend to fall into these wakes and form
structures. After creating these structures, closed strings would dissipate
through the emission of gravity waves, and today all that might remain
would be the structures they seeded, as well as relic gravity waves.
With improved observations, the likelihood of cosmic strings even
existing, let alone taking an important role in structure formation, has
diminished signicantly. Cosmic strings make two important predictions
for structure in the universethat it would be strongly skewed to line-
like features, and that the density perturbations would be of a different
form than those predicted by ination. Neither of these predictions ap-
pear to match the universe we inhabit. Additionally, the gravitational
elds of open strings would produce very distinctive patterns of grav-
itational lensing, which have not been observed. Finally, as an open
string moved, it would create distinctive discontinuities in the cosmic
microwave background, called the KaiserStebbins effect. This has also
not been observed. It is therefore clear that if cosmic strings exist (and
there is no denitive evidence in their favor), they are ruled out as a
signicant source of structure formation in the early universe. As with all
the scientic speculations discussed so far, cosmic strings were voted off
the island not as a result of their personality but because they failed to
meet the rigorous tests demanded by the scientic method. Some theo-
ries (such as superstrings) are harder to test, but in the end, all theories
must submit to the demands of the scientic method, even those that
ask perhaps the most tantalizing questions of allwhat is the likelihood
of life elsewhere in the universe?
THE FINE-TUNED UNIVERSE
In 1938, famed physicist Paul Dirac noted a curious relationship be-
tween the numerical size of various parameters in the universe, includ-
ing its then-presumed age, namely that various combinations of these
148 Cosmology 101
constants had values around 10
39
, which became known as the large
number hypothesis. Dirac believed that this could not be a coincidence,
and instead revealed some important relationship between cosmology
and the subatomic realm. Robert Dicke pointed out in 1961 that since
the age of the universe would obviously change over time, the various
other parameters would also have to change to keep their ratio close
to Diracs special value. Since this is not accepted by most physicists, he
found another way to explain the supposed relationship. He pointed
out the fact that we as observers can only exist for a certain period of the
history of the universe. For example, we could not make observations of
the universe before sufcient generations of stars had previously died
and made enough carbon to construct our bodies. We can also only
exist as long as there are stars and planets of just the right properties to
sustain us.
This was the beginning of a realization that the universe seems to be
ne-tuned in such a way as to make carbon-based life-forms possible.
For example, if the strong force were much stronger, all protons would
pair off and there would be no normal hydrogen (and hence no water).
If it were much weaker, large atomic nuclei could not exist. If gravity
had been much weaker, structures such as stars and galaxies would never
have been created, but if it were much stronger, the universe would have
recollapsed before now. If the universe had been created with more than
the three spatial dimensions we currently enjoy, neither atoms nor plan-
etary orbits would be stable, but in fewer than three spatial dimensions
basic biological processes such as circulation of blood and digestion are
not possible. Various explanations for this ne-tuning have been pro-
posed by scientists, philosophers, and religious writers. Some have taken
a nonscientic approach and explained this as evidence of some grand
designer. Withinthe scientic sphere, the anthropic principle and multi-
verse theories are both somewhat controversial alternative explanations
to the religious concept of ne-tuning by design.
The Anthropic Principle
A scientic discussion of the anthropic principle is best done within
the framework of careful denitions, as there are several concepts
housed under the anthropic umbrella. The most common version of
the principle is the so-called weak version attributed to Robert Dicke
that was discussed above. Essentially the cause of the ne-tuning of the
universe is not explained, other than to say that if the constants were
not the values they currently are, we as observers would not be here to
observe them. In some ways it explains why intelligent life exists at this
The Road Goes Ever On 149
point in the universe rather than earlier, in that the conditions (such as
the amount of carbon made within stars) necessary for life simply were
not met until this point. Some have called this weak anthropic principle
(WAP) a tautology with no predictive power.
The strong anthropic principle (SAP) was developed by Brandon Carter
in 1974 and suggests that the universe had no choice in its ne-tuning
because the eventual existence of intelligent observers was a necessity.
Rather than to merely note that the ne-tuning of the universe makes
our existence possible, the SAP seeks to answer why the universe is
ne-tuned by appealing to our very existence. As speculative (and con-
troversial) as the SAP is considered, even more so are the nal anthropic
principle (FAP) and participatory anthropic principle (PAP). The former
claims that not only is the eventual existence of intelligent life a neces-
sity for the universe, but that once it has arisen it will continue to exist
indenitely. The PAP is related to questions of the role of the observer
in quantum mechanics that are beyond the scope of this book. In a
nutshell, it claims that observers are necessary to bring the universe into
full existence. In this view, events can only be said to have reality if they
are observed by a human mind. However, the SAP, FAP, and PAP appear
unlikely, as the only form of intelligent observers currently known to
exist (humans) did not arise until the universe was well over 13 billion
years old and well after all its basic structures (such as stars, galaxies,
and superclusters) had already been formed.
Among the physics community, reaction to the anthropic principle(s)
includes open hostility and resigned acceptance (until theres a better
explanation, it will have to do). In their review of the standard model
of the early universe, Kolb and Turner editorialized It is unclear to one
of the authors how a concept as lame as the anthropic idea was ever
elevated to the status of a principle (1990, 269). String theorist Leonard
Susskind also paints a rather vivid picture: The Anthropic Principle af-
fects most theoretical physicists the same way that a truckload of tourists
in the African bush affects an angry bull elephant (2006, 172). The
anthropic principle even became the subject of parody; for example,
mathematician David Shotwell coined the entomologic principle (the
universe is ne-tuned to produce insects) and Carl Sagan offered a sim-
ilar lithic principle (the universe is ne-tuned to produce rocks). On
the other side of the spectrum, the anthropic principle has been used
for many years by Stephen Hawking and other theoretical physicists to
constrain the parameters of various models by selecting out those ver-
sions which allow for the existence of observers. For example, if a model
allows a variety of types of universes, the anthropic principle can be used
150 Cosmology 101
to only accept those types in which life is possible (because any types of
universes that do not allow life are obviously not relevant to a discussion
of our particular universe).
Multiverses
The other major scientic explanation for the ne-tuning of the uni-
verse is the multiple universe or multiverse paradigm. This is a set of
theories which derives from the inationary model of cosmology, based
on the prediction that ination is eternal. Separate inationary regimes
or pocket universes will continuously be produced as they nucleate out
from the original false vacuum state. The collection of all these pocket
universes (of which our observed universe is just one) is referred to as
a multiverse. A possible explanation for the observed ne-tuning of con-
stants in our pocket universe can be made on statistical grounds. Given
a large (possibly innite) sample of pocket universes, there should be
a large range of possible combinations manifested for the fundamental
constants. In at least one of those pocket universes the combination of
constants is appropriate for the formation of intelligent carbon-based
life, and this is the pocket universe in which we live. Other pocket uni-
verses could have conditions completely averse to the origin of life.
There is no intention, no design, in some sense there is just dumb luck.
Gordon Kane compares it to winning the lottery: someone had to win,
and no one selected who that was, except randomly. Just because a uni-
verse has a unique set of laws and parameters should not lead one to
wonder whether that set was designed (2002, 24). Again, the argument
is that scientically there is no need for a designer. Any theological dis-
cussions are separate and distinct, and are not ruled out by any of the
scientic arguments presented here.
It has been suggested that the anthropic principle could be combined
with the multiverse model, to provide an explanation for the tiny size
of the cosmological constant. If it is an environmental variable and
varies from one pocket universe to another, we can use the anthropic
principle to state that we could only exist in (and hence observe) a
pocket universe in which the value of the cosmological constant is small
enough to allow for the formation of atoms, stars, and galaxies. In the
words of cosmologist Michael Turner,
If the multiverse is right, it could be that we live in a little oasis. . . . Martin
Rees has speculated that we could well be the only intelligent beings in the
observable universe . . . Wed better be really careful not to extinguish it.
(Yulsman 2003, 346)
The Road Goes Ever On 151
THE SEARCH FOR EXTRATERRESTRIAL INTELLIGENCE
Are we truly the only intelligent species in the universe capable of cos-
mological queries? In 1600, controversial Italian philosopher Giordano
Bruno was burned at the stake by the Inquisition. Among his crimes
was proposing that There are then innumerable suns, and an innite
number of earths revolve around those suns, and that on those Earths
there could exist life (Munitz 1957, 183). In 1995, Swiss astronomers
Michel Mayor and Didier Queloz discovered a strange planet orbiting
the star 51 Pegasi with a period of 4.2 days. This was the rst extraso-
lar planet, or exoplanet, discovered by the telltale gravitational wobble
it inicts as it orbits its parent star. The wobble manifests itself in a
tiny, cyclical alternating blueshift and redshift of the parent star. Over
200 exoplanets have been discovered since this time, the vast majority
by this Doppler oscillation method. It is sensitive to planets of large
masses and small orbits, as the size of the spectral Doppler shift de-
pends on the strength of the gravitational tug the planet inicts on its
star.
To nd planets closer to Earths mass, two other methods have
employedmicrolensing and transits. Microlensing can be used when
a small object passes in front of another small object and the image of
the background object is temporarily brightened through gravitational
lensing. In the case of exoplanet searches, microlensing is well suited
to nding planets of small mass, like the earth, and with larger orbits
than those found through the Doppler shift method. In transit events,
the planet crosses the face of the star, resulting in a characteristic dim-
ming of the star. At least ten stars are known to have planets that can
be monitored in this way. With the explosion of known exoplanets in
the past decade, it is natural to speculate whether any of them (or the
host of planets undoubtedly still undiscovered in our galactic neck of
the woods) are currently inhabited.
The possibility of life elsewhere in the cosmos has fascinated scien-
tists and laypersons alike for centuries. There is currently no scientic
evidence that there is life elsewhere in the universe, despite uncor-
roborated reports of abductions by little green men and sightings
of weather balloons, ocks of birds, and even the planet Venus in-
correctly attributed to alien spacecraft. The scientic study of the
possible properties of lifeforms that might exist outside of the earth
is termed astrobiology, while the search for intelligent lifeforms out-
side of the earth is called SETIthe search for extraterrestrial intelli-
gence.
152 Cosmology 101
Life as We Know It (LAWKI)
Astrobiology is based on our current understanding of life on Earth,
or as it is sometimes termed, LAWKIlife as we know it. It is clear
that our knowledge is therefore limited, and there is no reason why
life on other worlds must be identical or even similar to life on our
world. On Earth life is based on four basic elements, carbon, hydrogen,
oxygen and nitrogen, which are relatively commoninthe universe today.
Carbon is a remarkable atom, which can form unique long chains and
rings which serve as the backbone for organic moleculesthe basic
building blocks of life. In science ction stories, silicon is sometimes
used as the basic component for alien lifeforms, since it lies just below
carbon on the periodic table and therefore has some similar properties.
Unfortunately, silicon is not believed to be able to easily make complex
molecules similar to those that are vital to carbon-based life.
It is necessary to dene what we mean by life in the rst place. Perhaps
the most simplistic denition is that life has the ability to reproduce at
the cellular level and evolve. Life on Earth reproduces via DNA(deoxyri-
bonucleic acid), the famous double helix molecule. Although there
is no reason that life on other worlds should utilize DNA, it is highly
probable that some similar type of molecule should be involved. DNA is
composed of four different base amino acids, and these and other amino
acids make up proteins, important compounds in the function of living
organisms. All four of these base amino acids, plus 70 more (including
55 that do not occur naturally in terrestrial life) have been identied
in the Murchison meteorite, which landed in Australia in 1972. Amino
acids have also been found in the Murray meteorite, as well as in the
interstellar medium, leading to the conclusion that the basic building
blocks of life are prevalent in the universe today. But amino acids are
only the building blocks of life, and the steps between the formation of
these chemicals and creation of the rst reproducing lifeforms are not
denitively understood. This lesson was clearly demonstrated in front of
a worldwide audience in the case of the supposed Martian microfossils.
In August, 1996, NASA held a special press conference to announce
that signs of ancient life had been found in a meteorite that had been
blasted off of Mars. The 3.9-billion-year-old rock, named ALH84001,
had been collected in Antarctica in 1984, and like over a dozen other
meteorites, had been veried as Martian in origin by the composition of
gases trapped within the rock (which matched observations taken from
the surface of Mars by the Viking landers). Inside the greenish rock
scientists had found what they believed to be four different signatures
The Road Goes Ever On 153
of life. The rst was rosette-shaped nodules of carbonate about 0.05
mm across that were suggested to have been produced by bacteria. The
second evidence was tiny grains of magnetite about 50 nm long (less
than 1/100th the width of a human hair). Magnetite is a magnetic
mineral used by some bacteria on Earth as internal compasses and also
produced as a waste product to rid the bacteria of excess iron. The third
evidence presented was complex organic molecules called polycyclic
aromatic hydrocarbons, or PAHs, which could be a sign of life. The most
controversial evidence was tiny tube-shaped objects that some scientists
interpreted as microfossils of Martian bacteria.
As soon as the results were announced, scientists from around the
world began the process of peer review, rechecking the analysis of the
NASA team and searching for alternative explanations and interpreta-
tions for the four pieces of evidence. Within several years, the claim of
life on Mars had begun to unravel. Logical nonbiological explanations
for the carbonate rosettes and magnetite grains were soon published.
Although the PAHs were found to be likely Martian in origin and not
contamination from the meteorites time on Earth, the presence of or-
ganic molecules does not prove the existence of life (as in the case
of other meteorites found on Earth). The supposed microfossils were
carefully scrutinized by biologists, with the result that they have been
largely refuted as evidence of life (at least as we understand it). The
objects were only a few dozen nanometers across, and research strongly
suggests that a size at least 10 times that would be necessary to success-
fully house the most basic molecular machinery of life (Kerr 1998,
1398). In comparison, the common E.coli bacteria has a diameter of
about 2,000 nm. Therefore, at the time these words are being typed,
there is no evidence for past or present life on Mars, although either
possibility certainly cannot be ruled out by our limited exploration of
the Red Planet.
Since life has not been found on the planet next door, where should
we begin our search for the proverbial needle in a haystack? Although
humans can only exist within a relatively narrow range of temperatures,
and any aquariumowner knows that sh can only thrive within a window
of pH levels, there are organisms called extremophiles that can exist
under conditions once deemed impossible. For example, thiococcus,
a type of bacteria, can actually perform photosynthesis using infrared
rather than visible light. For the sake of conversation, we will restrict our
search to planets similar to Earth in terms of temperature, composition,
and the amount of harmful radiation received from their stars. We will
also assume that since it took approximately a billion years for the rst
154 Cosmology 101
life to arise on Earth, and another 3.5 billion years for intelligent life to
evolve, the timescale will be similar on other worlds. Therefore we will
only look at stars whose lifespan is long enough to allow intelligent life
to evolve. Having done this, we are left with spectral class F5 through K5
stars, limiting our search to about 25 percent of the stars in our galaxy.
Now that we have narrowed our search to a mere 25 billion stars, is there
any way to estimate how many intelligent civilizations await contact?
In 1961, radio astronomer Frank Drake came up with a mathematical
expression for this estimate, now called the Drake equation. Although it
is cited in slightly different forms in different references the concepts
are rather basic. There are a number of parameters which must be
considered in estimating the total number of intelligent civilizations
in our galaxy, and since most of these parameters are not known with
certainty, the number of intelligent civilizations can vary from as low
as one to as high as several million or more. The various parameters
are the number of stars in the Milky Way, the fraction of stars similar
to the sun, the fraction of sun-like stars with planets and the number
of habitable planets per star, the fraction of such planets where life
actually arises, the fraction of those that are inhabited by intelligent
life that communicates via modern technology, and the average lifespan
of such intelligent civilizations. It is actually this last parameter which
has the most variation, since we only have one civilization to use as a
modelour own.
Talking to Aliens
Given the uncertainty in the number of candidate civilizations in the
Milky Way, and no way to know in which direction to look, astronomers
are left with a quandarydo we sit and listen, hoping that our galactic
neighbors are talkative, or do we broadcast, like the Whos in Dr. Seuss
classic tale Horton Hears a Who, our faint voices shouting to the cosmos
Were here! Were here! Both tactics have been taken by astronomers
over the past few decades. In 1974, Frank Drake and colleagues beamed
a message toward the globular cluster M13 in Hercules. It was a digital
message which contained basic pictures of a human being, the solar
system, and the double helix of DNA.
Most scientists involved in SETI are using a sit and wait tactic, hop-
ing to pick up evidence of an intelligent signal. But even this is a mon-
umental task, given the number of possible wavelengths that could be
used. Since water is a vital component of life on Earth, one can as-
sume it might also be important for extraterrestrial life forms. Water
The Road Goes Ever On 155
can be broken into atomic oxygen (O) and the hydroxide radical (OH),
which are called the dissociation products of water. Each has distinctive
spectral lines, the most important of which lie between frequencies of
1400 and 1700 MHz. In 1976 physicist Bernard Oliver suggested that
searches for extraterrestrial intelligence should focus on this range of
frequencies, which was dubbed the water hole. It was argued that in-
telligent species would meet around this technological water hole just as
creatures habitually do on Earth around more literal water holes. Carl
Sagan suggested that since the 21-cm radiation of cool hydrogen should
be known to any technological society, multiples of that frequency (1420
MHz) might be a good choice for a broadcast, especially frequencies
such as 1420 or 1420/. This is the assumption he used in his novel
Contact.
Besides radio signals, humans have also sent physical artifacts into
space that can tell extraterrestrial travelers about our planet. For exam-
ple, the Pioneer 10 and 11 spacecraft that traveled by Jupiter in the 1970s
were slingshotted out of the solar system by the giant planets gravity.
Although they are not slated to pass by any known stars in the foresee-
able future, each has a gold-plated plaque designed by Carl Sagan and
Frank Drake which tells the simple story of the species that sent them.
Next to a diagram of the spacecraft (shown for scale) are the naked
forms of a male and female human, the males hand raised in a gesture
of friendship. A diagram of the solar system with Earth highlighted is
also included, along with several other pictograms. Several years later
the two Voyager spacecraft were launched on a tour of the giant planets
and beyond. Each contained a gold-plated record and materials (and
instructions) to build a record player. As Carl Sagan explained, each
contained
Greetings in sixty human tongues, as well as the hellos of the humpback
whales. We sent photographs of humans from all over the world caring for
one another, learning, fabricating tools and art and responding to chal-
lenges. There is an hour and a half of exquisite music from many cultures,
some of it expressing our sense of cosmic loneliness, our wish to end our
isolation, our longing to make contact with other beings in the Cosmos.
(1980, 287)
Is this search in vain? Is there anyone out there? If so, could they
really understand the messages we have sent? Would they care enough to
answer us? Only one thing is certainnothing is truly in vain in science
156 Cosmology 101
if it allows us to improve our technology and expand our knowledge
of the universe, even if that knowledge leads us to the conclusion that
we are alone. In the words of Carl Sagan, Many, perhaps most, of our
messages will be indecipherable. But we have sent them because it is
important to try (Ibid.).
Conclusion
We have now reached the end of our journey through space and time,
from the origin of the universe to its demise, from our solar system to
the faintest and most distant galaxies and quasars. It is certainly impres-
sive how much we know about the universe we live in, given our xed
position in space and time within it. We began with the concept that the
earth was the center of the cosmos, and have been left with the idea that
we are not only denied a central position, but we are not even made
of the most prevalent materials in the universe (namely dark matter
and dark energy). If eternal ination is correct, our visible universe is
also one of perhaps an innite number in a larger multiverse, further
removing us from a place of importance. This concept was called the
extreme Copernican Principle by cosmologist David Schramm.
Despite all that we know about the universe, there remain serious
unanswered questions, including the nature of dark energy, the prop-
erties of the correct grand unied theory and quantum gravity model,
and the identity of the inaton eld. Once gravity waves are directly de-
tected, inationary theories may be further bolstered and the ekpyrotic
model refuted, or perhaps vice versa. The scientic method is a cruel
mistress, and no matter the beauty of the theory, it can fall victim to the
blades of observation and experimentation in the ash of an eye.
Never was the scientic method more evident to the general public
than in the case of Pluto. When I began writing this book, Pluto was
classied as a planet. By the time I had nished, the International As-
tronomical Union had reclassied it as a dwarf planet, and the news was
awash with debate over the decision, much of it emotionally charged.
The response of many people to the news of Plutos demotion from
planetary status points out a drastic misunderstanding in todays culture
of the nature of science. As we have seen, science is the investigation
158 Cosmology 101
of the natural world, a never-ending search for answers to current ques-
tions about how the universe works, as well as the search for new and
better questions to ask. Sometimes discoveries are made and not com-
pletely understood at the time, leading to erroneous assumptions being
made. Unfortunately, as we have seen, even in science misconceptions
can be stubborn and are not always swift to be remedied. Such is the
case with Pluto.
Decades after its discovery, astronomers had learned more about
Plutos properties, and it was obvious that it was not only very different
from the four Earth-like rock and metal planets and the four Jupiter-like
gas giant planets, but that it shared more in common with the myriad
icy bodies in the outer solar system. The suggestion was originally made
by those in the International Astronomical Union (IAU) that Pluto be
cross-classied as a planet and Kuiper Belt object, a member of the icy
asteroid belt of sorts beyond the orbit of Neptune. The outcry from
the general public was surprising, and rather alarming. As you have
learned, scientists admit their mistakes all the time, reclassifying objects,
acknowledging that experiments and observations were erroneous, even
acknowledging the occasional embarrassing error in a theoretical calcu-
lation. All of this passes without a blink from John Q. Public. Where was
the outcry several years ago when the IAU extended the classication
of stars to include the new classes L and T? Astronomy books had to
be changed and students scrambled to create a new mnemonic device
for the extended classication system (if anyone has a really good one
for OBAFGKMLT, Im all ears). When rabbits were reclassied from
rodents to lagomorphs in the early 1900s, the Easter bunny did not
suddenly change his trademark confectionary gifts.
Why is Pluto so different? Plutos proposed reclassication brought
claims that textbooks would need to be changed, and that children
would have to learn new facts. Yes, they will, and they have, throughout
the history of science! Astronomical discoveries happen at such a fast
and furious rate these days that textbooks are out of date before their ink
is even dry. The number of crossed-out sentences and hastily scribbled
additions to the margins of my lecture notes drives home that point.
When classes start every semester, I preface my comments on the rst day
of class with the warning that 10 percent of what I will teach the students
may eventually be proven wrong (a rather conservative estimate). As
members of a technological society, we have had to learn to deal with
constant change, which is admittedly not always comfortable. I still have
a box of eight-track tapes in my basement, for reasons that have nothing
to do with science. Perhaps Plutos demotion hits some people harder
Conclusion 159
than others for a similar reason. But as Max Tegmark hopefully offers,
Perhaps we will gradually get more usedto the weirdways of our cosmos,
and even nd its strangeness to be part of its charm (2004, 490).
It is this very same ever-changing nature of scientic knowledge which
assures that there will always be questions to ask and experiments to
run. There is always that one further test which might prove a long-
cherished theory wrong, or point the way to an improvement in our
current understanding (as in the case of ination and the big bang
theory). Science does not belong to the scientists, and the universe
does not belong to cosmologists. We are all citizens of this grand and
innite cosmos, our bodies made of the very atoms createdinthe nuclear
furnaces of long-dead stars, our minds free to soar through space and
time to the very beginning of the universe. Take the time to go out on
the next clear night, away from the distracting lights of modern society,
and take a moment to gaze upward at our stellar cousins, to whom
we owe our very existence. In the words of Annie Jump Cannon, who
became intimately familiar with more than 350,000 stars during her
distinguished career, starlovers and stargazers are linked together in
bonds of friendship while the same sky overarches us all, and we do our
best to increase the sum of human knowledge as pertains to the Story of
Starlight (1941, 61).
Glossary
Absolute magnitude. How bright a star would appear at a distance of
10 pc.
Accretion disk. A disk of hot material spiraling onto a stellar corpse, such
as a white dwarf.
Active galactic nuclei. A unied explanation for QSOs, Seyfert galaxies,
radio galaxies, and blazars based onthe interactions betweena supermassive
black hole and its local environment.
AdS/CFT. The concept that a quantum theory in a higher-dimensional
space can be modeled by another theory that inhabits the boundary of that
space.
Angular power spectrum. A graph of the cosmic microwave background
showing the relative strength of temperature variations of various angular
sizes.
Apparent magnitude. The observedbrightness of a star as seenfromEarth.
Astronomical unit. The average sun-Earth distance (150 million km).
Asymptotic branch. The low temperature, high luminosity section of the
HR diagram that dying stars of solar-type mass inhabit after ceasing the
triple-alpha cycle.
Barred spiral. A spiral galaxy whose center features a bar-like structure.
Baryogenesis. The physical processes in the early universe that selected
out matter over antimatter.
162 Glossary
Baryonic matter. Normal matter composed of protons and neutrons.
Big bang. The scientic theory that the universe began in a hot, dense
state and has expanded ever since.
BL Lac object. A class of powerful and highly variable nuclei of galaxies.
Black hole. An object whose gravitational eld is so intense that not even
light can escape.
Blackbody. A hypothetical body that is a perfect absorber and perfect
emitter.
Blazar. A term for BL Lac objects and optically violent variable quasars.
Boson. A subatomic particle that does not obey the Pauli exclusion prin-
ciple, such as the photon.
Brane. An object with two or more spatial dimensions predicted by
M-theory.
Brown dwarf. An object more massive than a planet but whose mass was
too low to become a star.
Carbon cycle. The chain of hydrogen fusion reactions used by massive
main sequence stars to generate energy, where carbon acts as a catalyst.
Cepheids. Unstable stars whose period of variation depends on their
average luminosity.
Closed universe. A universe that will collapse in the future because the
density of matter and energy is above the critical density.
Cold dark matter. Dark matter that moves much slower than the speed of
light and clumps together to seed galaxy formation.
Cosmic distance ladder. A series of corroborating methods used to esti-
mate distances to stars and galaxies.
Cosmic magnication. A weak variety of gravitational lensing that bright-
ens the images of distant galaxies.
Cosmic microwave background. Apervasive leftover energy fromthe early
universe which has signicantly cooled over time to the current 2.7 K.
Glossary 163
Cosmic shear. A weak variety of gravitational lensing that distorts the
shapes of distant galaxies.
Cosmic string. A one-dimensional defect in space-time predicted by some
GUTS.
Cosmological constant. A constant term in Einsteins eld equations of
general relativity, originally conceived by Einstein to balance gravity and
keep the universe static, now used to describe a leading candidate for dark
energy.
Cosmological principle. On average, the universe looks the same viewed
from any location and in any direction.
Cosmology. The scientic study of the composition, structure, and evolu-
tion of the universe.
Critical density. The density of matter and energy in the universe required
for the expansion of the universe to be exactly balanced by gravity.
Dark age. The period of time between recombination and the rst gener-
ation of stars when the universe was dark.
Dark energy. A generic name for the unknown cause of the current accel-
erating expansion of the universe.
Dark matter. Invisible material whose presence is discovered through its
gravitational effect.
Deceleration parameter. Ameasure of the rate of change of the expansion
of the universe.
Deferent. In the Ptolemaic cosmology, the main circle comprising a
planets orbit around Earth.
Degeneracy pressure. The pressure created by the Pauli exclusion princi-
ple that prevents further collapse of a stellar corpse.
Density perturbation. Slight variations in the density of matter that were
created in the early universe and led to the formation of structures such as
galaxies.
164 Glossary
Density wave theory. Spiral arms form when gas, dust, and stars slow their
revolution around the galaxy and temporarily clump together like a trafc
jam.
Distance modulus formula. The distance to an object is determined by
comparing its apparent and true brightness.
Doppler broadening. The widening of a spectral line due to the rotation
of an object along our line of sight.
Doppler effect. Wavelengths of sound or light lengthen as the source
moves away from the observer and shorten as the source approaches.
Drake equation. A mathematical estimate of the number of technological
societies in the universe
Duality. A connection between two physical theories that demonstrates
that they give the same results.
Einstein eld equations. The fundamental equations of general relativity
that describe how matter and energy warp space-time.
Ekpyrotic model. An alternative to ination in which the universe is cre-
ated by the collision of two branes.
Electromagnetic spectrum. The continuum of all possible wavelengths of
light (electromagnetic waves).
Electroweak force. The high-energy unication of the electromagnetic
and weak nuclear forces.
Elliptical galaxy. A galaxy with a round to elliptical shape and no spiral
arms.
Epicycle. In the Ptolemaic cosmology, the small circle on which a planet
is attached, which in turn is attached to the deferent.
Equant. In the Ptolemaic cosmology, the center of uniform motion of a
planets orbit.
Eternal ination. The concept that once inationbegins it will never cease.
Extinction. Interstellar dust makes distant objects appear articially dim.
Glossary 165
False vacuum. A local minimum value of energy that is not the overall
lowest possible energy state.
Fermion. A subatomic particle that obeys the Pauli exclusion principle,
such as the electron.
Field. A physical quantity whose properties can be dened at a particular
location.
Final anthropic principle. The suggestion that once life arises in the uni-
verse it can never be extinguished.
Flat universe. A universe that has achieved the critical density.
Friedmann-Robertson-Walker model. A mathematical model of the uni-
verse that can be open, closed, or at (in a geometrical sense).
Geocentric universe. An ancient model of the universe with Earth at the
center.
Globular cluster. A dense, spherical mass of hundreds of thousands of
older stars.
Grand unied theories (GUT). A class of models that seek to unite the
strong nuclear and electroweak forces.
Gravitational lensing. Multiple images of a distant object (such as a QSO)
caused by the gravitational eld of an intervening object (such as a galaxy).
Gravity waves. Ripples in space-time generated by the motion of massive
objects.
Hadron. A particle that interacts via the strong nuclear force.
Half-life. The time required for half of a radioactive sample to decay.
Heisenberg uncertainty principle. In quantum mechanical systems, it is
impossible to know both the location and motion of a particle, or the
duration and energy of an event, simultaneously to absolute certainty.
Heliocentric universe. A model of the universe that places the sun at the
center.
166 Glossary
Hertzsprung-Russell (HR) diagram. A plot of the spectral class (tempera-
ture) and luminosity (absolute magnitude) of stars.
HI region. A cool hydrogen cloud with a distinctive radio emission at
21 cm.
Hierarchy problem. The puzzling fact that gravity is so much weaker than
the other forces.
Higgs eld. A quantum mechanical eld supposed to make subatomic
particles massive rather than massless.
HII region. A hot cloud of glowing hydrogen.
Horizontal branch. The section of the HR diagram between the red giant
branch and main sequence where stars undergoing core helium burning
are located.
Hot dark matter. Dark matter that travels near the speed of light and can
only form very large structures.
Hubble tuning fork diagram. A pictorial representation of the various
classes and subclasses of galaxies.
Hubbles law. The more distant a galaxy, the faster it appears to recede,
evidence for the expansion of the universe.
Hypothesis. Aproposed scientic explanation for some aspect of the phys-
ical world.
Ination. A class of models in which our observable universe underwent
an early period of exponential expansion.
Inationary CDM model. Also called the concordance model, a com-
bined model of cold dark matter, dark energy, and ination that is sup-
ported by many observations.
Interstellar reddening. Interstellar dust scatters the blue component of
starlight, making the star appear less blue (or redder).
Island universe hypothesis. The suggestion in the late 1700s that there
might be other galaxies outside the Milky Way.
Glossary 167
Keplerian rotation curve. The rapidly declining rotation curve generated
by a system where the mass is concentrated in the center (like the solar
system).
Lenticular galaxy. A lens-shaped galaxy resembling a spiral without arms.
Local Group. The several dozen galaxies clustered around the Milky Way
and Andromeda Galaxy.
Magnetic monopole. A hypothetical particle that represents an isolated
north or south magnetic pole.
Main sequence. The diagonal line on the HR diagram that represents
normal stars creating energy by fusion of hydrogen in their core.
Main sequence turnoff. The position of the top of the main sequence on
an HR diagram for a star cluster, determined by the age of the cluster.
Malmquist bias. Dim members of any population will be overlooked in
surveys of distant objects.
Metallicity. The percentage of a star (by mass) composed of elements
heavier than helium.
Microlensing. A point-like object passes in front of another point-like ob-
ject (such as stars), focusing the light and temporarily brightening the
system.
M-theory. A speculative theory invoking extra dimensions that includes
superstrings and supergravity as special cases.
Multiverse. A collection of pocket universes (or other multiple universes).
Nebula. A cloud of gas and/or dust in space.
Neutron star. A high density, medium-mass stellar corpse made of
neutrons.
Non-Hubble ow. Also called peculiar motion, motion of galaxies inde-
pendent of the expansion of the universe.
Nova. Originally a generic term for any exploded star, more correctly the
outburst of a white dwarf in a binary star system.
168 Glossary
Nuclear cosmochronometry. The use of the decay of radioactive elements
in old stars to estimate their age.
OB association. A loose aggregation of very hot, young stars.
Olbers paradox. The principle that if the universe were eternal and in-
nite the night sky could not be dark.
Open cluster. An irregular group of dozens to hundreds of young stars.
Open universe. A universe whose density is below the critical density that
will expand forever.
Pair-instability supernova. The implosive death of the supermassive rst
generation of stars.
Parallax. The apparent shift in the positions of nearby stars relative to
more distant stars due to the motion of Earth around the sun.
Parsec. The distance of a star whose parallax would be one arc second
(3.26 light years).
Participatory anthropic principle. The idea that the events in the universe
are not actualized until observed, necessitating the existence of intelligent
life.
Pauli exclusion principle. Identical fermions cannot occupy the same
quantum state at the same time.
Perfect cosmological principle. The universe has always looked the same,
everywhere and at all times.
Period-luminosity relationship. The relationship between average lumi-
nosity and period of pulsation for Cepheid variables, used to determine
distances to galaxies.
Planck curve. A graph showing the intensities of a range of wavelengths
of light for an object of a particular temperature.
Planck era. The rst 10
43
seconds of the universe, when all four forces
were unied.
Planetary nebula. The outer layers of a dying star puffed off into space.
Glossary 169
Pocket universe. Separate universes independently created from the false
vacuum.
Polarization. A property of light that describes whether the directions of
oscillation of different waves are aligned or not.
Population I. Stars like the sun that are composed of recycled interstellar
material with higher metallicities.
Population II. Older stars that represent a previous generation and are
relatively lower in elements heavier than helium.
Population III. The still-unobserved rst generation of stars predicted to
be made of pure hydrogen and helium.
Primordial nucleosynthesis. The processes that generated helium in the
rst few minutes of the universe.
Protonproton cycle. The chain of hydrogen fusion reactions used by
solar-type main sequence stars to generate energy.
Protostar. An object in the process of becoming a star.
Pulsar. Arapidly rotating neutronstar detectedby its beams of radio waves.
Quantum chromodynamics. The quantum theory of strong nuclear force
interactions.
Quantum electrodynamics. The quantum theory of electromagnetic in-
teractions.
Quantum gravity. Any scientic model that attempts to unite general rel-
ativity and quantum mechanics.
Quantum mechanics. The principles that explain the behavior of micro-
scopic particles such as atoms.
Quantum tunneling. The ability of a quantum mechanical system to move
fromone side of anenergy barrier to another without having enoughenergy
to pass over the barrier.
Quark. A fundamental particle that comprises neutrons and protons.
170 Glossary
Quasi-stellar object (QSO). A generic term for star-like, highly energetic
galactic nuclei of high redshift, including quasars.
Quintessence. A type of dark energy that varies with time.
Radial velocity. Motion of an object toward or away from an observer.
Radio galaxy. A galaxy with unusually strong radio emissions.
Radiometric dating. Utilizing radioactive decay to estimate the age of an
object.
Recombination. When the universe cooled to a few thousand degrees K,
electrons were nally bound to nuclei to form stable atoms.
Red giant branch. The low temperature, high luminosity section of the
HR diagram inhabited by stars undergoing hydrogen shell burning.
Redshift. The wavelength of light emitted by a receding object appears
lengthened.
Rotation curve. A graph that plots the orbital speeds of objects located at
different distances from the center of a galaxy.
r-process. The rapid neutron capture process by which supernovae cre-
ate nuclei heavier than iron.
RR Lyrae stars. Unstable, low mass stars that pulsate within a day or less.
Scientic method. A continual process of hypothesis, observation, and
experimentation used by scientists to understand the world.
Seyfert galaxy. A spiral galaxy with an unusually bright core and other
unusual emissions.
Singularity. Ingeneral relativity, a situationwhere the eldequations break
down, such as at the center of a black hole or the origin of the universe.
Space-time. The fabric of the universe composed of three dimensions of
space interwoven with one dimension of time.
Spectral classes. The classication of stars by differences in their spectra,
due to different surface temperatures.
Glossary 171
Spectroscopic parallax. An HR diagram of stars of known luminosity and
spectral class can be used to estimate the luminosity, and fromthat distance,
for other stars.
Spectroscopy. The process of separating a beamof light intoits constituent
wavelengths for analyzing.
Spiral galaxy. A galaxy that has two or more arms composed of gas, dust,
and stars.
Spiral nebula. The term used for spiral galaxies before the 1940s.
Spontaneous symmetry breaking. A system automatically moves from a
state of perfect symmetry to a state of broken symmetry.
s-process. The slow neutron capture process used by red giants to create
nuclei heavier than iron.
Standard candle. Any astronomical object whose true brightness can be
estimated with some certainty.
Standard model. The current description of particle physics interactions,
including QCD and the electroweak theory.
Steady-state theory. A out-dated model of the universe as an eternal, un-
changing system.
Stefan-Boltzmann law. The mathematical relationship relating the lumi-
nosity of a star to its size and temperature.
Stellar nucleosynthesis. The processes by which stars create atomic nuclei
through fusion.
String theory. The concept that what are observed as elementary particles
are different modes of vibration of tiny string-like objects.
Strong anthropic principle. The suggestion that the universe must be ne-
tuned to allow for the existence of life.
Strong nuclear force. The strongest of the four fundamental forces, which
describes how quarks interact to form protons and neutrons.
Supercooling. A system cools to a critical temperature and does not auto-
matically undergo the appropriate transition, such as in freezing rain.
172 Glossary
Supergravity. A theory that unites supersymmetry and general relativity.
Supermassive black hole. A black hole millions of times more massive
than the sun, found at the heart of a galaxy.
Superstring. A string theory that unies bosons and fermions.
Supersymmetry. A generic term for models that unies bosons and
fermions.
Synchrotron radiation. A characteristic radio signal created by electrons
moving in a magnetic eld.
Theory. A hypothesis that has been rigorously tested and is generally con-
sidered among the best current explanations.
Triple-alpha cycle. The fusion cycle by which carbon and oxygen are cre-
ated from helium nuclei.
Type Ia supernova. The explosion of a white dwarf.
Type II supernova. The explosion of a red supergiant.
Weak anthropic principle. The concept that if the universe were different,
we would not be here to study it.
Weak nuclear force. The fundamental force that mediates radioactive
decay.
Weakly interacting massive particle (WIMP). Theorized particles of cold
dark matter, including supersymmetric partners and axions.
White dwarf. A compact stellar corpse about the mass of the sun and the
diameter of Earth.
Wiens law. The hotter the star, the shorter the wavelength at which it
emits most of its light, making it appear bluer.
Zone of avoidance. Dense dust and gas in the plane of the Milky Way
impede observations in that portion of the sky.
Selected Bibliography
Adams, Fred C. and Gregory Laughlin. 1999. The Five Ages of the Universe. New York:
Free Press. This is the seminal popular-level work on the fate of the universe,
written by the researchers themselves. A thought-provoking (and sometimes
depressing) survey of the ultimate fate of matter itself, long after the demise of
the solar system.
Alpher, Ralph A. and Robert Herman. 2001. Genesis of the Big Bang. Oxford: Oxford
University Press. Two of the main characters in the development of the big bang
theory recount their roles and the work of famed physicist George Gamow. The
personal nature of this text is a unique window into the world of science and
scientists.
arXiv preprint archive (http://arXiv.org). Given that it can take a year for scientic
papers to be published in professional journals, scientists have taken to posting
preprints of their papers inonline archives. arXiv is the largest and best known
such archive in the physics and astronomy communities. The interested reader
can search for papers by subject, and will nd the most up-to-date research on
a variety of cutting-edge topics.
Asimov, Isaac. 1956. The Last Question. Available at http://www.multivax.com/last
question.html (accessed September 1, 2006). This classic short story may seem
a little dated, but it is an excellent overview of the concept of the fate of the
universe, written by a master of science ction.
Baade, Walter. 1956. The Period-luminosity Relation of the Cepheids. Publications of
the Astronomical Society of the Pacic 68: 516. A transcript of a lecture given in
honor of his receiving the Bruce Medal from the ASP, this article is an excellent
and readable summary of the history of the calibration of the Cepheid period-
luminosity relationship.
Barrow, John D. and Frank J. Tipler. 1988. The Anthropic Principle. Oxford: Oxford
University Press. This technical book is considered the seminal work on the
anthropic principle. The interested and dedicated reader will nd this work
challenging but fascinating.
Bromm, Volker. 2003. Cosmic Renaissance. Mercury 32(5): 2533. Bromm, one of the
active researchers in the eld of population III stars, presents a highly readable
survey of current theories of the properties of the rst generation of stars.
174 Selected Bibliography
Burbidge, E. Margaret, G. R. Burbidge, William A. Fowler, and F. Hoyle. 1957.
Synthesis of the Elements in Stars. Reviews of Modern Physics 29(4): 547650. The
so-called B
2
FH paper demonstrated that all elements heavier than hydrogen
could be created through the lives and death of stars, results that have largely
remained the same to this day. One of the most important papers in twentieth-
century physics, this lengthy and detailed work is largely meant for a technical
audience.
Cannon, Annie J. 1941. The Story of Starlight. The Telescope 8(3): 5661. This article
is the transcript of a popular-level radio talk by the woman who single-handedly
classied more stellar spectra than any other human being. Her overview of the
importance of spectroscopy is written in a comfortable, personal style, and her
genuine affection for her friends, the stars, shines throughout.
Crowe, Michael J. 1994. Modern Theories of the Universe from Herschel to Hubble. New
York: Dover. Crowe surveys the development of modern cosmology through
the multiple lenses of science, history, and philosophy. This text is a valuable
balance of excerpts from original texts and insightful commentary.
Davies, P. C. W. and J. Brown, eds. 1988. Superstrings: A Theory of Everything? Cam-
bridge: Cambridge University Press. Althoughsomewhat dated, this is a valuable
review of the early history of string theory. Much of the work is a series of in-
terviews with the important players in string theory in the 1980s, who provide
their own personal insight into the controversial theory.
Einstein, Albert. 1961. Relativity: The Special and General Theory, 2nd ed. New York:
Crown. Although intended as an overview of both the special and general
theories of relativity for the general reader, it is not an easy read. It is valuable
as an explanation of Einsteins seminal work through his own words, including
unique insights into his thinking.
Flamsteed, Sam. March 1995. Crisis in the Cosmos. Discover, 6677. Despite its sensa-
tional title, this popular-level article covers Wendy Freedmans results concern-
ing the Hubble constant and other unsettling observations clearly and with
brutal honesty. It paints a vivid example of the scientic method in action in
cosmology.
Freedman, Wendy L., Barry L. Madore, Jeremy R. Mould, et al. 1994. Distance to
the Virgo Cluster Galaxy M100 from Hubble Space Telescope Observations
of Cepheids. Nature 371: 757762. This classic paper afrmed that the value
of the Hubble constant could not be accommodated with the accepted ages
of globular clusters in a at universe without a cosmological constant. The
immediate result of the publication of this paper was a urry of popular-level,
sensationalized articles declaring the big bang model to be in trouble, causing
some scientists to consider the possibility that the cosmological constant might
not be zero after all (anticipating the discovery of dark energy). Although
written for a technical audience, the writing style is clear and carefully lays out
how their experiment was done and their conclusions reached, making this an
excellent example of the scientic method in action.
Freedman Wendy L., Barry F. Madore, Brad K. Gibson, et al. 2001. Final Results
fromthe Hubble Space Telescope Key Project to Measure the Hubble Constant.
Astrophysical Journal 553: 4772. As the title explains, this is the nal report of
Freedmans team concerning the value of the Hubble constant. Although this
detailed paper was intended for their professional colleagues, the interested
Selected Bibliography 175
reader will get a sense of the excruciating care taken by astronomers in collect-
ing their data.
Gamow, George.The Evolutionary Universe. September 1956. Scientic American,
136165. A nontechnical review of the cosmological models leading up to the
development of the big bang, written by one of the theorys founders. Gamow
nishes with a summary of his own model, which he interestingly does not
call the big bang (due to the terms original pejorative nature). An important
popular-level paper in the history of modern cosmology.
Gefter, Amanda. 2003. Decoding the Mystery of Dark Energy. Mercury 32(5): 3440.
Various models of dark energy and dark matter are reviewed at a popular-level.
This is a highly recommended rst introduction to dark energy for readers of
limited science background.
Geller, Margaret J. 1990. Mapping the Universe: Slices and Bubbles. Mercury 19(3):
6676. Geller leads the nontechnical reader through a brief history of map-
ping our own world before discussing her seminal research in the mapping of
structure in the universe. Her personal account of the groundbreaking discov-
ery of the bubble-like system of laments and voids makes this an important
reference.
Goldsmith, Donald and Tobias Owen. 2002. The Search for Life in the Universe, 3rd
ed. Sausalito, CA: University Science Books. Touted as the premiere text for
courses dealing with astrobiology, this work lives up to its press. Basics of
astrophysics and biology are surveyed, as well as aspects of interstellar commu-
nication and travel. This text is highly recommended for all readers interested
in life beyond Earth and the problems of interstellar communication.
Gould, Stephen J. 1999. Rocks of Ages. NewYork: Ballantine. Respected paleontologist
and popular-level science writer Stephen Jay Gould investigates the relationship
between science and religion and nds a place of peace in his concept of
nonoverlapping magisteria (NOMA). This work is strongly recommended for
anyone interested in the interface between theology and cosmology, or science
and religion in general.
Guth, Alan. 1997. The Inationary Universe. Reading, MA: Addison-Wesley. This en-
gaging rst-person account of the development of the inationary model is a
must-read for any cosmology enthusiast. Guths humor and humility provide
a refreshing perspective on the process of science and the sometimes thorny
relationships between scientists.
Gyatso, Tenzin (Dalai Lama). 2005. The Universe in a Single Atom. New York: Morgan
Road Books. The widely recognized spiritual leader of Tibetan Buddhism and
Nobel Peace Prize laureate demonstrates that science and religion cannot only
coexist, but can enrich the experience of the other. Although written for a
popular audience, some of the Buddhist philosophy may be confusing to the
casual reader.
Harrison, Edward. 1987. Darkness at Night. Cambridge, MA: Harvard University Press.
Harrison delivers a detailed but accessible survey of the development, history,
and solution of Olbers paradox. Although narrow in focus, it is the seminal
survey of this topic.
. 2000. Cosmology, 2nd ed. Cambridge: Cambridge University Press. Known for
his deep philosophical musings about cosmology, Harrison weaves together his-
torical developments inscience withaninsightful reectionas to their meaning.
176 Selected Bibliography
Although it sometimes reads like a textbook, it is such a unique work that it
belongs on the bookshelf of every serious devotee of cosmology.
Hawking, Stephen W. 1993. Black Holes and Baby Universes and Other Essays. New
York: Bantam Books. This popular-level collection of essays is a much easier
read than Hawkings famous A Brief History of Time, and provides an overview of
both Hawkings research into black holes and Hawking the man, including his
physical struggles with ALS.
, ed. 2002. On the Shoulders of Giants. Philadelphia: Running Press. Hawking
edited this collection of the most seminal works of Copernicus, Kepler, Galileo,
Newton, and Einstein and wrote brief but informative overviews of each scien-
tists life. This thick tome provides the most important works of the Copernican
revolution in one volume, as well as a selection from the work of Einstein, who
overthrew the mechanistic Newtonian universe after nearly 400 years of unri-
valled success. This work is recommended for the reader interested in exploring
the Copernican Revolution through primary texts.
Hawking, Stephen and Roger Penrose. 1996. The Nature of Space and Time. Princeton:
Princeton University Press. This volume is based on a series of alternating
lectures given by the two famed theoretical physicists on the nature of the
interplay between cosmology and quantum mechanics. Humor and occasional
mathematics are intertwined, making this a unique work. It is recommended
for the serious reader.
Heath, Thomas L. 1991. Greek Astronomy. New York: Dover. First published in 1932,
this work begins with a detailed introduction to the contributions of early Greek
astronomers. The remainder of the text is judiciously selected excerpts from
primary texts, some not easily accessible elsewhere.
Hetherington, Norriss, ed. 1993. Cosmology. New York: Garland Publishing. The
secondary title of this work Historical, Literary, Philosophical, Religious, and
Scientic Perspectives accurately summarizes the importance of this collec-
tion of essays. Unlike similar volumes, this work includes cosmologies from
non-Western traditions. Highly recommended for the reader interested in the
inuence cosmology has had on other aspects of human society.
Hubble, Edwin. 1929. A Relation Between Distance and Radial Velocity Among
Extra-galactic Nebulae. Proceedings of the National Academy of Sciences 15: 168
174. Although not the rst scientic paper to announce the linear relationship
between distance and redshift, Hubbles paper was the seminal work on the
topic, and earned the relationship his name. It was clearly written for a technical
audience, so its value is largely historical.
Inglis, Mike. 2003. Observers Guide to Stellar Evolution. London: Springer. Inglis of-
fers an overview of spectral classication and stellar evolution for amateur
astronomers. Although meant for the serious observer of the night sky, the
nontechnical reader will nd this a useful survey of important concepts.
Jones, Bessie Zaban and Lyle Gifford Boyd. 1971. The Harvard College Observatory.
Cambridge, MA: Harvard University Press. This detailed history of the HCO
through 1919 includes a frank discussion of the role of women in the early
twentieth century (including Antonia Maury, Annie Jump Cannon, Williamina
Fleming, and Henrietta Leavitt). Highly recommended for anyone interested
in the important contributions of these women and the conditions under which
they worked.
Selected Bibliography 177
Jones, Mark H. and Robert J. A. Lambourne, eds. 2004. An Introduction to Galaxies
and Cosmology. Cambridge: Cambridge University Press. A textbook designed
for lower-level undergraduates, this work provides an excellent background for
the reader interested in galactic astronomy. The visual learner will appreciate
the plentiful diagrams and illustrations.
Kane, Gordon. 2002. Anthropic Principles. Phi Kappa Phi Forum 82(4): 2124. Kane
addresses the apparent ne-tuning of the universe in accessible language. This
article is a useful summary of various scientic explanations, including the
anthropic principle and multiverse theory.
Kerr, Richard A. 1998. Requiem for Life on Mars? Support for Microbes Fades.
Science 282: 13981400. This concise article demonstrates the scientic method
in action, as evidence for fossil life on Mars was critiqued by the scientic
community and found wanting. This is a valuable summary of an event that
created quite a stir in both scientic and media circles a decade ago, but has
been largely forgotten since.
Kirshner, Robert P. 2002. The Extravagant Universe. Princeton: Princeton University
Press. Harvards Kirshner played a lead role in the discovery of the accelerating
universe. An excellent public speaker, Kirshners style is engaging and reects
the excitement of the process of discovery in science.
Kolb, Edward W. and Michael S. Turner. 1990. The Early Universe. Redwood City, CA:
Addison-Wesley. Two leading particle physicists review the interface between
their eld and cosmology (circa 1990) for a technical audience. Although dense
with mathematics, the text is clearly written and punctuated by the authors wit.
A companion volume contains reprints of seminal papers in particle physics
and cosmology.
Koopmans, Leon V. E. and Roger D. Blanchard. 2004. Gravitational Lenses. Physics
Today 57(6): 4551. This article is an overview of various aspects of gravitational
lensing including microlensing, written at an introductory level. Theory is kept
to a minimum, as the articles main thrust is the important applications of
lensing and the breadth of information that can be obtained.
Kragh, Helge. 1996. Cosmology and Controversy. Princeton: Princeton University Press.
Kraghs seminal work is a detailed study of the development of the big bang
theory and its eventual victory over the rival steady-state. Written by a historian
of science, it presents a different point of viewcompared to similar works written
by scientists. Densely referenced, it gives the interested reader an impressive
variety of avenues for further exploration.
Krauss, Lawrence. 1999. Cosmological Antigravity. Scientic American 280(1): 3441.
Krauss presents an overview of the history of the cosmological constant, and the
reasons for its resurgence (the age crisis and the accelerating universe). Writ-
ten by an important researcher in the elds of globular cluster ages and particle
physics, this paper is important not only for its accessibility but its timeliness
(published at the time of the rst evidence for the accelerating universe).
. 2005. School Boards Want to Teach the Controversy. What Controversy?
Physics and Society 34(3): 9. Written in response to various attempts to include
Intelligent Design in the science classroom, physicist Larry Krauss clearly
lays out his own support for a reasonable coexistence between science and
religion, utilizing specic examples fromboth. This insightful editorial is highly
recommended.
178 Selected Bibliography
Leavitt, Henrietta S. 1912. Periods of Twenty-ve Variable Stars in the Small
Magellanic Cloud. Harvard College Observatory Circular 173: 13. This brief work
is simply one of the most important astronomical papers of the early twen-
tieth century. Leavitt clearly lays out her argument and observations, laying
the groundwork for the most successful method of distance determination in
cosmology. The historical signicance of this paper (including the fact that a
woman was the sole author) cannot be overly stressed.
Lemonick, Michael D. Before the Big Bang. February 2004. Discover, 3541. This
article contains a popular-level discussion of the ekpyrotic and cyclic models.
Quotes from both proponents and skeptics set the models within the larger
process of the scientic method.
Lemonick, Michael D. and J. Madeleine Nash. Unraveling Universe. March 6, 1995.
Time, 7684. One of the many popular-level articles published in the wake
of Wendy Freedmans conrmation that the Hubble constant was in conict
with a at, zero-cosmological constant universe. Both sides of the issue are
represented, as well as the difculties inherent in making such observations. A
avor of the sometimes colorful personalities involved gives this article some
spice.
Lightman, Alan and Roberta Brawer. 1990. Origins. Cambridge, MA: Cambridge
University Press. This text is a valuable collection of interviews with famous cos-
mologists and particle physicists, including several Nobel laureates. It balances
scientic opinions and insights with unique literary snapshots of the physicists
as human beings.
Macpherson, Hector. 1919. The Problem of Island Universes. The Observatory
42(543): 329334. Macpherson offers an excellent and timely overview of cos-
mology in the early twentieth century written for a popular-level audience. It is
highly recommended for readers interested in the tension in cosmology during
this period of time.
Martnez, Vicent, Virginia Trimble, and Mara Jes us Pons-Bordera, eds. 2001. His-
torical Development of Modern Cosmology. San Francisco: Astronomical Society of
the Pacic. These proceedings from a summer school on the history of modern
cosmology contain a rich variety of articles written by some of the heavy-hitters
in the eld as well as lesser-known researchers. Most articles are easily accessible
for the popular-level reader, and provide information not usually included in
introductory trade books.
Munitz, Milton K., ed. 1957. Theories of the Universe. Glencoe, IL: The Free Press.
Munitz has assembled a valuable collection of excerpts from primary texts
that reect cosmology from ancient cultures through the early days of the big
bang/steady-state debate, with few editorial comments. This work is a must-have
for all those interested in reading about the early history of cosmology through
original voices.
Naeye, Robert. Delving into Extra Dimensions. June 2003. Sky and Telescope, 3844.
This article is a clearly written, nontechnical survey of theories incorporating
extra dimensions, including several different brane worlds. Candid comments
by supporters and skeptics provide a balanced approach to these speculative
theories.
Peebles, P. J. E. andBharat Ratra. 2003. The Cosmological Constant andDark Energy.
Reviews of ModernPhysics 75: 559606. This work is a scholarly reviewof the history
Selected Bibliography 179
of the concept of the cosmological constant as well as current observations and
theoretical models. Despite its appearance in a scholarly journal, the paper
contains relatively little mathematics and is recommended for the interested
reader.
Perivolaropoulos, L. 2005. The Rise and Fall of the Cosmic String Theory
for Cosmological Perturbations. Available at http://arxiv.org/PS cache/astro-
ph/pdf/0501/0501590.pdf (accessed September 2, 2006). This paper presents
a review of the history of cosmic strings and their implications for cosmology.
Although written for a technical audience, the serious reader will nd this a
useful and interesting work.
. 2006. Accelerating Universe: Observational Status and Theoretical Implica-
tions. Available at http://arxiv.org/PS cache/astro-ph/pdf/0601/0601014.pdf
(accessed September 1, 2006). The author has written a thoughtful and tech-
nical overview of the discovery of the accelerating universe and the various
models of dark energy. Alternate interpretations and observational difculties
are recounted with honesty and detail, something often ignored in similar pa-
pers. The interested reader can easily skip the occasional mathematics in order
to gain an understanding of the inherent difculties in this area of research.
Randall, Lisa. 2005. Warped Passages. NewYork: HarperCollins. Apopular-level survey
of string and brane theories written by a leading researcher in the eld of brane
worlds. This thick volume will satisfy the reader who is interested in the nuts and
bolts of branes but who lacks the mathematical background to read professional
papers.
Rees, Martin. 2000. New Perspectives in Astrophysical Cosmology, 2nd ed. Cambridge:
Cambridge University Press. This thin volume gives the popular-level reader a
taste of some of the successes and remaining questions of twentieth century
astronomy. Written by a radio astronomer involved in resolving the steady-
state/big bang debate, his personal insights and sharp wit provide a unique
perspective.
Rubin, Vera. 1997. Bright Galaxies Dark Matters. New York: Springer Verlag. In this
series of essays and short scientic papers, Rubin surveys her considerable con-
tributions to our understanding of the role of dark matter in galaxies. Equally
important, her biographical writings paint a vivid picture of the discrimination
she faced during the early years of her career as the rare woman in cosmology.
Sagan, Carl. 1980. Cosmos. New York: Random House. Based on his classic television
series of the same name, Sagan demonstrates why he is the ultimate popularizer
of science. The reader journeys with Sagan through the universe in search of
the ultimate answers, including whether or not we are alone. In the process, the
reader learns more about what it fundamentally means to be human. Although
some of the science is dated, this text (and the revised video series recently
released) are highly recommended to all those with an interest in cosmology.
. 1996. The Demon-Haunted World. New York: Random House. Master popular
science writer Sagan teaches the scientic method through engaging examples.
More importantly, he lays out the difference betweenscience andpseudoscience
and explains the dangers the latter poses to a technological society.
Singh, Simon. 2004. Big Bang. New York: HarperCollins. In this charming popular-
level volume, Singh takes the reader from ancient Greece to cutting-edge re-
search in various aspects of cosmology. Scientic discoveries are placed in their
180 Selected Bibliography
proper context with historical facts not recounted elsewhere. Highly recom-
mended for popular-level readers.
Smith, Sinclair. 1936. The Mass of the Virgo Cluster. Astrophysical Journal 83: 23
30. This is perhaps the rst paper written in English to suggest that unseen
material (dark matter) inhabits galaxy clusters. Although written for a technical
audience, the introduction and conclusion are highly readable and of historical
importance.
Spergel, D. N., R. Bean, O. Dore, et al. 2006. Wilkinson Microwave Anisotropy
Probe (WMAP) Three Year Results: Implications for Cosmology. Avail-
able at http://arxiv.org/PS cache/astro-ph/pdf/0603/0603449.pdf (accessed
September 1, 2006). This technical paper is dense with observations from the
WMAP cosmic microwave background probe and clearly demonstrates how
these observations bolster the concordance model. The importance of these
results made this preprint one of the most highly cited references in cosmology
shortly after its release.
Sproul, Barbara. 1979. Primal Myths. San Francisco: HarperCollins. This work fea-
tures an impressive compilation of creation myths from around the world.
Although organized geographically, this work demonstrates the universality of
certain aspects of creation myths, no matter the country of origin.
Starkman, Glenn D. and Dominik Schwarz. 2005. Is the Universe Out of Tune?
Scientic American 293(2): 4855. This popular-level article clearly explains the
importance of acoustic waves in the early universe. Successes and lingering
problems with the concordance model with regard to WMAP data are laid out
without bias.
Susskind, Leonard. 2006. The Cosmic Landscape. New York: Little, Brown and Co.
Susskind, a leading researcher instring theory, masterfully explains his speciality
as well as topics such as brane theory and the anthropic principle with keen
insight and considerable humor. This book is highly recommended for the
reader interested in these topics but who lacks a technical background.
Tegmark, Max. 2004. Parallel Universes. In Science and Ultimate Reality, ed. John
D. Barrow, Paul C. Davies and Charles L. Harper, Jr., 459491. Cambridge:
Cambridge University Press. One of the most detailed surveys of various multi-
verse models, where the denition of multiverse is taken in its broadest terms.
Tegmark reviews the properties of each model as well as possible means of test-
ing/observing each. The treatment is not mathematical but recommended for
the serious rather than casual reader.
Trimble, Virginia. 1995. The 1920 Shapley-Curtis Discussion: Backgrounds, Issues,
and Aftermath. Publications of the Astronomical Society of the Pacic 107: 1133
1144. Trimble offers a detailed review of the Great Debate, including the
personalities involved and the larger issues facing the society of the time. The
points of contention are clearly laid out, giving each man his due in terms
of where he was correct and where he was limited by the technology of the
day. It is highly recommended for readers interested in early twentieth century
cosmology.
Van den Bergh, Sidney. 1988. Novae, Supernovae, and the Island Universe Hypoth-
esis. Publications of the Astronomical Society of the Pacic 100: 817. An excellent
scientic and historical overview of the original confusion between novae and
supernovae and the effect this had on the island universe debate. This work
Selected Bibliography 181
is recommended for readers interested in this important era in the history of
modern cosmology.
. 2000. The Galaxies of the Local Group. Cambridge: Cambridge University Press.
This text is a detailed study of the galaxies in our cosmic neighborhood. Dense
with facts and references, this volume will be of interest to readers curious about
the vital statistics of our Local Group and its citizens.
VonHippel, Ted, Chris Simpson, andNadine Manset, eds. 2001. Astrophysical Ages and
Time Scales. SanFrancisco: Astronomical Society of the Pacic. These conference
proceedings cover a wide variety of techniques, successes, and problems in
determining the ages of various astronomical objects and the universe at large.
Most of the articles are written for a technical audience, but the serious reader
will gain at least a avor of the successes and challenges of these difcult areas
of research.
Waller, William H. and Paul H. Hodge. 2003. Galaxies and the Cosmic Frontier. Cam-
bridge, MA: Harvard University Press. This work is a detailed yet accessible
survey of current understanding of galaxies and large-scale structure written by
researchers in the eld. It is recommended for the reader especially interested
in these aspects of cosmology.
Webb, Stephen. 1999. Measuring the Universe. London: Springer. Written for the
undergraduate astronomy major, the mathematics can easily be ignored by the
interested popular-level reader. It is an excellent survey of the various rungs of
the distance ladder and the techniques utilized at each level.
Wesson, Paul S., K. Valle, and R. Stabell. 1987. The Extragalactic Background Light
and a Denitive Resolution of Olberss Paradox. Astrophysical Journal 37: 601
606. The authors offer a denitive explanation of the solution of Olbers para-
dox. Although appearing in a technical journal, the introduction and conclu-
sion are accessible to the interested reader.
Wiita, Paul J. 2006. Active Galactic Nuclei: Unication, Blazar Variability and the
Radio Galaxy/Cosmology Interface. Available at http://arxiv.org/PS cache/
astro-ph/pdf/0603/0603728.pdf (accessed September 1, 2006). Wiita has writ-
ten a concise yet detailed survey of the various types of active galactic nuclei
and the successes and limitations of the unied model. Written for a technical
audience, the rst half of the article is accessible to the interested reader, while
the details in the latter half make it difcult going.
Yulsman, Tom. 2003. Origins. Bristol, UK: Institute of Physics. A journey through
modern cosmology as seen through the eyes of key researchers, this popular-
level book balances observation and theory, establishment and speculation.
Index
21-cm radiation, 13, 14, 24, 89, 100,
155
Accretion disk, 34, 35, 87
Acoustic waves, 13439
Active galactic nuclei, unied model,
87, 88
Alpher, Ralph, 69, 72, 142
Ancient Greek astronomy, 3740
Andromeda Galaxy (M31), 33, 50,
5456, 82, 83, 88, 89, 121
Angular power spectrum. See Cosmic
microwave background, angular
power spectrum
Anthropic principle, 14850
Aristotle, 38, 40, 4244, 46, 94,
134
Astrobiology, 15153
Baryogenesis, 114, 117, 130
Big bang, 64, 65, 6873, 109, 113,
12228, 146, 159. See also Universe,
early history; Universe, origin
Blackbody, 16, 17, 72
Black hole, 7, 35, 36, 61, 63, 116,
120, 121; supermassive, 6, 36, 71,
81, 82, 84, 87, 88
Blazars, 86, 87. See also Active galactic
nuclei, unied model
BL Lac objects. See Blazars
Bondi, Hermann, 70, 71
Brahe, Tycho, 27, 35, 43, 44
Branes and braneworlds, 14446
Brown dwarf, 20, 26, 121, 129
Burbidge, E. Margaret, 31, 71, 101
Burbidge, Geoffrey, 31, 71, 101
Cannon, Annie Jump, 18, 20, 35, 159
Cepheids, 3133, 54, 56, 77, 96, 98,
99, 106, 107, 131. See also
Periodluminosity relationship
Closed universe, 6266, 122, 124
Concordance model. See Inationary
CDM model
Copernican revolution, 4147, 51
Copernicus, Nicolaus, 41, 45, 46
Cosmic microwave background, 6, 16,
119, 120, 123, 124, 126, 127, 134;
angular power spectrum, 13640,
143, 147; discovery, 72; prediction,
72
Cosmic strings, 146, 147
Cosmological constant, 6264, 66, 68,
69, 105, 106, 108, 13134, 143,
144, 150
Cosmological principle, 6263, 70, 71,
92, 93, 119, 123
Creation myths, xv, xvi, 37, 69
Critical density, 6466, 122
Curtis, Heber, 54, 55
Dark energy, 13134, 138, 140, 146
157
Dark matter, 8891, 12931, 133, 136,
138, 139. See also Inationary
CDM model
184 Index
Degeneracy pressure, 26, 34, 35.
See also Pauli exclusion principle
Density perturbations, 121, 123, 124,
12629, 138, 139, 146, 147
De Sitter space, 62, 63, 125, 144
Dicke, Robert, 72, 122, 148
Distance, measuring, 2, 3, 23, 43, 55,
56, 93101, 106, 107. See also
Distance modulus formula; Parallax;
Period-luminosity relationship
Distance modulus formula, 9, 23, 32,
56, 97
Doppler broadening, 84, 100
Doppler effect, 17, 5658, 67, 71, 96,
104, 105, 151; and rotation curves,
89, 90
Drake equation, 154
Drake, Frank, 154, 155
Eclipse, xv, 37, 39, 43, 61
Eddington, Arthur, 61, 69
Einstein, Albert, 5968, 90, 91, 108,
114, 141
Einstein eld equations, 5968, 109,
125, 141
Ekpyrotic model, 145, 146
Electromagnetic radiation, 3, 4
Electromagnetic spectrum, 47, 86
Electroweak force, 113, 117, 124
Exoplanets, 25, 119, 151, 153, 154
Extinction, 23, 55, 56, 97, 98, 132
Faber, Sandra, 92, 101
False vacuum, 12528, 150
Finger of God effect, 93, 105
Flat universe, 6366, 1057, 122, 123,
127, 128, 131, 138, 140
Fleming, Williamina Paton, 19, 20
Forces, fundamental 11117, 144,
145, 148; unication of 11216,
14143, 145
Fowler, William, 31, 71, 101
Fraunhofer, Josef, 14, 18
Freedman, Wendy, xviii, 107, 108
Friedmann-Robertson-Walker models,
6366, 68, 105, 122
Fusion. See Stars, energy generation
Galaxy: barred spiral, 75, 76, 7981,
83; clusters, 82, 88, 9092, 105,
107, 121, 130, 131, 139 (see also
Local Group; Virgo cluster); dwarf,
77, 82, 83; elliptical, 7678, 82,
8486, 99, 101; formation, 77, 78,
80, 82, 83, 120, 130, 131, 146;
interactions, 77, 78, 8284, 86, 87,
121; irregular, 83; lenticular, 76;
rotation, 5357, 7880, 8890, 100,
139; spiral, 7578, 8385, 8890,
99100, 107 (see also Spiral nebulae);
spiral arms, 26, 33, 52, 75, 76,
7880, 100
Galilei, Galileo, 9, 25, 38, 45, 46, 48
Gamma ray astronomy, 7, 81, 86, 120
Gamow, George, 6972
General theory of relativity, 35, 5969,
90, 109, 114, 116, 141, 142
Geocentric cosmology, 38, 40, 42, 45
Globular clusters. See Stars, clusters
Gold, Thomas, 70, 71
Grand unied theory, 11316, 124,
146, 157
Gravitational lensing, 91, 139, 147,
151. See also Microlensing
Gravity waves, 61, 129, 140, 143, 146,
147, 157
Guth, Alan, 12426
Halley, Edmund, 46, 48, 94, 96
Hawking, Stephen, xix, xx, 121, 125,
126, 136, 142, 149
Heisenberg uncertainty principle, 110
Heliocentric cosmology, 39, 4146
Herman, Robert, 69, 72, 142
Herschel, John, 49, 51, 52, 56
Herschel, William, 5, 5052
Hertzsprung, Ejnar, 21, 22, 32
Hertzsprung-Russell (HR) diagram, 22,
23, 2931, 33, 9799, 103
Higgs, eld and particle, 115, 116,
124, 125, 127
HII regions, 24, 79, 80, 83, 106
Hipparcos, 8, 39, 40, 96
HI regions, 13, 24, 100
Hoyle, Fred, 31, 70, 71, 101
Index 185
Hubble constant, 57, 58, 71, 101,
1048, 131, 132, 134, 138
Hubble, Edwin, 5658, 63, 68, 82, 106
Hubbles law, 5658, 63, 64, 70, 71,
85, 104, 105, 131; discovered by
Lematre, 68; and distance
determinations, 92, 93, 101, 104,
105
Hubble tuning fork diagram, 7577
Huggins, William, 14, 17, 52
Ination, 73, 106, 107, 117, 12433,
13640, 14447, 150, 159; eternal,
12729, 150, 157; new, 126, 127,
145; old, 12426
Inationary CDM model, 134, 136,
13840, 143
Infrared astronomy, 5, 25, 78, 80, 81,
85, 87, 100, 119
Interstellar medium, 23, 24, 31, 48,
49, 54, 55, 81, 120, 152. See also
Nebulae
Interstellar reddening, 24
Island universe hypothesis, 49, 5155,
57
Kapteyn, Jacobus, 53, 55
Kepler, Johannes, 44, 45, 4750, 89,
135
Large Magellanic Cloud. See Magellanic
clouds
Large-scale structure, 9193, 139
Leavitt, Henrietta Swan, 31, 32
Lematre, Georges, xx, 63, 6870, 106,
132
Light. See Electromagnetic radiation;
Electromagnetic spectrum
Linde, Andrei, 126, 128
Local Group, 77, 8284, 91, 99, 134.
See also Andromeda galaxy;
Magellanic clouds
MACHOS, 129
Magellanic clouds, 31, 32, 76, 83, 84
Magnetic monopoles, 114, 122, 124,
126, 146
Magnitude scale, 8, 9, 21, 39
Main sequence. See Stars, main
sequence
Martian meteorites, possible life found
in, 152, 153
Maury, Antonia, 1922
Messier catalog, 50, 51
Microlensing, 129, 151
Microwave astronomy 5, 6, 72. See also
Cosmic microwave background;
Wilkinson Microwave Anisotropy
Probe
Milky Way, 58, 75, 77, 8284, 103,
121; evolution, 8082, 87; structure,
13, 14, 26, 27, 4955, 7982, 87,
89
M-theory, 143, 144
Multiple universes. See Multiverse
Multiverse, 128, 148, 150, 157
Nebulae, 23, 24, 4955, 79, 84, 119
Neutron star, 7, 35, 61, 121
Newton, Isaac, 4, 11, 12, 37, 4648,
59, 60, 116, 139
Nova, 34, 35, 43, 5456, 99, 100
Nuclear cosmochronometry. See
Radioactivity
Nucleosynthesis: primordial, 69, 71,
72, 11719, 129; stellar, 30, 31, 71,
72, 101, 102, 148
Olbers paradox, 48, 49, 62, 66, 67
Open universe, 6366, 105, 106, 122
Parallax, 32, 38, 41, 43, 9498
Particle physics, 10916, 127, 132,
133, 141, 142; standard model,
10913, 115, 116, 145
Pauli exclusion principle, 34, 35, 111,
121
Peer review, xvii, xviii, 140, 153
Period-luminosity relationship, 32, 33,
54, 56, 9699, 106
Pickering, E.C., 1921
Planck length, 116, 141
Planck, Max, 7, 16, 116
Planetary nebula, 24, 30, 34, 99
186 Index
Pocket universes, 128, 150
Ptolemy, 8, 39, 40, 43, 134
Pulsar, 35, 85
Quantum chromodynamics (QCD),
112, 141
Quantum electrodynamics, 112, 113
Quantum eld theory, 109, 110, 115,
144
Quantum gravity, 11416, 144, 157
Quantum mechanics, 7, 8, 1517, 34,
109, 110, 11416, 125, 142, 149
Quasars. See Quasi-stellar objects
Quasi-stellar objects, 71, 8587, 91,
139, 140. See also Active galactic
nuclei, unied model
Quintessence 133, 146
Radial velocity, 17, 56, 58, 96. See also
Doppler effect
Radio astronomy, 6, 13, 14, 71, 81,
8487. See also 21-cm radiation
Radio galaxies, 35, 71, 8486. See also
Active galactic nuclei, unied model
Radioactivity, 101, 102, 112, 114, 117,
118, 121, 128
Randall, Lisa, 111, 143, 145
Recombination, 72, 118, 119, 134,
136, 138, 139
Red giants, 21, 22, 29, 34, 99. See also
Stars, evolution
Redshift. See Doppler effect; Universe,
expansion
Redshift surveys, 92, 93, 105, 130
Religion and science, xv, xvi, xixxxi,
18, 38, 50, 122, 136; and anthropic
principle, 14851; and big bang,
6870; and Copernican revolution,
4043, 4548
RR Lyrae stars, 31, 80, 83, 99
Rubin, Vera, xvii, xviii, 89
Russell, Henry Norris, 21, 22
Sagan, Carl, xxi, 31, 149, 155, 156
Sandage, Allan, 106, 107
Scientic method, xviixix, 6063,
7073, 128, 129, 14143, 147,
15759
Search for Extraterrestrial Intelligence
(SETI), 15156
Seyfert galaxies, 84, 86, 87. See also
Active galactic nuclei, unied model
Shapley, Harlow, 33, 5456
Singularity, 35, 63, 68, 72, 73, 116,
146
Slipher, Vesto, 56, 88
Small Magellanic Cloud. See Magellanic
clouds
Space-time, 35, 5961, 79, 109, 116,
128, 141, 146
Special theory of relativity, 3
Spectroscopy, 4, 1421, 85, 102, 155.
See also Stars, spectral classication
Spiral nebulae, 5057, 75. See also
Galaxy, spiral
Spontaneous symmetry breaking,
11517, 124, 125
Standard model. See Particle physics,
standard model
Stars: clusters, 26, 27, 33, 5355,
7677, 79, 80, 83, 84, 9699, 120;
cluster ages, 99, 103, 104; 1068;
131, 132; death, 24, 3436, 120,
121 (see also Supernova); energy
generation, 2530, 120; evolution,
2731, 99, 100, 103, 104, 119, 120;
formation, 2426, 7779, 83,
11921; luminosity classes, 22, 23,
97, 98 (see also Hertzsprung-Russell
(HR) diagram); main sequence, 22,
2629, 98103; metallicity, 33, 34,
8082, 98, 99, 103, 119, 120;
populations, 23, 33, 34, 76, 80, 98,
99, 102, 11921; spectral
classication, 1823, 97, 98, 103,
158
Steady-state model, 7072
Steinhardt, Paul, 126, 145, 146
Stellar nucleosynthesis. See
Nucleosynthesis, stellar
String theory, 14147
Supercooling, 12427
Supergravity, 114, 115, 144
Supernova, 24, 31, 34, 35, 55, 85,
121; pair-instability, 120; type Ia, 34,
100, 131, 132; type II, 34, 35, 100
Index 187
Superstrings. See String theory
Supersymmetry, 114, 115, 130, 142
Susskind, Leonard, 133, 149
Telescopes: optical, 912, 45, 46,
4952, 107; radio, 13, 14
Turner, Michael, 107, 146, 149, 150
Ultraviolet astronomy, 6, 24, 85, 87
Universe: accelerating, 66, 68, 69,
131, 132; age, 66, 67, 1014, 124,
131, 132; early history, 6873,
11320, 12227, 13538 (see also
Ination); expansion, 6271, 105,
116, 118, 119, 12428, 13133
(see also Hubbles law; Universe,
accelerating); fate, 6466, 121, 122,
133, 134; origin, xvi, 6873, 116,
127, 128; static, 38, 47, 48, 59, 62,
63, 66
Vacuum energy, 110, 13234
Van Maanen, Adriaan, 5357, 88, 89,
100
Virgo Cluster, 56, 57, 88, 91, 105,
107
Wheeler, John A., 35, 60
White dwarf, 6, 22, 30, 34, 61, 100,
102, 103, 121
Wilkinson Microwave Anisotropy Probe,
120, 13640
WIMPS, 130
Witten, Edward, 141, 144
X-ray astronomy, 6, 7, 35, 36, 81, 84,
85, 87, 90, 91
Zeldovich, Yakov, 72, 130
Zone of avoidance, 52, 54, 55, 79, 83,
92
About the Author
KRISTINE M. LARSEN is Professor of Astronomy and Physics at Central
Connecticut State University. She is the author of Stephen Hawking: A
Biography (Greenwood, 2005).