The Dark Universe Matter Energy and Gravity

Space Telescope Science Institute Symposium

Series 1st Edition Mario Livio (Editor) pdf
download
https://ebookgate.com/product/the-dark-universe-matter-energy-
and-gravity-space-telescope-science-institute-symposium-
series-1st-edition-mario-livio-editor/

Get the full ebook with Bonus Features for a Better Reading Experience on ebookgate.com
Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

A Decade of Extrasolar Planets around Normal Stars Space

Telescope Science Institute Symposium Series Unknown

https://ebookgate.com/product/a-decade-of-extrasolar-planets-around-
normal-stars-space-telescope-science-institute-symposium-series-
unknown/
ebookgate.com

Dark Matter in the Universe John N. Bahcall

https://ebookgate.com/product/dark-matter-in-the-universe-john-n-
bahcall/

ebookgate.com

Mario Bava all the colors of the dark 1st ed Edition Lucas

https://ebookgate.com/product/mario-bava-all-the-colors-of-the-
dark-1st-ed-edition-lucas/

ebookgate.com

Dark Matter 5 By Dust Consumed 1st Edition Don

Bassingthwaite

https://ebookgate.com/product/dark-matter-5-by-dust-consumed-1st-
edition-don-bassingthwaite/

ebookgate.com
The Mirror the Window and the Telescope How Renaissance
Linear Perspective Changed Our Vision of the Universe 1st
Edition Samuel Y. Edgerton
https://ebookgate.com/product/the-mirror-the-window-and-the-telescope-
how-renaissance-linear-perspective-changed-our-vision-of-the-
universe-1st-edition-samuel-y-edgerton/
ebookgate.com

Introduction to Renewable Energy Energy and the

Environment series 1st Edition Nelson

https://ebookgate.com/product/introduction-to-renewable-energy-energy-
and-the-environment-series-1st-edition-nelson/

ebookgate.com

The Space Book Activities for Experiencing the Universe

and the Night Sky 1st Edition Marc Mccutcheon

https://ebookgate.com/product/the-space-book-activities-for-
experiencing-the-universe-and-the-night-sky-1st-edition-marc-
mccutcheon/
ebookgate.com

Causality and Modern Science Third Revised Edition Mario

Bunge

https://ebookgate.com/product/causality-and-modern-science-third-
revised-edition-mario-bunge/

ebookgate.com

Unmanned Space Missions An Explorer s Guide to the

Universe 1st Edition Erik Gregersen

https://ebookgate.com/product/unmanned-space-missions-an-explorer-s-
guide-to-the-universe-1st-edition-erik-gregersen/

ebookgate.com
This page intentionally left blank
SPACE TELESCOPE SCIENCE INSTITUTE

SYMPOSIUM SERIES: 15

Series Editor S. Michael Fall, Space Telescope Science Institute

THE DARK UNIVERSE:

MATTER, ENERGY AND GRAVITY

This book reviews the recent findings on the composition of the universe, its dynamics,
and the implications of both for the evolution of large-scale structure and for fundamental
theories of the universe. With each chapter written by a leading expert in the field, topics
include Massive Compact Halo Objects, the oldest white dwarfs, hot gas in clusters of
galaxies, primordial nucleosynthesis, Modified Newtonian Dynamics, the cosmic mass
density, the growth of large-scale structure, and a discussion of dark energy. The book is
an invaluable resource for professional astronomers and graduate students in this cutting-
edge area of research.
Other titles in the Space Telescope Science Institute Series.

1 Stellar Populations
Edited by C. A. Norman, A. Renzini and M. Tosi 1987 0 521 33380 6
2 Quasar Absorption Lines
Edited by C. Blades, C. A. Norman and D. Turnshek 1988 0 521 34561 8
3 The Formation and Evolution of Planetary Systems
Edited by H. A. Weaver and L. Danly 1989 0 521 36633 X
4 Clusters of Galaxies
Edited by W. R. Oegerle, M. J. Fitchet and L. Danly 1990 0 521 38462 1
5 Massive Stars in Starbursts
Edited by C. Leitherer, N. R. Walborn, T. M. Heckman and C. A. Norman
1991 0 521 40465 7
6 Astrophysical Jets
Edited by D. Burgarella, M. Livio and C. P. O’Dea 1993 0 521 44221 4
7 Extragalactic Background Radiation
Edited by D. Calzetti, M. Livio and P. Madau 1995 0 521 49558 X
8 The Analysis of Emission Lines
Edited by R. E. Williams and M. Livio 1995 0 521 48081 7
9 The Collision of Comet Shoemaker-Levy 9 and Jupiter
Edited by K. S. Noll, H. A. Weaver and P. D. Feldman
1996 0 521 56192 2
10 The Extragalactic Distance Scale
Edited by M. Livio, M. Donahue and N. Panagia 1997 0 521 59164 2
11 The Hubble Deep Field
Edited by M. Livio, S. M. Fall and P. Madau 1998 0 521 63097 5
12 Unsolved Problems in Stellar Evolution
Edited by M. Livio 2000 0 521 78091 8
13 Supernovae and Gamma-Ray Bursts
Edited by M. Livio, N. Panagia and K. Sahu 2001 0 521 79141 3
14 A decade of Hubble Space Telescope science
Edited by M. Livio, K. Noll and M. Stiavelli 2003 0 521 82459 1
The Dark Universe: Matter, Energy
and Gravity

Proceedings of the
Space Telescope Science Institute Symposium,
held in Baltimore, Maryland
April 2–5, 2001

Edited by
MARIO LIVIO
Space Telescope Science Institute, Baltimore, MD 21218, USA

Published for the Space Telescope Science Institute

CAMBRIDGE UNIVERSITY PRESS
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press

The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521822275

© Cambridge University Press 2003

This publication is in copyright. Subject to statutory exception and to the provision of

relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.

First published in print format 2004

ISBN-13 978-0-511-26450-4 eBook (EBL)

ISBN-10 0-511-26450-X eBook (EBL)

ISBN-13 978-0-521-82227-5 hardback

ISBN-10 0-521-82227-0 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
 Contents

Participants vi
Preface ix
A brief history of dark matter
V. C. Rubin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Microlensing towards the Magellanic Clouds: Nature of the lenses and implications
on dark matter
K. Sahu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Searching for the Galactic dark matter
H. B. Richer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Hot gas in clusters of galaxies and ΩM
M. E. Donahue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Tracking the baryon density from the Big Bang to the present
G. Steigman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Modified Newtonian Dynamics and its implications
R. H. Sanders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
Cosmological parameters and quintessence from radio galaxies
R. A. Daly & E. J. Guerra . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
The mass density of the Universe
N. A. Bahcall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Growth of structure in the Universe
J. A. Peacock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
Cosmological implications of the most distant supernova (known)
A. G. Riess . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Dynamical probes of the Halo Mass Function
C. S. Kochanek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
Detection of gravitational waves from inflation
M. Kamionkowski & A. H. Jaffe . . . . . . . . . . . . . . . . . . . . . . . . 162
Cosmological constant problems and their solutions
A. Vilenkin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Dark matter and dark energy: A physicist’s perspective
M. Dine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183

v
 Participants

AbdelSalam, Hanadi Kapteyn Institute for Astronomy

Alam, S. M. Khairul Ohio State University
Albrow, Michael Space Telescope Science Institute
Alcaniz, Jailson Universidade Federal do Rio Grande do Norte
Alcock, Charles Lawrence Livermore National Laboratory
Allen, Ron Space Telescope Science Institute
Anchev, Joseph
Andrews, Thomas
Avera, Randy NASA/FAA
Bahcall, John Institute for Advanced Study
Bahcall, Neta Princeton University
Barish, Barry California Institute of Technology
Beckwith, Steven Space Telescope Science Institute
Benitez, Narciso The Johns Hopkins University
Bennett, David University of Notre Dame
Bergvall, Nils Astronomical Observatory of Uppsala
Bernabei, Rita Universita di Roma II “Tor Vergata”
Blakeslee, John The Johns Hopkins University
Bludman, Sidney DESY-T
Burg, Richard NASA/Goddard Space Flight Center
Burke, Christopher Ohio State University
Burkert, Andreas Max Plancke Institut für Astronomie
Caldwell, Robert Dartmouth College
Canzian, Blaise U.S. Naval Observatory
Carpenter, Kenneth NASA/Goddard Space Flight Center
Casertano, Stefano Space Telescope Science Institute
Chaname, Julio Ohio State University
Cheslow, Melvyn
Chou, C. K. National Central University
Christian, Carol Space Telescope Science Institute
Conselice, Christopher Space Telescope Science Institute
Daly, Ruth Pennsylvania State University
D’Amario, James Harford Community College
de Jong, Jelte Kapteyn Institute
Delahaye, Franck Ohio State University
Dell’Antonio, Ian Brown University
Dine, Michael University of California at Santa Cruz
Donahue, Megan Space Telescope Science Institute
Drake, Andrew Lawrence Livermore National Laboratory
Duerbeck, Hilmar W. Brussels Free University (VUB)
Dvali, Gia New York University
Fang, Taotao Massachusetts Institute of Technology
Fazio, Giovanni Harvard-Smithsonian Center for Astrophysics
Felten, James NASA/Goddard Space Flight Center
Ferguson, Harry Space Telescope Science Institute
Fields, Dale Ohio State University
Freudling, Wolfram ST-European Southern Observatory
Gaitskell, Richard University of California at Berkeley

vi
 Participants vii
Gerhard, Ortwin University of Basel
Gerke, Brian Cambridge University
Giavalisco, Mauro Space Telescope Science Institute
Glicenstein, Jean-Francos CEA-Saclay
Godon, Patrick Space Telescope Science Institute
Graber, James Library of Congress
Greyber, Howard Greyber Associates
Guimaraes, Antonio C. Brown University
Hämmerle, Hannelore Universität Bonn
Hansen, Bradley Princeton University
Hartnet, Kevin NASA/Goddard Space Flight Center
Hauser, Michael Space Telescope Science Institute
Henriksen, Mark University of Maryland, Baltimore County
Hoekstra, Henk CITA
Jain, Bhuvnesh University of Pennsylvania
Jeletic, James NASA/Goddard Space Flight Center
Kamionkowski, Marc California Institute of Technology
Kassin, Susan Ohio State University
Kazanas, Demosthenes LHEA/NASA/Goddard Space Flight Center
Kimble, Randy NASA/Goddard Space Flight Center
King, Lindsay University of Bonn
Kirkham, Barry TRW/NASA/Goddard Space Flight Center
Kochanek, Chris Harvard-Smithsonian Center for Astrophysics
Leckrone, David NASA/Goddard Space Flight Center
Lima, Jose Ademir Universidade Federal do Rio Grande do Norte
Lin, Yi-Hui National Central University
Livio, Mario Space Telescope Science Institute
Macchetto, Duccio Space Telescope Science Institute
Maoz, Dan Columbia University
Margon, Bruce Space Telescope Science Institute
Marochnik, Leonid Space Telescope Science Institute
Marshall, Jennifer Ohio State University
Mashchenko, Sergiy University of Montreal
Mathews, Grant University of Notre Dame
McKernan, Barry University College of Dublin
Medvedev, Mikhail CITA, University of Toronto
Meylan, Georges Space Telescope Science Institute
Miralles, Joan-Marc Space Telescope European Coordinating Facility
Natarajan, Priyamvada Yale University
Nelson, Cailin IGPP/Lawrence Livermore National Laboratory
Netterfield, C. Barth University of Toronto
Niedner, Mal NASA/Goddard Space Flight Center
Nomoto, Ken’icki University of Tokyo
Onken, Christopher Ohio State University
Osmer, Patrick Ohio State University
Paulin-Henriksson, Stephane PCC, College of France
Peacock, John European Southern Observatory
Perlmutter, Saul Lawrence Berkeley Laboratory
Pringle, James Institute of Astronomy, Cambridge
Rauscher, Bernard J. Space Telescope Science Institute
viii Participants
Reid, Iain Neill Space Telescope Science Institute
Rhodes, Jason NASA/Goddard Space Flight Center
Richer, Harvey University of British Columbia
Riffeser, Arno Sternwarte München
Riess, Adam Space Telescope Science Institute
Rosati, Piero European Southern Observatory
Rubin, Vera Carnegie Institute of Washington
Runyan, Marcus California Institute of Technology
Ryden, Barbara Ohio State University
Sahu, Kailash Space Telescope Science Institute
Sancisi, Renzo Kapteyn Sterrekundig Institute
Sanders, Bob Kapteyn Sterrekundig Institute
Schild, Rudolph Smithsonian Astrophysical Observatory
Schneider, Peter Universität Bonn
Schreier, Ethan Space Telescope Science Institute
Seitter, Waltraut C. Münster University
Shanks, Tom University of Durham
Silverberg, Robert NASA/Goddard Space Flight Center
Sparmo, Joe NASA/Goddard Space Flight Center
Stecher, Theodore NASA/Goddard Space Flight Center
Steigman, Gary Ohio State University
Steinhardt, Paul Princeton University
Stiavelli, Massimo Space Telescope Science Institute
Struble, Mitchell University of Pennsylvania, Lockheed Martin
Swaters, Robert Carnegie Institute of Washington
Tavarez, Maritza University of Michigan
Tinker, Jeremy Ohio State University
Tripp, Todd Princeton University
Turner, Michael University of Chicago
Tyson, Anthony Bell Laboratories, Lucent Technology
Urry, C. Megan Space Telescope Science Institute
Vilenkin, Alex Tufts University
Williams, Bob Space Telescope Science Institute
Wilson, Gillian Brown University
Woodgate, Bruce NASA/Goddard Space Flight Center
Xie, Gaofeng Purple Mountain Observatory
Yamamoto, Kazuhiro Hiroshima University
Yaqoob, Tahir The Johns Hopkins University
Zheng, Zheng Ohio State University
 Preface. Through a glass, darkly

The planet Uranus was discovered in 1781 by the British astronomer William Her-
schel. Not long after its discovery, astronomers charting the orbit of Uranus found small
discrepancies between the predicted and observed positions of the planet. In Septem-
ber 1845, British astronomer John Adams proved mathematically that the deviations
in Uranus’ orbit could not result merely from the gravitational pull of the other known
planets and he predicted the existence of another, previously undetected planet in the
solar system. The eventual discovery of the planet Neptune in September 1846 by the
German astronomer Johann Galle thus marked the first detection of astronomical “dark
matter” whose presence was first deduced by its gravitational effects. However, in the
history of physics, we also find a case in which the assumption about the existence of
an unseen medium was later proven to be totally wrong. Until 1887, physicists assumed
that aether —a substance that pervades all space—was a necessary medium for the prop-
agation of light. A famous experiment by American researchers Albert Michelson and
Howard Morley not only showed unambiguously that this medium does not exist, but
the experimental results also set Einstein on the road to a new theory of space and
time—special relativity.
Astrophysicists today are faced with a similar “Neptune vs. aether” dilemma. On the
face of it, there are many indications that about 90% of the matter in our universe is in the
form of “dark matter”—matter whose constituents do not emit electromagnetic radiation
and that interact very weakly with ordinary matter. The luminous galaxies we see are
just like the tiny minilights on a huge, dark, Christmas tree. The existence and amount
of the “dark matter” is deduced, for example, from the speeds of galaxies in clusters of
galaxies. In equilibrium, the gravitational force of all the matter in the cluster exactly
balances the proneness of the galaxies to scatter in all directions. Careful determinations
of the speeds thus “weigh” the cluster. Other observations, like gravitational lensing—the
bending of light from distant sources by the cluster’s gravity—also confirm that about
90% of the mass in clusters is dark.
The most likely candidates for the constituents of the dark matter are some exotic
elementary particles that are relics of the very early, high-energy universe. Elementary
particle theories that link fermions (that have a fractional quantum mechanical spin)
and bosons (with integer spin) are known as supersymmetry (or, affectionately, SUSY)
theories. Supersymmetry requires the existence of (yet undiscovered) fractional spin,
neutral, massive partners to integer spin particles like the photon. The lightest members
of this menagerie of SUSY particles are known as neutralinos and they are the leading
candidates for dark matter.
However, there is another possibility, in principle, to explain the extra gravity usually
attributed to dark matter. The idea behind this alternative is similar in spirit to the lesson
learned from the aether. Instead of requiring the existence of an unseen medium, maybe
the theory of gravity itself needs to be changed. One proposed modification suggests that
our three-dimensional (plus time) universe with all of its elementary particles is stuck
to a (three-dimensional) membrane that exists in a higher-dimensional space. Particles
like protons and electrons cannot move in the extra dimensions and neither can the
electromagnetic fields (a bit like electrons being confined to move along a copper wire).
Gravity and its carrier—the graviton—can, on the other hand, extend and travel into the
higher-dimensional space. In this model, the gravitational effects we attribute to dark
matter could simply represent the gravity of matter that resides in a membrane/universe
parallel to ours. Photons cannot travel throughout the extra dimension separating the
ix
x Preface. Through a glass, darkly
parallel universes and consequently the matter in the parallel universe is necessarily
“dark” to our detectors.
Luckily, experiments planned for the coming decade will be able to distinguish between
the “Neptune” and “Aether” options. The Large Hadron Collider (LHC), the world’s
most powerful particle accelerator which is being built in Geneva, is less than a decade
away from achieving the energy range (proton beams with 7-on-7 TeV) needed to discover
neutralinos. The LHC could also discover particles predicted to exist by the new theories
of gravity. Furthermore, the new theory predicts deviations from Newton’s inverse square
law at submillimeter distances. No fewer than four experiments are expected to test
gravity at these small distances during the coming few months to years.
As if the existence of dark matter was not puzzling enough, since 1998 there exists
strong evidence that most of the universe’s total energy density is in the form of an
even more mysterious “dark energy.” Observing a few dozen stellar explosions known
as Type Ia supernovae at redshifts of order z ∼ 0.5–1, two teams discovered that the
expansion of the universe is accelerating!
Type Ia supernovae are extremely bright (occasionally outshining an entire galaxy)
events representing the complete thermonuclear disruptions of white dwarf stars. Since
Type Ia supernovae are nearly perfect “standard candles” (their small deviations from a
constant luminosity are well calibrated), they can be used as superb distance indicators to
distances spanning half the universe’s age. The expectation prior to 1998 was that distant
supernovae would reveal that the universe had been expanding in the past faster than
predicted by a simple Hubble expansion, because of the deceleration caused by gravity.
Instead, the two teams found (independently) that the distant supernovae were receding
slower than the Hubble law, implying an accelerating cosmic expansion propelled by
some “dark energy.” The pressure associated with this dark energy is negative, resulting
in gravity becoming a repulsive force. The observations, together with measurements of
the anisotropy of the cosmic microwave background radiation, suggest that the energy
density in the dark energy is about 73% that required for a geometrically flat universe.
The precise nature of the dark energy is probably the greatest mystery of today’s
physics. It is generally assumed that this dark energy represents the energy associated
with the physical vacuum. However, the value of the observed energy density is some
55 orders of magnitude smaller than that expected from supersymmetry considerations.
Currently, it is not even clear if the dark energy density is constant in time, as would
be expected for Einstein’s “Cosmological Constant” (introduced to produce a static uni-
verse), or evolving as some uniform scalar field (dubbed “quintessence”). It is also pos-
sible, in principle, that the accelerating universe and the deduced dark energy are also
manifestations of the need for a new theory of gravity.
The Space Telescope Science Institute Symposium on “The Dark Universe: Matter,
Energy, and Gravity” took place during 2–5 April 2001.
These proceedings represent a part of the invited talks that were presented at the
symposium, in order of presentation. We thank the contributing authors for preparing
their papers.
We thank Sharon Toolan of ST ScI for her help in preparing this volume for publication.

Mario Livio
Space Telescope Science Institute
Baltimore, Maryland
 A brief history of dark matter
By V E R A C. R U B I N
Department of Terrestrial Magnetism, Carnegie Institution of Washington

1. Introduction
The title not withstanding, this is not a history of dark matter. Until we know what
the dark matter is, we cannot know its history. Instead, this is a brief history of how
astronomers converged to the view that most of the matter in the universe is dark.
This paper deals principally with the early studies which helped to answer the questions
“Are rotation curves flat? If so, why?” It also includes some early history in deciphering
the signature of clusters of galaxies as gravitational lenses, which seems to have been
little investigated. This account covers the years up to 1980; achievements since 1980
are science, not history. Several excellent, informative brief histories exist, and interested
readers should see Trimble (1987, 1995) and van den Bergh (1999). We can all thank
Sidney van den Bergh for correctly translating Zwicky’s “dunkle (kalte) materie” as
“dark (cold) matter” and finally putting to rest the myth that Zwicky called it “missing
matter.”
The notion that there are stars that are dark was a common one in the 18th and 19th
Century. Walt Whitman’s (1855) lines in Leaves of Grass, “The bright suns I see and the
dark suns I cannot see are in their place” and Bessel’s “Foundation of an Astronomy of
the Invisible” (Clerke 1885 and reference therein) are early manifestations of this belief.
Based on his decade-long observations of Sirius and Procyon, Bessel announced in 1844
that each was a binary star system, whose irregular motion on the sky was due to the
presence of its invisible companion. When Maria Mitchell taught her students at the new
(1865) Vassar Female College, one of the classes featured Dark Stars.
Even earlier, the Reverend Mr. Mitchell (1784) had imagined a star, 500 times the
solar radius but of equal density, and realized that “all light emitted from such a body
would be made to return toward it, by its own proper gravity.” Moreover, “if there really
should exist in nature [such] bodies... since their light could not arrive at us, we could
have no information from light. Yet if any other luminous bodies should happen to revolve
about them, we might still perhaps from the motions of these revolving bodies infer the
existence of the central ones.” An impressive intuition, almost 200 years before we knew
of black holes.
And of course, Vincent van Gogh’s “Starry Night” (1889) is surely every optical as-
tronomer’s nightmare of what the night sky would look like if the dark matter were not
dark.

2. The early 20th century

Kapteyn (1922), in his efforts to study the arrangement and motion of the sidereal
system, estimated “the amount of dark matter” in a universe with the sun at the center
of similar ellipsoids of revolution. Using star counts and physics, he concluded that “it
appears at once that this (dark) mass cannot be excessive.” Kapteyn references Jeans
(1919, p. 239) for a dark star determination. However, in my 1919 edition of Jeans’ book,
there is no mention of dark stars on the cited page 239. Instead, the only estimate I have
located in a quick perusal is on page 222, where Jeans writes “these estimates evaluate
1
2 V. C. Rubin: A brief history of dark matter
the density of matter in the bright stars only; the dark stars, of which it is impossible
even to guess at the number, will increase the density to a quite unknown extent, so that
the estimates only provide lower limits to the true density.” But in a few years Jeans had
changed his mind. With later work, Jeans (1922) counts “about three dark stars in the
universe for every bright star.” Trimble (1995) has pointed out that this range of dark
matter density matches closely the range of dark matter density currently discussed. A
decade later, Oort’s (1932) study of the mass density in the Galactic plane also left its
mark with the name “Oort limit.”
Zwicky’s (1933; also 1937c) analysis of the velocity dispersion for galaxies in the Coma
cluster marks the beginning of the contemporary study of dark matter in the universe,
albeit a slow beginning. His study, plus that of Smith (1936) for the Virgo cluster, noted
that the large relative motions for individual galaxies would disrupt the clusters, unless
each galaxy has a mass about 100 times the accepted mass. Zwicky also cited good
evidence that clusters are not dissolving.
The discrepancy between the high galaxy masses calculated from the viral mass of
the clusters, and the low masses calculated from the very inner rotation curves for five
galaxies, troubled Hubble (1936). “The discrepancy seems to be real and is important,”
he wrote. It is not surprising that these early absorption line rotation curves extended
only over the brightest nuclear regions, and were poor indicators of galaxy mass. Several
decades would pass before the cluster dark matter would be associated with the flat
rotation curves derived for individual galaxies.
Observations of M31’s rotation by Babcock (1939) and Mayall (1951) extended major
axis rotation velocities to almost 120 from the nucleus, but exposure times were tens of
hours, and spectrographs had stability problems. Interestingly, both Babcock’s velocities
for M31 and Humason’s unpublished velocities for NGC 3115 showed the last measured
point to have a rotation velocity of over 400 km s−1 (almost two times actual), but
consequently raised questions of mass distribution.
At a symposium celebrating the dedication of the McDonald Observatory in 1939, Oort
(1940) noted that “... the distribution of mass [in NGC 3115] appears to bear almost no
relation to that of the light.” His conclusion, “The strongly condensed luminous system
appears embedded in a large more or less homogeneous mass of great density,” was a
clear statement of the puzzle that would grip astronomers again in the 1970s. However,
it seems to have impressed few in 1940 and in the decades following.

3. Instrumentation starts catching up: Mid-century

Early radio observations of neutral hydrogen in external galaxies showed a slowly
falling rotation curve for M31 (van de Hulst el al. 1957) and a flat rotation curve for M33
(Volders 1959). Because the flatness could be attributed to the side lobes of the beam,
it was consequently ignored. Louise Volders must also have realized that a flat rotation
curve conflicted with the value of the Oort constants for our Galaxy, which implied a
falling rotation curve at the position of the sun. Jan Oort was one of her thesis professors.
With eight rotation curves available by 1959, de Vaucouleurs (1959) concluded “In
all cases the rotation curve consists of a straight inner region... beyond which the rota-
tional velocity decreases with increasing distance to the center and tends asymptotically
toward Kepler’s third law.” From a reanalysis of the same scattered velocities for M31,
Schwarzschild (1954) reached the opposite conclusion. He stated that the approximately
flat rotation curve was “not discordant with the assumption of equal mass and light
distribution.” With the 20/20 vision of hindsight, plots of the data reveal only a scatter
of points, from which no certain conclusions can be drawn.
 V. C. Rubin: A brief history of dark matter 3
A paper that was more in the spirit of what was to come than what had taken place
was Kahn and Woltjer’s (1959) investigation of the dynamical stability of the Local
Group. They concluded that the Local Group must contain an appreciable amount of
invisible matter. In a sense, this was a contemporary formulation of Zwicky’s virial cluster
problem.
Although Zwicky’s cluster results were not forgotten, they only came to the forefront of
astronomy in discussions of the stability of clusters. At two symposia in Santa Barbara
preceding the Berkeley 1961 General Assembly, Ambartsumian (1961) had significant
support for his view that clusters were explosively disintegrating (but see van den Bergh
1961; 1999). Much of the discussion centered on galaxy radial velocities which were
beginning to be obtained in fairly large numbers. My notes from the symposia mention
dark matter once, in connection with the disks of early type spirals.
In an effort to learn how our Galaxy “ended,” my graduate students at Georgetown and
I made a study of the three dimensional velocities of almost 1000 O and B stars beyond
the solar circle (Rubin et al. 1962). Our 1962 conclusion, “the stellar (rotation) curve is
flat, and does not decrease as is expected for Keplerian orbits,” apparently influenced very
few, and was not emphasized by the senior author when she returned to the problem of
galaxy rotation a decade later. In my earliest Kitt Peak observing, I attempted to obtain
a rotation curve for the Galaxy beyond the solar circle by observing O and B stars near
the anticenter direction (Rubin 1965). It was clear that while many studies of the center
of the Galaxy were underway, there was little attention being paid to the outer limits of
galaxies.
The many galaxies studied by Margaret and Geoffrey Burbidge (e.g. Burbidge, Bur-
bidge, & Prendergast 1962) generally showed an inner velocity rise; velocities were then
expected to fall. For at least one galaxy, NGC 7331 (Rubin, Burbidge, & Burbidge 1965),
we showed three possible density laws which extended the velocity curve beyond the
turnover; one predicted rotation curve is rising slightly, one is slightly falling. This was
our way of saying that we did not know what happens beyond the final measured veloc-
ities.

4. The decade of seeing is believing: The seventies

Science often advances when ideas, formerly very disparate, are united. In retrospect,
it took a long time for astronomers to relate Zwicky’s dark matter to the flat rotation
curves for some galaxies that were beginning to attract attention. If I were to choose
a date when astronomers decided that dark matter must “really” exist, I would pick
1978. In 1977, many astronomers hoped that dark matter might be avoided; by 1979 the
Annual Review article by Gallagher and Faber convinced most of the remaining skeptics.
Of the two questions to be answered, “Are rotation curves flat? If so, why?” we would
arrive at an answer to the first.
Freeman (1970) discussed 21-cm velocity maps of a number of nearby galaxies. For
NGC 300 and M33 he concluded “if [these data] are correct, there must be in these
galaxies additional matter that is not detected, either optically or at 21 cm. Its mass must
be as large as the mass of the detected galaxy, and its distribution must be quite different
from the exponential distribution which holds for the optical galaxy.” His remarks are
reminiscent of Oort’s (1940) concerning NGC 3115, but technology had now made it
possible to measure velocities beyond the optical galaxy.
In the same year, Kent Ford and I completed our study of the velocities of emission
regions in M31 (Rubin & Ford 1970), and produced a rotation curve which extended to
120 , the extent of the optical disk. The curve was flat over the last 30% of the galaxy.
4 V. C. Rubin: A brief history of dark matter

Figure 1. Rotation curves for the galaxies M31, M101, M81 and the rotation curve of our
Galaxy, from Roberts and Rots (1973).

Perhaps that is why we chose not to employ the typical mass models that assumed a
Keplerian fall beyond the last observed region, but instead wrote that “extrapolation (of
the mass) beyond that distance is clearly a matter of taste.” In an era when rotation
curves were routinely extended in a Keplerian fashion beyond the final observed point,
we chose not to do so.
More flat 21-cm rotation curves followed. Rogstad and Shostak (1972) and later Krumm
and Salpeter (1976) obtained flat 21-cm rotation curves for more galaxies, but rumors of
sidelobe problems continued to plague such studies. Mort Roberts, whose very extended
rotation curve of M31 (1995) would lead to the acceptance of flat rotation curves, de-
layed that acceptance by publishing (Roberts & Rots 1973) a plot (Fig. 1) of superposed
rotation curves for M31, M101, our Galaxy, and M81. For our Galaxy and M81 the outer
velocity decreases were so substantial that the eye of the beholder remembered mostly
the falling parts.
However, the theorists had their eyes wide open. Ostriker, Peebles, and Yahill (1974)
introduced their paper on galaxy masses with the stunning sentence “There are reasons,
increasing in number and quality, to believe that the masses of ordinary galaxies may
have been underestimated by a factor of 10 or more.” This work, plus that of Einasto,
Kaasik, and Saar (1974) “The mass of galactic coronas exceeds that mass of populations
of known stars by one order of magnitude, as do the effective dimensions,” made use of
arguments both observational and theoretical. They emphasized the evidence showing
that masses of nearby giant spirals increase linearly with radius from 20 to 100 kpc and
to as much as 500 kpc. These papers, coupled with an earlier paper (Ostriker & Peebles
1973) which demonstrated (with 150 to 500 mass points) that disk galaxies were “found
 V. C. Rubin: A brief history of dark matter 5

Figure 2. Rotation velocities in M31 as a function of distance from the nucleus. Optical Hα
velocities come from Rubin and Ford (1970); 21-cm velocities come from Roberts and Whitehurst
(1975).

Figure 3. Rotation velocities for NGC 3115 from stellar absorption lines from Rubin,
Peterson, and Ford (1976; @American Astronomical Society).

to be rapidly and grossly unstable to barlike modes,” attracted considerable attention.

It is surprising that it did not send scores of observers scurrying to their telescopes.
By 1975, the 21-cm studies by Mort Roberts and his collaborators produced (Roberts
& Whitehurst 1975) a rotation curve for M31 (Fig. 2) that extended to 35 kpc (175 ),
almost 50% farther than the optical curve (Rubin & Ford 1970). At least for M31, there
was no longer room for argument: the rotation curve was mathematically flat, over more
than 50% of the detectable disk, much of it beyond the optical galaxy. For this best
observed disk galaxy, astronomers knew that mass rose linearly with radius, and did not
asymptotically approach a limit.
6 V. C. Rubin: A brief history of dark matter
Flat rotation curves were not restricted to spirals. Rubin, Peterson, and Ford (1976)
submitted an AAS abstract that consisted of only a plot and a title: The Rotation Curve
of the E7/S0 Galaxy NGC 3115 (Fig. 3). It was a textbook plot: lat, steep, flat.
For a few astronomers’ views of rotation curves in 1977, we have an impeccable source:
a discussion at the Yale meeting, The Evolution of Galaxies and Stellar Populations
(Tinsley & Larson 1977). I reproduce some of the discussion which followed the talk by
Freeman.
FREEMAN (to Krumm and Salpeter): Rumor has it that your flat rotation curves
may be affected by sidelobe problems. Could you comment please?
SALPETER: Sidelobe problems affect only the last point on a rotation curve.
Even if the last point were removed, a large enough range of radial distances remains
to demonstrate flat rotation curves.
FREEMAN: Is anyone prepared to offer any alternative way of explaining the
flat rotation curves aside from having a massive halo?
M. ROBERTS: Why do you need a massive halo? Why not an unmassive disk?
The mass goes up with the radius so it only need to be doubled. All you need is
the most common type of star in the solar neighborhood, M dwarfs, to account for
the M/L.
The problem with the flat rotation curves is not peculiar to only one telescope,
but is common to all the large telescopes that have been used. Any given case may
be arguable, but overall there do seem to be flat rotation curves—although not
ALL rotation curves are flat.
L. SMITH: If it were true that 50% of stars formed were below 0.1 M and if
it were true that only MASSIVE star formation stops at ≈ 13 kpc galactic radius,
would that solve the problem.
M. ROBERTS: Yes.
KING: Although I’m not particularly fond of massive halos, I can think of one
argument that would favor an extended mass being in a halo rather than in a disk.
This is based on the Westerbork observations of the high incidence of disks with
twisted edges. If a twisted disk has much mass in its outer parts, it is very hard
for it to maintain a clean twist without turning into a washboard. But it is much
easier to maintain the twisted shape if the mass is in a halo instead.
OSTRIKER: From observations by Hy Spinrad and myself, it does seem difficult
to account for the mass with ordinary late M dwarfs, for they would give too much
light. Of course, stars of even lower mass are possible. My other point is that if all
the mass is in a flat, cold disk it is very likely to be unstable. But since the mass is
invisible, it could be in a flat, hot disk.
As these exchanges indicate, astronomers were willing, in 1977, to accept that some
rotation curves were flat. But at the same meeting, Rees (1977) focused his talk, Galaxy
Formation, on the “implications of massive halos and ‘missing mass’ (which, if the par-
ticipants in this conference are an unbiased sample, are seriously believed to exist...).”
Yet not a single author referenced Zwicky’s studies of dark matter in clusters of galaxies.
The next year, Bosma (1978) completed his thesis, observing and compiling 21-cm
rotation curves for 25 galaxies (Fig. 4). All but a few had flat or almost flat curves.
Only M81, M51, and M101 showed significant outer falling velocities; explanations in
terms of tidal interactions would ultimately arise. Also in 1978, Kent Ford and I (Rubin
& Ford 1978) published data for eight rotation curves of high luminosity spirals, and
photos (Fig. 5) which showed their emission line spectra all strikingly flat to the eye.
I think that Bosma’s plots, plus the visual spectra, convinced many astronomers that
rotation curves are flat. Not flat was the exception. But there were still non-believers.
One eminent astronomer said to me “When you observe low luminosity galaxies, you’ll
find Keplerian falling rotation curves.” Not so, of course. We know now that the lower
 V. C. Rubin: A brief history of dark matter 7

Figure 4. 21-cm rotation curves for 25 galaxies, from Bosma’s (1978) thesis.

the luminosity, the fractionally more dark matter required. Kalnajs’ (1983) insistence
that dark matter is not required, at least for a few galaxies with spatially limited data,
convinced a few astronomers that dark matter could be avoided. In retrospect, I think it
is fair to say that many astronomers hoped that Kalnajs was right; dark matter was to
be avoided, if at all possible.
In their review “The Kinematics of Spiral and Irregular Galaxies,” van der Kruit and
Allen (1978) waffled in their conclusion. “It is certainly true that more mass is waiting
to be found beyond the last measured HI points on many rotation curves... However,
the great increase (factors of 10 to 100) in masses favored by Einasto et al. (1974) and
by Ostriker et al. (1974) involve estimates at much greater distances, from 200–500 kpc.
...There is no evidence in favor of such massive halos within the visible disks of galaxies...
8 V. C. Rubin: A brief history of dark matter

Figure 5. Spectra showing emission lines (dark) of Hα and [NII] for galaxies of different Hubble
types, taken with the Kitt Peak 4-m spectrograph plus Carnegie image tube. Exposures range
from 120 to 200 minutes (Rubin and Ford 1978).

We must conclude that the results from rotation curves are not inconsistent with the
existence of extensive, massive halos around galaxies, although the prime evidence for
them comes from studies of binary galaxies and outlying globular clusters (e.g. Turner
and Ostriker 1977, Sargent 1977).”
Only one year later, a comprehensive review by Faber and Gallagher (1979) concluded
more emphatically “After reviewing all the evidence, it is our opinion that the case for
invisible mass in the Universe is very strong and getting stronger.” Finally, a paper
(since Hubble 1936) had united in print Zwicky’s dark matter with flat rotation curves
of galaxies.
Ostriker (1999) chose Zwicky’s (1937c) Astrophysical Journal paper “On the Masses
of Nebulae and Clusters of Nebulae” for reprinting in the ApJ Centennial Issue (Abt
1999). In a glowing discussion of Zwicky’s paper, Ostriker wrote “Thus, we (Ostriker
et al. 1994) took comfort in the fact that our estimates for the total mass were consistent
with those reached by Zwicky decades earlier.” Yet even this ground-breaking paper of
 V. C. Rubin: A brief history of dark matter 9
Ostriker et al. (1994) failed to reference Zwicky. However, these authors were in good
company. A decidedly incomplete survey of papers in the 1960s and 1970s (but which
does include most of the relevant papers referenced in the present paper) turned up not
a single paper pre-Faber and Gallagher (1979) that referenced Zwicky.
Following the influential Faber and Gallagher (1979) review, it was “general belief”
that rotation curves were flat. As more rotation curves accrued, dark matter became
the accepted cause. But in as much as we have not yet succeeded in identifying the
composition of dark matter, attributing flat rotation curves to ‘dark matter’ seems at
times only a semantic construct.

5. Another approach: Gravitational lenses

As usual, Zwicky (1937a, 1937b, 1937c) was right. Gravitational lenses do offer a
method for inferring the existence of dark matter in clusters of galaxies.
Early work concentrated on single galaxies as lenses. Just a few years after detecting the
deflection of light by the sun during the 1919 total solar eclipse, Frost (1923, mentioned in
Zwicky 1937b) suggested searching for the gravitational deflection of background sources
by stars. A decade later, Zwicky (1937a, 1937b, 1937c) realized that the more massive
galaxies would be far more important as lenses. Refsdal (1964) cites the literature through
1964, and Barnothy (1989) traces the detailed history preceding his suggestion (Barnothy
1965) that QSOs were lensed Seyfert galaxies.
Clusters of galaxies as gravitational lenses have a convoluted history. During tests
of a SIT Vidicon, Jim Westphal (1973) observed the cluster Abell 370 with the 200-
inch telescope. Images comparing the cluster photographed with the SIT Vidicon and
an unaided photographic plate were published by Science magazine. In retrospect, both
images are very noisy, but do show a gravitational arc, not mentioned. Hoag (1981)
discussed a filament-like feature (the arc) in A370 at an AAS meeting; the source of the
image is a KPNO 4-m prime focus plate.
A decade after the SIT Vidicon images, Butcher, Oemler, and Wells (1983) published a
beautiful Kitt Peak 4-m prime focus image of A370 (Fig. 6), taken for their detailed study
of galaxies in clusters. The very sharp and prominent arc is not referred to in the paper.
When I sent an email request to Butcher and Oemler to reproduce the image in this
paper, I gently warned them that is was included in papers which had been published,
but had not noted the arc. Oemler replied (almost instantly) “Of course you may use
it, and of course you may mention that we stared at that damn arc for endless hours
without recognizing it—I always do.”
Several years later, Lynds and Petrosian (1986) presented a paper “On the Giant Lu-
minous Arcs in Clusters” at the AAS meeting. They noted their properties: spatially
coherent, narrow arc-like shape of enormous lengths, whose radii of curvature point to-
ward the central cD galaxies and the centers of gravity of the clusters. As in the previous
work, a gravitational lens was not mentioned. A discussion with Lynds (private com-
munication, 2000) suggests that he and Petrosian were considering various explanations,
but were waiting to get a good spectrum of the arc.
Soucail and colleagues solved the puzzle, following a false start. Initially they (Soucail
et al. 1987a) attributed the arcs “in” clusters to a characteristic of the cluster, perhaps
a cooling flow.” Katz (1987) and Milgrom (1987) thought them to be light echos from a
beaming cluster source. Later, from initial spectra of poor quality, Soucail et al. (1987b)
identified the arc in A370 as a gravitational lens. The next year, Soucail et al. (1988), in
a paper received by A&A on November 17, 1987, showed a beautiful spectrum taken with
an arc-shaped slit, which confirmed that the arc is the image of a galaxy at z = 0.724,
10 V. C. Rubin: A brief history of dark matter

Figure 6. Kitt Peak 4-m prime focus image of Abell cluster A370, from Butcher, Oemler, and
Well (1983). Note the arc, which was not mentioned in the paper.

gravitationally lensed by A370 at z = 0.374. It is true, as textbooks instructed, that

lensing is optimal for a mass midway between the observer and the lensed object.
At about the same time in late 1987, Lynds and Petrosian (1988) submitted a late
paper for the AAS meeting in Austin, Texas, January 1988. They too had obtained
spectra and redshifts and announced the arc as the lensed image of a background galaxy.
Their detailed paper (Lynds and Petrosian 1989) gives a slight history leading up to this
conclusion.

6. Conclusion
By the end of the decade of the 70s, astronomers generally agreed that rotation veloc-
ities did not show a Keplerian decline toward the edge of the optical disk. An extended
spheroidal distribution of matter, roughly centered on the galaxy center, more extended
than the optical galaxy, and more massive that the luminous galaxy, was the generally
adopted solution. Today, although we think we know the quantity of luminous matter in
the universe, and can put a limit on the amount of dark matter in the universe, we have
made little progress in deducing the composition of the dark matter.
 V. C. Rubin: A brief history of dark matter 11
For a few scientists (Finzi 1963; Bekenstein & Milgrom 1984, Milgrom 1983; Sanders
1990, McGaugh and de Blok 1998), flat rotation curves have been a sufficient reason
to investigate alternatives to Newtonian gravitational theory. Rather than discuss these
alternatives which are generally as valid as dark matter in fitting the observations (al-
though gravitational lensing might stress the theories), I will close with a few historical
comments relating to scales in nature.
At the close of the 19th century, physicists were discovering phenomena that did not
fit into the science they had grown up with. Late in life, in discussing the early years
of quantum mechanics, Heisenberg (1961) wrote “Although the previous laws of nature,
e.g., Newtonian mechanics, contained so-called constants, the constants referred to the
properties of objects... On the other hand, Planck’s action quantum, which is the char-
acteristic constant in his law of radiation, does not represent a property of objects, but
a property of nature. It establishes a scale in nature.”
Heisenberg continues “While the laws of former physics, e.g., Newtonian mechanics,
should basically be valid for all orders of magnitude (the movement of the moon around
the earth should obey the same laws as the fall of the apple from the tree or the deviation
of an alpha particle that grazes the nucleus of an atom), Planck’s law of radiation shows
for the first time that there are scales in nature, that phenomena in different ranges of
magnitude are not necessarily of the same type.” It is interesting that Heisenberg’s large
scale example does not extend beyond the solar system.
Feynman, in “The Character of Physical Law” (1965) starts with the solar system,
and the problem with the apparent motion of Mercury which “was shown by Einstein
that Newton’s Laws were slightly off and then they had to be modified. The question is,
how far does this law extend? Does it extend outside the solar system?” After discussing
a binary star system and a globular cluster, he shows a “typical galaxy, and it is clear
once again that this thing is held together by some force, and the only candidate that
is reasonable is gravitation. When we get to this size we have no way of checking the
inverse square law, but there seem to be no doubt that in these great agglomerations of
stars... gravity is extending even over these distances.”
He ends by naming a characteristic that gravity shares with other physical laws. “It
is mathematical, ...it is not exact; Einstein had to modify it, and we know that it is not
quite right yet, because we have still to put the quantum theory in. That is the same
with all our other laws—they are not exact. There is always an edge of mystery, a place
where we have some fiddling around to do yet. This may or may not be a property of
Nature, but it is certainly common to all the laws as we know them today. It may be
only a lack of knowledge.”
Only the future will tell us what the dark matter is, or whether our lack of knowledge
of gravitation on the largest scales has fooled us. It will be exciting to follow the path
that leads us from this edge of mystery to answer the question, “What is dark matter?”

REFERENCES
Abt, H. A. 1999 American Astronomical Society Centennial Issue 525, Number 1C, Part 3.
Ambartsumian, V. A. 1961 AJ 66, 536.
Babcock, H. W. 1939 Lick Obs. Bull. 19, 41.
Barnothy, J. M. 1965 AJ 70, 666.
Barnothy, J. M. 1989, in Gravitational Lenses: Proceedings of a Conference Held in Honor of
Bernard F. Burke’s 60th Birthday (eds. J. M. Moran, J. N. Hewitt, & K. Y. Lo). p. 23.
Springer.
Bekenstein, J. & Milgrom, M. 1984 ApJ 286, 7.
Bosma, A. 1978 Ph.D. Thesis, University of Groningen.
12 V. C. Rubin: A brief history of dark matter
Burbidge, E. M., Burbidge, G. R., & Prendergast, K. H. 1962 ApJ 136, 128.
Butcher, H., Oemler, A., & Wells, D. C. 1983 ApJS 52, 183.
Clerke, Agnes Mary 1885, in A Popular History of Astronomy during the Nineteenth Century,
Adam and Charles Black. p. 41
de Vaucouleurs, G. 1959, in Handbuch der Physics, Astrophysik IV 53, 310.
Einasto, J., Kaasik, A., & Saar, E. 1974 Nature 250, 309.
Faber, S. M. & Gallagher, J. S. 1979 ARA&A 17, 135.
Feynman, R. 1965, in The Character of Physical Law. MIT Press.
Finzi, A. 1963 MNRAS 127, 21.
Freeman, K. C. 1970 ApJ 160, 811.
Heisenberg, W., Born, M., Schrodinger, E., & Auger, P. 1961 On Modern Physics. p. 7.
Paolo Boringhieri Editore.
Hoag, A. 1981 BAAS 13, 799.
Hubble, E. 1936, in The Realm of the Nebulae. p. 180. Yale Univ. Press.
Jeans, J. H. 1919, in Problems of Cosmogony and Stellar Dynamics. p. 239. Cambridge Univ.
Press.
Jeans, J. H. 1922 MNRAS 82, 130.
Kahn, F. D. & Woltjer, L. 1959 ApJ 130, 705.
Kalnajs, A. J. 1983, in Internal Kinematics and Dynamics of Galaxies, IAU Symposium 100
(ed. A. Athanassoula). p. 87. Reidel.
Kapteyn, J. C. 1922 ApJ 55, 302.
Katz, J. I. 1987 A&A 182, L19.
Krumm, N. & Salpeter, E. E. 1976 ApJ 208, L7.
Lynds, R. & Petrosian, V. 1986 BAAS 18, 1014.
Lynds, R. & Petrosian, V. 1988 BAAS 20, 644.
Lynds, R. & Petrosian, V. 1989 ApJ 336, 1.
McGaugh, S. S. & de Blok, W. J. G. 1998 ApJ 499, 66.
Mayall, N. U. 1951, in The Structure of the Galaxy, p. 19. University of Michigan Press.
Milgrom, M. 1983 ApJ 270, 371.
Milgrom, M. 1987 A&A 182, L21.
Mitchell, Rev. John 1784 Phil. Trans. Royal. Soc. London.
Oort, J. H. 1932 Bull. Ast. Neth. 6, 249.
Oort, J. H. 1940 ApJ 91, 273.
Ostriker, J. P. 1999, in American Astronomical Society Centennial Issue (ed. H. A. Abt) 525,
Number 1C, Part 3, p. 297.
Ostriker, J. P. & Peebles, P. J. E. 1973 ApJ 186, 467.
Ostriker, J. P., Peebles, P. J. E., & Yahill, A. 1974 ApJ 193, L1.
Rees, M. J. 1977, in The Evolution of Galaxies and Stellar Populations (eds. B. M. Tinsley &
R. B. Larson). p. 339. Yale University Observatory.
Refsdal, S. 1964 MNRAS 128, 23.
Roberts, M. A. & Rots, A. H. 1973 A&A 26, 483.
Roberts, M. S. & Whitehurst, R. N. 1975 ApJ 201, 327.
Rogstad, D. H. & Shostak, G. S. 1972 ApJ 176, 315.
Rubin, V. C., Burley, J., Kiasatpoor, A., Klock, B., Pease, G., Rutscheidt, E., &
Smith, C. 1962 AJ 67, 491 [I had to fight with the AJ editor to get the students named
joint authors].
Rubin, V. C. 1965 ApJ 142, 934.
Rubin, V. C., Burbidge, E. M., Burbidge, G. R., & Crampin, D. J. 1965 ApJ 141, 759.
Rubin, V. C. & Ford, W. K. Jr. 1970 ApJ 159, 379.
Rubin, V. C. & Ford, W. K. Jr. 1978 ApJ 225, L107.
Rubin, V. C., Peterson, C. J., & Ford, W. K. Jr. 1976 BAAS 8, 297.
Sanders, R. H. 1990 A&AR 2, 1.
Sargent, W. L. W. 1977, in The Evolution of Galaxies and Stellar Populations (eds. B. M.
Tinsley & R. B. Larson). p. 427. Yale University Observatory.
Schwarzschild, M. 1954 AJ 59, 273.
 V. C. Rubin: A brief history of dark matter 13
Smith, S. 1936 ApJ 83, 23.
Soucail, G., Fort, B., Mellier, Y., & Pical, J. P. 1987a A&A 172, L14.
Soucail, G., Mellier, Y., Fort, B., Hammer, F., & Mathez, G. 1987b A&A 184, L7.
Soucail, G., Mellier, Y., Fort, B., Mathez, G., and Cailloux, M. 1988 A&A 191, L19.
Tinsley, B. M. & Larson, R. B. (eds.) 1977 The Evolution of Galaxies and Stellar Populations,
p. 174. Yale University Observatory.
Trimble, V. 1987 ARA&A 25, 425.
Trimble, V. 1995, in Dark Matter (eds. S. S. Holt & C. L. Bennett). p. 57. AIP.
Turner, E. L. & Ostriker, J. P. 1977 ApJ 217, 24.
van de Hulst, H. C., Raimond, E., & van Woerden, H. 1957 Bull. Astro. Neth. 14, 1.
van den Bergh, S. 1961 AJ 66, 566.
van den Bergh, S. 1999 PASP 111, 657.
van der Kruit, P. C. & Allen, R. S. 1987 ARA&A 16, 103.
Volders, L. 1959 BAN 1, 323.
Whitman, Walt 1855, in Song of Myself.
Westphal, J. A. 1973 Science 181, 930.
Zwicky, F. 1933 Helv. Phys. Acta. 6, 110.
Zwicky, F. 1937a Phys. Rev. 51, 290.
Zwicky, F. 1937b Phys. Rev. 51, 679.
Zwicky, F. 1937c ApJ 86, 217.
Microlensing towards the Magellanic Clouds:
Nature of the lenses and implications on
dark matter
By K A I L A S H C. S A H U
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA;
[email protected]

A close scrutiny of the microlensing results towards the Magellanic clouds reveals that the stars
within the Magellanic clouds are major contributors as lenses, and the contribution of MACHOs
to dark matter is 0 to 5%. The principal results which lead to this conclusion are the following:
(i) Out of the ∼17 events detected so far towards the Magellanic Clouds, the lens location
has been securely determined for one binary-lens event through its caustic-crossing timescale.
In this case, the lens was found to be within the Magellanic Clouds. Although less certain, lens
locations have been determined for three other events and in each of these three events, the lens
is most likely within the Magellanic clouds.
(ii) If most of the lenses are MACHOs in the Galactic halo, the timescales would imply that
the MACHOs in the direction of the LMC have masses of the order of 0.5 M , and the MACHOs
in the direction of the SMC have masses of the order of 2 to 3 M . This is inconsistent with
even the most flattened model of the Galaxy. If, on the other hand, they are caused by stars
within the Magellanic Clouds, the masses of the stars are of the order of 0.2 M for both the
LMC as well as the SMC.
(iii) If 50% of the lenses are in binary systems similar to the stars in the solar neighborhood,
∼10% of the events are expected to show binary characteristics. The fact that the two observed
binary events are caused by lenses within the Magellanic Clouds would imply that there should
be a total of about 20 events caused by lenses within the Magellanic Clouds. This implies
that most of the microlensing events observed so far are probably caused by stars within the
Magellanic Clouds.
(iv) If the microlensing events are caused by MACHOs of 0.5 M , as claimed from the
LMC observations, about 15 events should have been detected by now towards the SMC, with
timescales of ∼40 days. The fact that no event has been detected towards the SMC caused by
MACHOs (both the events detected towards the SMC have been shown to be due to self-lensing)
places severe constraints on the MACHO contribution and suggests that the contribution of
MACHOs to dark matter is consistent with zero, with an upper limit of 5%.

1. Introduction
Most spiral galaxies are known to have flat rotation curves (e.g. Begeman et al. 1991).
In the outer parts of the galaxies in particular, the visible matter falls far short of what
is required to explain the flat rotation curves. In our own Galaxy, the rotation curve
is observed to be flat up to at least 16 kpc from the center (e.g. Fich et al. 1989).
The scenarios proposed to explain the flat rotation curves are either departure from
Newtonian dynamics in large scale, or presence of dark matter (for a review see Sanders
1990). In the latter and more conventional scenario, the rotation curves would imply
that a significant part of the mass of the galaxies, including our own, resides in the halo
in some form of dark matter. The nature of this dark matter has been hypothesized to
be in either of the two forms: MACHOs (i.e. massive compact halo objects which is a
collective term for ‘Jupiters,’ brown dwarfs, red dwarfs, white dwarfs, neutron stars or
black holes); or elementary particles such as massive neutrinos, axions or WIMPs (weakly
interacting massive particles such as photinos, etc.). One important distinguishing factor
14
 K. C. Sahu: Microlensing towards the Magellanic Clouds 15
of the MACHOs is that they can have detectable gravitational effects and can cause
“gravitational microlensing” of background stars.
Einstein (1936) was the first to point out that a star can act as a gravitational lens
for another background star, if the two are sufficiently close to each other in the line of
sight. Given the observational capabilities of that time, Einstein had however considered
this to be a purely theoretical exercise since there was “no hope of observing such a
phenomenon directly.”
Paczyński (1986) worked out the probability of such microlensing events by MACHOs
and showed that if the halo of our Galaxy is made up of MACHOs, the probability
of finding any given star being lensed by them is 5 × 10−7 , independent of their mass
distribution. He suggested an experiment to look for such events using the LMC stars.
Such an experiment was taken up by two groups who reported their first results in 1993
(Alcock et al. 1993; Aubourg et al. 1993). In six years’ monitoring of millions of stars
towards the Magellanic Clouds, so far about 17 events have been detected towards the
Large Magellanic Cloud (LMC) and two events towards the Small Magellanic Cloud
(SMC).
The observed microlensing optical depth as derived from these observations is ∼ 1 ×
10−7 . This is about a factor of 5 lower than what would be expected if the dark matter is
entirely made up of MACHOs. To complicate the issue further, the simple microlensing
light curves cannot tell us the location of the lenses because of a degeneracy between
the distance, mass and the proper motion of the lens. So the lenses can be stars in
the Galactic disk (Gould 1994; Evans et al. 2001; Gates & Gyuk 2001), MACHOs in
the Galactic halo (Alcock et al. 2000a), or stars within the LMC itself (Sahu 1994a,b;
Evans & Eamonn 2000). As a result, the interpretations of these lensing events, and their
locations in particular, have remained uncertain.

2. Overview of the problem

The contribution of known stars to the microlensing optical depth had not been cal-
culated until after the first microlensing events were reported towards the LMC. Such
calculations revealed that the microlensing optical depth due to the known stars within
the LMC is close to the observed optical depth, particularly in the region of the bar
where the microlensing event was detected (Sahu 1994a,b; Wu 1994). This was used to
argue that the stars within the Magellanic Clouds must play a significant role as lenses.
Lensing by stars within the Magellanic clouds is now commonly referred to as “self-
lensing.” It has been known for a long time (e.g. Schneider et al. 1992), and it was re-
derived by Gould (1995) that the self-lensing microlensing optical depth can be expressed
as a function of the velocity dispersion
τ = 2 × sec2 (θ)v2 /c2 , (2.1)
where θ is the inclination of the disk of the LMC (the best estimated value of which is
34.7 ± 6 deg, van der Marel and Cioni, 2001). Since the velocity dispersion of the stars
in the disk of the LMC is about 20 km s−1 , the above equation would imply that the
self-lensing optical depth is about 2 × 10−8 , which is too low to explain the observed
microlensing events.
However, there are some problems in such an approach. First, eq. (2.1) can be applied
only for a virialized system. The Magellanic Clouds are not relaxed, and far from virialized
systems: they are dynamically disturbed, they show tidal-tail structures, and the velocity
dispersions are different for low and high-mass objects (for a review, see Westerlund 1997).
Second, the lenses are low-mass objects for which the velocity dispersions are unknown.
16 K. C. Sahu: Microlensing towards the Magellanic Clouds
Third, the velocity dispersions in the region of the bar—where most of the events have
been found so far—is also unknown.
A cautionary note seems appropriate here. A similar story had repeated towards the
Galactic bulge. The velocity dispersion of the Galactic bulge stars would imply a mi-
crolensing optical depth which is a factor of at least three smaller than the observed
optical depth. In the case of the Galactic bulge, the re-discovery of the Galactic bar,
with an inclination of 15◦ to the line of sight helped in solving the puzzle of the dis-
crepancy between the observed and the expected microlensing optical depth (Kiraga &
Paczyński, 1994; Paczyński et al. 1995).
In the meantime, several papers have appeared with detailed calculations of the optical
depth taking into account the extra effects such as ‘diffusion’ of small-mass objects, which
causes the velocity dispersion of less massive objects to be larger (e.g. Salati et al. 1999).
The full range of the calculated optical depth due to the stars within the LMC, as
published in the literature so far, is 1 × 10−8 to 2 × 10−7 (e.g. Gyuk et al. 2000, Graff
2001, Salati et al. 1999, Aubourg et al. 1999). The high end of this self-lensing optical
depth would imply that all the observed events are caused by stars within the LMC
and that the contribution of MACHOs to the dark matter is negligible. The low-end of
the self-lensing optical depth would imply that none of the observed events are caused
by stars within the Magellanic clouds, and hence the events are caused most likely by
MACHOs in the halo. In the extreme case where all the events are caused by MACHOs
in the Galactic halo, the implied contribution of MACHOs to dark matter is about 25%
(Alcock et al. 2000a; Afonso et al. 2002).
Clearly, it is important to know the exact location and nature of the lenses, which can
shed some light on the whether MACHOs can account for the long-sought dark matter in
the halo. We obviously need some observational tests to guide us in resolving this puzzle.
The rest of the paper mainly deals with such tests and the current status of the results
from these tests.
It is worth pointing out here that, in the self-lensing hypothesis, it is implicit that
the sources and the lenses reside within the LMC. The sources cannot be considered
as background sources since the derivation of the observed microlensing optical depth
assumes that all the monitored stars which are within the Magellanic clouds contribute
to the microlensing optical depth.

3. Observational tests
Fortunately, there are several observational tests which can be used to infer the nature
of the lenses, including some tests which can be used for direct determinations of their
locations. The observational tests that I will discuss here, some of which have already
provided clear results, are the following:
(i) Determination of lens location through “caustic-crossing” timescale,
(ii) Other direct determinations of lens locations,
(iii) Frequency of observed binary-lens events,
(iv) Timescales of SMC vs. LMC events, and
(v) Spatial distribution of the microlensing events.

3.1. “Caustic-crossing” timescale

Clearly, the best test would be to measure the location of the lens directly. In a few special
cases, such a direct determination of the lens location is indeed possible. One such special
case is when the lens is a binary, and the source crosses the caustics produced by the
binary lens. Since a caustic is essentially a straight line in space, the time taken by the
 K. C. Sahu: Microlensing towards the Magellanic Clouds 17

Figure 1. Light curve of the PLANET data for MACHO-98-SMC-1. Shown are the data from
the SAAO 1 m (circles), the CTIO 0.9 m (squares), the CTIO-Yale 1 m (triangles), and the
Canopus 1 m (asterisks). The inset covers about 0.6 days, corresponding to less than one tick
mark on the main figure. (Taken from Albrow et al. 1999)

caustic to cross the source provides a direct measure of the proper motion of the lens
projected onto the source plane. If the lens is in the halo at a distance of ∼10 kpc, then
the expected proper motion of the lens is about 200 km s−1 which, projected onto the
source plane, is about 1000 km s−1 . So the time for the caustic to cross the source would
be of the order of half an hour. If, on the other hand, the lens is within the Magellanic
Clouds, the expected proper motion is about 50 km s−1 , and hence the caustic crossing
time is expected to be of the order of 10 hours. Thus, monitoring a caustic crossing
provides a powerful method to determine the location of the lens.

3.1.1. MACHO 98-SMC-1

Such an opportunity presented itself after the first binary-lens event, MACHO 98-
SMC-1, was discovered by the MACHO collaboration in 1998 (Alcock et al. 1999). After
the first caustic crossing was reported, the PLANET collaboration, with its 24-hour
access to telescopes around the world, predicted a second caustic crossing and began
monitoring this event, with a particular interest in fully sampling the second caustic
crossing (Albrow et al. 1999). The time for the caustic to cross the source was found to
be 8.5 hours (Fig. 1), which demonstrated that the lens is within the SMC (Fig. 2). The
result was further confirmed by the EROS, MACHO and MPS collaborations, who also
concluded that the lens is most likely within the SMC (Afonso et al. 1998, 2000, Alcock
et al. 1999, Rhie et al. 1999).

3.1.2. MACHO-LMC-9
The second binary-lens event, MACHO-LMC-9, was discovered by the MACHO col-
laboration towards the LMC, for which the observations are not as dense (Alcock et al.
18 K. C. Sahu: Microlensing towards the Magellanic Clouds
0.1

0.08

0.06

0.04

0.02

0
10 100 1000

Figure 2. Predicted velocity distributions (at the source plane) for halo and SMC lenses. The
measured value of the velocity and error bars are shown as vertical lines. The observed value is
inconsistent with the halo hypothesis. The lens and source could possess normal SMC kinematics.
(Taken from Alcock et al. 1999)

1997a). The caustic crossing timescale of the event also suggests that the lens is most
likely within the LMC (Fig. 3), although the sparse sampling of this event quality has led
to some doubt as to whether this is a microlensing event at all (Alcock et al. 2000a,b).

3.2. Other determinations of lens locations

3.2.1. MACHO 97-SMC-1
97-SMC-1 is the first microlensing event observed towards the SMC, which has a large
timescale of 220 days, in contrast to the average timescale of 40 days as observed for the
more than dozen microlensing events towards the LMC (Alcock et al. 1997b).
If the lens is a MACHO in the Galactic halo, the timescale of the event would suggest
that the mass of the lens is about 3 M . Clearly, this cannot be a normal star of 3 M since
it would be too bright and would be easily detectable. Indeed, a spectrum of the source
obtained by Sahu & Sahu (1999) shows the spectrum to be a pure B-type spectrum as
expected from a source within the SMC, without any contribution from the lens (Fig. 4).
This would suggest that the lens is either a black hole in the Galactic halo, or a small
mass object within the SMC.
A black hole in the Galactic halo is an unsatisfactory explanation since that would
imply that the MACHOs in the direction of the LMC are predominantly 0.5 solar mass
objects and the MACHOs in the direction of the SMC are predominantly 3 solar mass
objects. The locations of the LMC and the SMC too close to permit such a mass segre-
gation in these two directions even in the most extreme flattened model of the Galaxy.
 K. C. Sahu: Microlensing towards the Magellanic Clouds 19

Figure 3. Light curve of LMC-9 in the around the peak of the event, and the best-fit binary
lens light curve. The most likely lens location is within the LMC. (Taken from Alcock et al.
1997a.)

This leaves us with the only other alternative that the lens must be a small mass object
within the SMC.
Furthermore, as explained in the next section, this long timescale of the SMC events
is precisely what one would expect if the lenses are within the Magellanic Clouds, which
further supports the view that the lens is most likely within the SMC.
3.2.2. MACHO 96-LMC-2
Lens locations have been estimated for another event MACHO 96-LMC-2, where the
source is a binary (Alcock et al. 2001a). In this case too, the lens is most likely within
the LMC.
Thus, there are four cases so far for which we know the locations of the lenses. In
all the cases, without exception, the most likely locations of the lenses are within the
Magellanic Clouds.
It is worth noting here that the lens location has been recently determined for an event
caused by a star in the local Galactic disk (Alcock et al. 2001b). The events caused by
nearby lenses are expected to exhibit “parallax” effects which are easier to detect. The
fact that only one event caused by a star in the local Galactic disk has been detected is
consistent with the expected frequency of nearby stars acting as lenses.
3.3. Frequency of observed binary-lens events
For the sake of argument, let us start with the assumption that all the events except the
two binary events for which we know that the lenses are within the Magellanic clouds
are caused by MACHOs in the halo. This would be quite extra-ordinary since this would
20 K. C. Sahu: Microlensing towards the Magellanic Clouds

Figure 4. Observed spectrum of MACHO 97-SMC-1 taken on 30th May 1997, just after the
source had crossed the Einstein ring radius of the lens (solid curve), along with the best fit stellar
model spectrum (dashed line). The contribution from the lens is negligible which, combined with
the microlensing time scale, implies that the lens must be within the SMC. (Taken from Sahu
& Sahu 1998).

mean that the only two events that are caused by the stars within the Magellanic clouds
not only happen to be binaries, but they also show caustic crossing structure in their
light curves. If one event caused by a star in the Magellanic clouds is observed, the
probability that the lens would be a binary is about 50% (assuming 50% of the lenses
are in binary systems similar to the stars in the solar neighborhood). Since only 20% of
the binary-lens events are expected to show caustic-crossing structures (which is borne
out by the observations towards the Galactic bulge as well), the probability that it would
show a caustic-crossing structure is 10%. The probability that a second event caused by
a Magellanic cloud star would again be a binary and show caustic structure is again 10%.
The probability that two successive events caused by Magellanic cloud stars will both be
binaries and show caustic crossing structures is thus 1%, and hence extremely unlikely.
We can follow the same argument in reverse and ask the following question: if two events
which show binary characteristics are within the Magellanic clouds, how many of the
observed events should be caused by stars within the Magellanic clouds? Since 10% of
the events are expected to show binary characteristics, one would expect about 20 events
to be caused by stars within the Magellanic clouds. This is more or less the number of
events observed, which suggests that most of the lenses must be within the Magellanic
clouds.

3.4. Timescales of SMC vs. LMC events

The relative timescales of the LMC and the SMC events provide further indications on the
locations of the lenses. If the lenses are MACHOs in the Galactic halo, the characteristics
of the lenses towards the LMC and the SMC are expected to be the same and hence both
the LMC and the SMC events are expected to have similar timescales. On the other
hand, if the lenses are within the Magellanic Clouds, the expected timescales of the LMC
and the SMC events are expected to be very different. The LMC has a depth of less than
one kpc in a typical line of sight whereas the SMC as a much larger depth of about five
kpc along a typical line of sight. So the typical distance between the lens and the source
is much smaller in the case of LMC than in the case of the SMC. As a result, the size
 K. C. Sahu: Microlensing towards the Magellanic Clouds 21

Line of No. of Mlens (halo) Mlens (LMC/SMC) Mlens (local disk)

Sight Events (M ) (M ) (M )
LMC ∼17 ∼0.5 ∼0.2 ∼0.5
SMC ∼ 2 >2 ∼0.2 ∼2
Table 1. Lens masses for different scenarios

of the Einstein ring (which scales as the square root of the distance between the source
and the lens) is much smaller for the LMC events than for the SMC events. Since the
velocity dispersions of the stars within the LMC and the SMC are similar, the timescales
of the SMC events are expected to be much longer than the LMC events, if the lenses
are within the Magellanic Clouds.
Two events have been observed towards the SMC so far which have timescales of 75
and 125 days. This would correspond to lens masses of 2 to 3 solar masses if the lenses
are in the halo, and about 0.2 to 0.3 solar masses if the lenses are within the Magellanic
clouds. For the ∼17 events observed towards the LMC, the timescales are much shorter,
which would correspond to lens masses of ∼0.5 solar masses if the lenses are in the halo,
and ∼0.2 solar masses if the lenses are within the LMC. This is shown it tabular form
in Table 1. If the events are predominantly caused by MACHOs in the halo, one faces
the unrealistic consequence that the MACHOs in the direction of the LMC and the SMC
are of very different mass. Self-lensing is the only scenario which gives consistent masses
for both the LMC and the SMC events. The longer durations of the SMC events are a
natural consequence of the self-lensing hypothesis.

3.5. Spatial distribution of the events

Finally, the spatial distribution of the events provide some extra insight into the nature
of the lenses. As first pointed out by Sahu (1994 a,b), the events should be more concen-
trated towards the bar of the LMC if the lensing is caused by the stars within the LMC.
If the events are caused by MACHOs then the events (for a given number of monitored
stars) should be uniformly distributed over the whole of the LMC. Unfortunately, all the
analyses by the MACHO group so far have been confined to the region around the bar of
the LMC. The EROS group, however, have mostly monitored in the region outside the
bar, and have not detected any event. From their non-detection of microlensing events,
they derive a microlensing optical depth which is much smaller than the optical depth
determined by the MACHO group. This is consistent with the fact that stars within the
Magellanic clouds play a dominant role as lenses.
Monitoring some regions far from the bar of the LMC to look for microlensing events
would provide a clear test on whether the lensing events are caused by MACHOs or stars
within the LMC (e.g. Stubbs 1999).

4. Implications on dark matter

The most significant insight into the contribution of dark matter, in my view, comes
from a comparative study of the number of observed events towards the LMC and the
SMC, and their timescales. So far, two microlensing events have been detected towards
the SMC with time scales of 75 and 125 days. From these two events, the optical depth
towards the SMC has been estimated to be ∼ 2 × 10−7 , which is about the same as the
optical depth towards the LMC (Alcock et al. 1999). In terms of optical depth, these
22 K. C. Sahu: Microlensing towards the Magellanic Clouds
two events are equivalent to about 15 events with timescales of ∼40 days (optical depth
scales as the square of the timescale). Thus, if the microlensing events are caused by
0.5 solar mass MACHOs as claimed from the LMC observations, about 15 such events
should have been detected towards the SMC by now. Furthermore, as discussed earlier,
both of the observed SMC events have been shown to be caused by stars within the
SMC. Thus no microlensing event caused by a MACHO has been detected towards the
SMC, whereas 15 such events should have been detected if the contribution of MACHOs
to dark matter is indeed 20% as claimed from the detection of ∼20 microlensing events
towards the LMC. This leads us to the inevitable conclusion that the contribution of
MACHOs to dark matter is less than 2%, with a strong upper limit of 5%.

5. Conclusions
Results from several observational tests are already available which can be used to
clearly distinguish between different scenarios for the nature of the lenses. I have argued
that the results obtained so far point to the fact that the observed microlensing events
towards the Magellanic clouds are predominantly caused by the stars within the Magel-
lanic Clouds. Consequently, the contribution of MACHOs to the dark matter is consistent
with zero, with a strong upper limit of 5%. However, some estimates suggest that the
known stars within the Magellanic Clouds fall short by a factor of a few to explain the
observed microlensing events. The contribution to the microlensing optical depth by the
known stars is, however, very uncertain, and further work would be needed to resolve
this issue.

REFERENCES
Afonso, C., et al. 1998 A&A 337, L17.
Afonso, C., et al. 2000 ApJ 532, 340.
Afonso, C., et al. (EROS Collaboration) 2002 A&A, 400, 951; astro-ph/0212176.
Albrow, M., et al. 1999 ApJ 512, 672.
Alcock, C., et al. 1993 Nature 365, 621.
Alcock, C., et al. 1997a ApJ 491, L11.
Alcock, C., et al. 1997b ApJ 486, 697.
Alcock, C., et al. 1999 ApJ 518, 44.
Alcock, C., et al. 2000a ApJ 542, 281.
Alcock, C., et al. 2000b ApJ 541, 270.
Alcock, C., et al. 2001a ApJ 552, 259.
Alcock, C., et al. 2001b Nature 414, 617.
Aubourg, E., et al. 1993 Nature 365, 623.
Aubourg, E., et al. 1999 A&A 347, 850.
Begeman, K. G., Broeils, A. H., & Sanders, R. H. 1991 MNRAS 249, 523.
Einstein, A. 1936 Science 84, 506.
Evans, E. N. & Eamonn, K. 2000 ApJ 529, 917.
Evans, N. W., Gyuk, G., Turner, M. S., & Binney, J. 1998 ApJ 501, L45.
Fich, M., Blitz, L., & Stark, A. 1989 ApJ 342, 272.
Gates, E. L. & Gyuk, G. 2001 ApJ 547, 786.
Gould, A. 1994 ApJ 421, L71.
Gould, A. 1995 ApJ 441, 77.
Graff, D. 2001. In Microlensing 2000: A New Era of Microlensing Astrophysics (eds. J. W.
Menzies & Penny D. Sackett). ASP Conference Proceedings, vol. 239, p. 73. ASP.
Gyuk, G., Dalal, N., & Griest, K. 2000 ApJ 535, 90.
Kiraga, M. & Pacyński, B. 1994 ApJ 430, L101.
 K. C. Sahu: Microlensing towards the Magellanic Clouds 23
Paczyński, B. 1986 ApJ 304, 1.
Paczyński, B., Stanek, K. Z., Udalski, A., Szymanski, M., Kaluzny, J., Kubiak, M.,
Mateo, M., & Krzeminski, W. 1994 ApJ 435, L113.
Palanque-Delabrouille, N., et al. (EROS collaboration) 1997 A&A 332, 1.
Palanque-Delabrouille, N. 2001 New Astron. Rev. 45, 395.
Rhie, S. H., et al. 1999 ApJ 522, 1037.
Sahu, K. C. 1994a Nature 370, 275.
Sahu, K. C. 1994b PASP 106, 942.
Sahu, K. C. & Sahu, M. S. 1998 ApJ 508, 147.
Salati, P., et al. 1999 A&A 350, L57.
Sanders, R. H. 1990 A&A 2, 1.
Schneider, P., Ehlers, J., & Falco, E. E. 1992 Gravitational Lensing. Springer-Verlag.
Stubbs, C. 1999. In The Third Stromlo Symposium: The Galactic Halo (eds. B. K. Gibson,
T. S. Axelrod, & M. E. Putman). ASP Conference Series, vol. 165, p. 503. ASP.
van der Marel, R. & Cioni, M. L. 2001 AJ 122, 1807.
Westerlund, B. E. 1997 The Magellanic Clouds. Cambridge University Press.
Wu, X-P. 1994 ApJ 435, 66.
 Searching for the Galactic dark matter
By H A R V E Y B. R I C H E R
Department of Physics & Astronomy, University of British Columbia, 6224 Agricultural Road,
Vancouver, B.C., V6T 1Z1, Canada; [email protected]

A straightforward interpretation of the MACHO microlensing results in the direction of the

Magellanic Clouds suggests that an important fraction of the baryonic dark matter component
of our Galaxy is in the form of old white dwarfs. If correct, this has serious implications for the
early generations of stars that formed in the Universe and also on the manner in which galaxies
formed and enriched themselves in heavy elements. I examine this scenario in some detail and
in particular explore whether the searches currently being carried out to locate local examples
of these MACHOs can shed any light at all on this scenario.

1. Introduction
A conservative estimate of the mass of the Galaxy out to a distance of about 2/3
of that of the Large Magellanic Cloud is MG = 4 × 1011 M (Fich & Tremaine 1991).
With a total luminosity in the V-band of 1.4 × 1010 L (Binney & Tremaine 1987) the
Galactic mass to light ratio in V (M/LV ) out to 35 kpc is ∼ 30. Since normal stellar
populations do not generally produce M/LV ratios higher than about 3, this is usually
taken as evidence for an important component of dark matter within an extended halo
surrounding the Galaxy.
A related result is that of the MACHO microlensing experiment in the direction of the
Magellanic Clouds which seems to indicate that about 20% (10–50% is the 90% CI) of
the dark matter in the Galaxy is tied up in objects with masses near 0.5 M (0.3–0.9
is the 90% CI; Alcock 2000). This assertion is only true if the bulk of the MACHOs are
located in the halo of our Galaxy. Moreover, recent studies suggest that objects with
masses from 10−7 to 0.02 M are now largely excluded (so planets and/or brown dwarfs
are not important dark matter candidates) and that a 100% MACHO halo is no longer
a viable model (Alcock 2000).
MACHO masses near 0.5 M are suggestive of white dwarfs although neutron stars
or even primordial black holes remain as unlikely but still possible candidates. A halo
currently consisting of 20% old white dwarfs implies that the precursors of these 0.5 M
objects would have accounted for ∼ 40% of Ωbaryon in the Universe. This estimate comes
from taking 2 M as the precursor masses, M/LV = 1500 for Ωcritical , 0.04 for Ωbaryon
and assuming that the Galactic halo is a fair sample of the types of mass seen in the
Universe. It may be that Galactic halos are baryon-rich in which case the derivation of
40% would be an over-estimate. This may also be an interesting way to constrain the
masses of the precursors as clearly in this analysis they could not exceed ∼ 5 M without
violating the observed baryonic content of the Universe. Even given all the uncertainties
in these estimates, it is clear in this scenario that an important fraction of all the baryons
in the Universe were funneled through this star formation mode. This will surely have
implications for star formation scenarios in the early Universe and perhaps also on the
manner in which galaxies were assembled and how they enriched themselves in heavy
elements.
The interpretation of the microlensing results is, however, not without controversy as
there are indications that at least some of the lenses may reside in the Magellanic Clouds
themselves (McGrath & Sahu 2000; Sahu 2002). This would have the effect of lowering
24
 H. B. Richer: Searching for the Galactic dark matter 25

Figure 1. The spectrum of a 2500 K hydrogen-rich white dwarf from Hansen (1999, private
communication). The depletion of flux in the near IR has been seen now in a dozen or more
old white dwarfs. No known object, however, exhibits significant H2 bands as illustrated in this
model spectrum.

both the lens masses (to 0.2–0.3 M ) and the contribution of the MACHOs to the
Galactic dark matter budget. The lower lens masses would be in line with expectations
from mass function considerations and the lower MACHO dark matter contribution would
alleviate such problems as the appearance of high redshift galaxies (at an epoch when
the white dwarf precursors were luminous), the chemical evolution of galaxies (the white
dwarf precursors producing too much helium and heavier elements) and the peculiar
mass function required of the precursors (a mass function truncated at both high and
low masses to avoid too much chemical enrichment from high mass stars and too many
halo white dwarfs currently evolving from the low mass ones; Chabrier 1999). There is
also the possibility that most of the MACHOs are located in a thick (Gates et al. 1998)
or flaring disk (Evans et al. 1998) of the Galaxy. This would have the effect of reducing
their scale height and again their total Galactic mass contribution.
The definitive answer as to whether old white dwarfs are important baryonic dark
matter contributors will come from searches which will (or will not) identify significant
numbers of high velocity local examples of the very cool white dwarfs that produce the
microlensing. Such searches are now underway and in the ensuing sections I will discuss
early results from them.

2. New models of very cool white dwarfs

Just recently, a new era opened in the study of old white dwarfs. Hansen (1998,
1999) first calculated emergent spectra from cool (Teff < 4000 K) white dwarfs with
atmospheres that included opacity from the H2 molecule. Similar models were also con-
structed by Saumon and Jacobson (1999) and earlier ones by Bergeron, Saumon & We-
semael (1995) for objects down to 4000K. All these models exhibited very strong col-
lisionally induced opacity in the red and near IR part of the spectrum and the effects
of this opacity can be seen clearly in the model spectrum shown in Figure 1 as well in
26 H. B. Richer: Searching for the Galactic dark matter

Figure 2. Cooling track of a 0.5 M white dwarf (Hansen 1999). The age of the white dwarf
in Gyrs is indicated along the plot. This age does not include the main sequence lifetime of the
white dwarf precursor.

several dozen real objects (see e.g. Oppenheimer et al. 2001b). The broad bands seen in
the spectrum in Figure 1 are due to H2 but it is of interest to point out that no cool
white dwarf has yet been observed that actually shows these spectral features. Perhaps it
is because no objects this cool have been discovered up until now, or that the models are
still somehow deficient, or that possibly the Universe is not old enough for hydrogen-rich
white dwarfs to have reached such low temperatures. The effects of collisionally induced
absorption are not limited, however, to extremely cool objects. Its influence can been
seen in white dwarfs at least as hot as 5400 K (see, e.g. the spectrum of LHS 1126 in
Bergeron et al. 1994).
The effect of the H2 opacity is to force the radiation out in the bluer spectral re-
gions. This has an enormous influence on their colors, the location of the white dwarf
cooling track in the color-magnitude diagram, and hence their observability. Down to a
temperature of about 4000 K (which corresponds to an age of about 7 Gyr for 0.5 M
white dwarfs) these stars become increasingly redder as they cool, the reddest color they
achieve is about V − I  1.2 in Hansen’s (1998, 1999) models. Note that other sets of
cooling models give somewhat different colors and ages here (e.g. Fontaine et al. 2001).
Older white dwarfs become progressively bluer in V − I as they continue to cool. An-
cient white dwarfs, of age 12 Gyr, and mass 0.5 M , have V − I ∼ −0.3 according to
these models. As has been noted several times in the past few years, old hydrogen-rich
white dwarfs are blue, not red, in V − I colors. Searches for such ancient objects are now
concentrating on bluish objects in these colors.
A cooling track from Hansen’s (1998, 1999) models illustrating all these features for
0.5 M white dwarfs is shown in Figure 2. Note that instead of getting arbitrarily faint
with age, old hydrogen-rich white dwarfs seem eventually to coast at approximately
constant MV for ages between about 10 and 14 Gyr, just the timescales which may
bracket globular cluster ages. The implication is clear—very old white dwarfs may be
easier to detect than was previously thought. Also potentially of interest is that the V − I
 H. B. Richer: Searching for the Galactic dark matter 27

Figure 3. Deep HST color magnitude diagram of the globular cluster M4 with a cooling track
for a 0.5 M white dwarf superimposed. The oldest white dwarfs detected are only about 7 Gyr
and while these should show some effects due to collisionally induced absorption, they are too
warm to exhibit the extreme effect of a bluing in the V − I color with decreasing temperature.

color now moves the bulk of the ancient white dwarfs away from the swarm of field K
and M-dwarfs (making confusion with such objects less likely) but at the same time
overlapping the colors of distant star-forming galaxies (making the confusion with such
objects more likely). The solution here will be to search for old white dwarfs using both
color and proper motion criteria, an approach taken by most current programs.
One problem with these models is that they are largely untested. Of serious concern
is that none of the cool white dwarfs found to date (see following sections) seem to
exhibit the extreme blue colors predicted by most of these current models containing
H2 in their atmospheres. This may be suggesting missing or inadequate physics. One
place where these models could potentially be tested is among the cool white dwarfs in
globular clusters. Such systems should contain white dwarfs almost as old as the clusters
themselves (likely in the 12–14 Gyr range), so the extreme effects of collisionally induced
opacity ought to be clearly seen. Figure 3 shows the current situation here. It displays the
deepest color magnitude diagram yet obtained for a globular cluster (Richer et al. 1995,
1997), an HST study of M4. White dwarfs perhaps as faint as MV = 16 are recorded
in this study. If these objects have masses near 0.5 M , they will have ages in excess
of 7 Gyr and Teff ∼ 4000 K in Hansen’s (1998, 1999) models; cool enough to exhibit
some of the effects of collisionally induced opacity but not old enough to help constrain
the appearance of truly ancient white dwarfs. In HST cycle 9, a group of colleagues and
I were awarded time on HST to attempt to locate the termination point of the white
dwarf cooling sequence in M4. With the current generation of models and a cluster age
in the range of 12–14 Gyr, this should occur near a V magnitude of about 30, certainly a
enormous challenge. It is not clear whether the data will actually go this deep, but even
if it does not, we should be in a position to provide some stringent tests of the current
generation of models.
In Figure 4 I attach a simulation of what we expect to see with these data. This is an
optimistic simulation in that it does not include the effects of charge transfer inefficiency
28 H. B. Richer: Searching for the Galactic dark matter

Figure 4. Simulation of deep HST data in the globular cluster M4. The WFPC2 exposure time
calculator was used to determine the photometric uncertainty as a function of apparent magni-
tude for the white dwarfs in the globular cluster M4 assuming ages of 10 and 14 Gyr. A suite
of Monte Carlo realizations, incorporating a proper treatment of the photometric uncertainties
involved and the higher background expected from scattered light of the bright giants, were then
undertaken. One such realization is shown for a cluster with each of these ages.

now known to be plaguing the HST WFPC2 CCDs and also assumes that all the data are
taken under low sky brightness conditions. Nevertheless, even if the faintest magnitudes
are not reached, the effects of collisionally induced opacity should be readily apparent
and quantitatively testable with these data.

3. Current Searches for Very Cool White Dwarfs

A number of searches for an ancient population of halo white dwarfs are currently un-
derway or have been completed. Most of these involve color and proper motion selection
with some follow-up spectroscopy. The proper motions have generally been derived from
digitized photographic plates with quite long time baselines (typically 20–50 years). I list
below in Table 1 a compilation of these surveys together with some additional informa-
tion. This table must be considered a rough guide only to the current situation. However,
it is indicative of the present state of our knowledge.
In Table 1 the first two columns give the name and the area searched for each program.
The third column is the limiting V magnitude of each survey. This is a fairly crude
estimate as some (e.g. Oppenheimer et al. 2001a) did not survey in the V-band and it
was necessary to transform their particular magnitude to V so that a homogeneous set
of statistics for each cluster could be derived. Column 4 is an estimate of the distance
probed in each survey for old white dwarfs. It is the distance out to which ancient white
dwarfs could have been detected in the individual surveys. This limiting distance for
detection of old white dwarfs was set to 10(10([Vlim −17]/5) ) pc; 17 was taken as the MV
of a typical old white dwarf (Hansen 1998, 1999; Saumon and Jacobsen 1999; Richer
et al. 2000). Column 5 lists the number of thick disk and spheroid white dwarfs that are
expected out to the limiting distance of the survey. This number was derived following
the prescription of Reid et al. (2001) and is dominated by the thick disk contribution.
Thin disk white dwarfs are not considered in this discussion as they would be eliminated
by their low velocities. We are only concerned here with high velocity objects which
could contaminate a true dark halo sample. Column 6 contains the number of observable
white dwarfs expected from the dark halo (Richer et al. 2000) assuming that the halo
consists of 20% by mass of MACHOs which are old 0.5 M white dwarfs, half of which
have hydrogen-rich outer atmospheres. The helium-rich ancient white dwarfs would have
 H. B. Richer: Searching for the Galactic dark matter 29

Survey Area (Deg2 ) Vlim Dmax (pc) No. Disk + Sph. No. Dark Halo No. Found
Opp† 4165 19.8 36 3.9 3.7 4
Monet‡ 1378 20 40 2.5 2.4 >1
Ibata¶ 790 20 40 1.4 1.4 2
SDSS 400 20.5 50 1.4 3.5 2
Jong†† 2.5 24 250 1.1 3.6 3
CFHT‡‡ 15 25 400 27 88 ?
HDF¶¶ 1.4 × 10−3 28 1600 0.2 0.5 0?
† Oppenheimer et al. 2001a, Science Online, March 22
‡ Monet et al. 2000, AJ 120, 1541
¶ Ibata et al. 2000, ApJ 532, L41
 Harris et al. 2001, ApJ 549, L109
†† de Jong et al. 2000 astro-ph/0009058
‡‡ Stetson et al. in progress
¶¶ Ibata et al. 1999, ApJ 524, L95
Table 1. Predicted and observed numbers of faint white dwarfs

cooled to very low luminosities by a Hubble Time and would be basically unobservable.
The seventh column lists the number of old white dwarfs actually found out to the
limiting distance only in each survey.
It is clear from examination of Table 1 that in all existing surveys thus far, there is no
obvious excess in the discovered number of old white dwarfs over that expected from the
thick disk and spheroid. Note, however, that the CFHT program, currently acquiring its
second epoch of data, should be extremely important in this context. It is quite different
from most of the other surveys as it is much deeper (Vlim ∼ 25). One caveat must be kept
in mind when perusing Table 1, however. In comparing, for example, columns 5 through
7 the assumption of 100% recovery of all moving objects is built in. If, for example, a
survey was only 50% complete in recovering moving objects, then the number in column 7
should be increased by a factor of 2. For some surveys, such as the SDSS, where proper
motions are not part of the selection criteria, the numbers in Table 1 are strictly correct.
However, until estimates for proper motion limits are accurately known, there remains no
clear evidence of a white dwarf dark matter halo signature in any of the current surveys.
Of course, at the moment, it is also true that the surveys do not rule out such a dark
matter halo either.

4. Temperature and age distribution of dark halo white dwarf

candidates
We can examine the totality of objects in the surveys of Table 1 from a different
perspective. If they are, in fact, local examples of the MACHOs, and thus important
Galactic dark matter contributors, they should be quite old, likely as old as the globular
clusters. This would place them in the 12–14 Gyr range if we take the current spread
quoted for globular cluster ages. From the location of the white dwarfs in color-magnitude
and color-color diagrams, temperatures can be inferred with the use of theoretical models.
From these temperatures, ages can then be assigned using models of cooling white dwarfs
after choosing an appropriate mass.
For the purposes of this exercise, I used the models of Hansen (1998, 1999) and those
discussed in Oppenheimer et al. (2001a) with white dwarf masses of 0.5 M , to estimate
the temperature and cooling time of each object found in the various surveys. Note that
30 H. B. Richer: Searching for the Galactic dark matter

Figure 5. Cumulative distribution of the temperatures of all the white dwarfs with published
colors from the surveys listed in Table 1. Here I have selected not just white dwarfs out to Dmax
for each survey, but the complete sample. The time for a 0.5 M white dwarf to cool to a given
temperature is indicated. The age distribution seems to come from a sample of objects that is
significantly younger than the halo.

we are not limiting ourselves to objects within Dmax here. The cumulative distribution
of temperatures is shown in Figure 5 and the time for a 0.5 M white dwarf to cool to a
given temperature is also indicated by the vertical lines. The immediate conclusion from
the plot is that the sample in Table 1 is not that of a halo sample, the objects are simply
too young. It seems more in line with that of a thick disk component which dominates
the predicted numbers in Table 1. This result is, however, strongly model dependent. A
different set of cooling models or mass would provide quite a different result. For example,
if I had chosen to assign a mass of 0.6 M to all the stars, the coolest object would have
a cooling age of 11.7 Gyr, almost 1 Gyr older than a 0.5 M white dwarf. Further, the
main sequence lifetimes of the white dwarf precursors are not included in the cooling
ages. If the masses of the precursors are in the range of 2 M this will add about 1 Gyr
to these numbers. Nevertheless, the general conclusion to be drawn from both Table 1
and Figure 5 is that within the context of the present generation of white dwarf cooling
models, truly ancient white dwarfs do not currently appear to have been discovered.

5. Moving objects in the Hubble Deep Field

The last row in Table 1 deserves some special comments. Based on two epochs of
exposures taken in the HDF separated by 2 years, Ibata et al. (1999) claimed to detect
two extremely faint objects with significant proper motion. One of these appeared to have
colors consistent with that expected from a very cool white dwarf. With lower statistical
significance, three other objects also appeared to be moving, one of which possessed
colors expected from an old white dwarf. The inferred space velocities for these objects
placed them squarely in the category of halo objects. This result was potentially very
important as the detection of even a few such objects in the HDF would be enough to
account for the entire MACHO contribution to the Galactic dark matter. The second
 H. B. Richer: Searching for the Galactic dark matter 31
epoch data were, however, taken in the non-optimal F814W filter for another program
(supernova search), and were quite noisy.
Based on this seemingly important result, a third epoch was secured in F606W, fully
five years after the original HDF data. These images were of significantly higher quality.
One of the objects suggested by Ibata et al. (1999) as possibly moving was number 4-141
in the Williams et al. (1996) compilation. Using a maximum likelihood technique, Ibata
et al. (1999) measured a motion of 35 mas/yr with an error of 8 mas/yr. With the current
superior data set we measure a proper motion of 10 ± 4 mas/yr, a proper motion which is
not statistically significant. None of the other 4 objects was found to possess significant
motion over the 5-year baseline.
A full discussion of our reanalysis of moving objects in the HDF will appear in Richer
et al. 2002.

6. A final comment
My personal view of the current landscape regarding the viability of old white dwarfs
as a component of the Galactic dark matter is that while there are some interesting hints,
there is no compelling evidence in the current statistics either for against the scenario.
Perhaps a listing of the critical issues will be instructive.
(a) Where are the lenses in the direction of the LMC and SMC located? This is a
fundamental issue—perhaps the most important one. There are suggestions that some,
perhaps even most, of the lenses are in the Clouds themselves. This issue may be settled
in a number of ways, some of which are possible with current technology. One road to
a solution is to observe along a different line of sight—a microlensing campaign in the
direction of M31 or M33 would be extremely useful here. There are programs currently
underway looking at M31 (e.g. Uglesich et al. 1999) but no comprehensive results are
available at this time. Some future satellite missions may be capable of detecting par-
allaxes (and masses) for the lenses (Gould and Salim 1999) and this would be critical
data for establishing their location. If the lenses are in the Clouds and are low mass main
sequence stars, then it may eventually be possible to observe them directly.
(b) Is there a population of cool white dwarfs with halo kinematics that are present
in numbers which exceed those expected from the thick disk and spheroid? As I have
discussed, based solely on number count arguments, the current searches do not as yet
provide a conclusive answer either in the affirmative or negative. Deep (V ∼ 25) proper
motion and color selected surveys covering 10 degrees or more of sky are required to be
definitive. There are several such searches underway and the results are eagerly antici-
pated.
(c) Do we have the last word on the models of old white dwarfs? There are two aspects
of the present theory of cooling white dwarfs which I find disturbing. First, there are
enormous problems in fitting the spectra of these cool white dwarfs with the current
generation of models. The situation is so dire that Bergeron (2001), one of the premier
practitioners of this art, was required to insert a fictitious opacity source into his models
to help in the fits. He then went on to state: “It now remains to be seen whether this new
opacity source, even when included with the approximate treatment introduced in this
paper, can help resolve the mystery surrounding several ultracool white dwarfs whose
energy distributions have yet failed to be successfully explained in terms of hydrogen
or mixed hydrogen/helium compositions.” The second concern is simply that no truly
ultracool white dwarfs have as yet been discovered, or, perhaps better put, no white
dwarfs are known whose colors and/or spectra suggest that the object’s age is in the
range of 12–14 Gyrs. This may be the same problem as the first one mentioned here,
32 H. B. Richer: Searching for the Galactic dark matter
that is that the models are still somehow inadequate, or it may be suggesting that such
objects are extremely rare or just do not exist. It is within this context that extremely
deep observations of globular cluster white dwarfs will be important.

The author would like to thank the Space Telescope Science Institute for organizing
a meeting that was all that a scientific should be; exciting new results, participants
with deep and wide interests, an excellent venue. In particular, deepest gratitude is
due to Mario Livio who organized the meeting and provided one of the best conference
summaries that this author has ever experienced.
The research of HBR is supported by grants from the Natural Sciences and Engineering
Research Council of Canada.

REFERENCES
Alcock, C. 2000 Science 287, 74.
Bergeron, P. 2001 ApJ 558, 369; astro-ph/0105333.
Bergeron, P., Ruiz, Maria-Teresa, Leggett, S. K., Saumon, D., & Wesemael, F. 1994
ApJ 423, 456.
Bergeron, P. Saumon, D., & Wesemael, F. 1995 ApJ 443, 764.
Binney, J. & Tremaine, S. 1987 Galactics Dynamics, Princeton Series in Astrophysics, Prince-
ton University Press.
Chabrier, G. 1999 ApJ 513, L103.
de Jong, J., Kuijken, K., & Neeser, M. 2000; astro-ph/0009058.
Evans, N. W., Gyuk, G., Turner, M. S., & Binney, J. 1998 Apj 501, L45.
Fich, M. & Tremaine, S. 1991 ARAA 29, 409.
Fontaine, G., Brassard, P., & Bergeron, P. 2001 PASP 113, 409.
Gates, E. I., Gyuk, G., Holder, G.P., & Turner, M.S. 1998 ApJ 500, L145.
Gould, A. & Salim, S. 1999 ApJ 524, 794.
Hansen, B. M. S. 1998 Nature 394, 860.
Hansen, B. M. S. 1999 ApJ 520, 680.
Harris, H. C., Hansen, B. M. S., Liebert, J., Vanden Berk, D. E., Anderson, S. F.,
Knapp, G. R., Fan, X., Margon, B., Munn, J. A., Nichol, R. C., Pier, J. R.,
Schneider, D. P., Smith, J. A., Winget, D. E., York, D. G., Anderson, J. E., Jr.,
Brinkmann, J., Burles, S., Chen, B., Connolly, A. J., Csabai, I., Frieman, J. A.,
Gunn, J. E., Hennessy, G. S., Hindsley, R. B., Ivezic, Z., Kent, S., Lamb, D. Q.,
Lupton, R. H., Newberg, H. J., Schlegel, D. J., Smee, S., Strauss, M. A., Thakar,
A. R., Uomoto, A., & Yanny, B. 2001 ApJ 549, L109.
Ibata, R., Irwin, M., Bienayme, O., Scholz, R., & Guibert, J. 2000 ApJ 532, L41.
Ibata, R., Richer, H. B., Gilliland, R. L., & Scott, D. 1999 ApJ 524, L95.
McGrath, E. J. & Sahu, K. C. 2000 AAS Meeting 197, #105.09.
Monet, D. G., Fisher, M. D., Liebert, J., Canzian, B., Harris, H. C., & Reid, I. N.
2000 AJ 120, 1541.
Oppenheimer, B. R., Hambly, N. C., Digby, A. P., & Saumon, D. 2001a Science 292, 698.
Oppenheimer, B. R., Saumon, D., Hodgkin, S. T., Jameson, R. F., Hambly, N. C.,
Chabrier, G., Filippenko, A. V., Coil, A. L., & Brown, M. E. 2001b ApJ 550, 448.
Reid, I. N., Sahu, K. C., & Hawley, S. L. 2001 ApJ, 559, 942; astro-ph/0104110.
Richer, H. B., et al. 2002, in preparation.
Richer, H. B., Fahlman, G. G., Ibata, R. A., Pryor, C., Bell, R. A., Bolte, M., Bond,
H. E., Harris, W. E., Hesser, J. E., Holland, S., Ivanans, N., Mandushev, G.,
Stetson, P. B., & Wood, M. A. 1997 ApJ 484, 741.
Richer, H. B., Fahlman, G. G., Ibata, R. A., Stetson, P. B., Bell, R. A., Bolte, M.,
Bond, H. E., Harris, W. E., Hesser, J. E., Mandushev, G., Pryor, C. & Vanden-
berg, D. A. 1995 ApJ 451, L17.
Richer, H. B., Hansen, B., Limongi, M., Chieffi, A., Straniero, O., & Fahlman, G. G.
2000 ApJ 529, 318.
 H. B. Richer: Searching for the Galactic dark matter 33
Sahu, K. C. 2002; see paper in this volume.
Saumon, D. & Jacobson, S. 1999 ApJ 511, L107.
Uglesich, R., Crotts, A. P. S., Tomaney, A. B., & Gyuk, G. 1999 AAS Meeting 194,
#90.04
Williams, R. E., Blacker, B., Dickinson, M., Dixon, W., Ferguson, H. C., Fruchter,
A. S., Giavalisco, M., Gilliland, R. L., Heyer, I., Katsanis, R., Levay, Z., Lucas,
R. A., McElroy, D. B., Petro, L., Postman, M., Adorf, H-M., & Hook, R. 1996
AJ 112, 1335.
Another Random Scribd Document
with Unrelated Content
Project Gutenberg™ is synonymous with the free distribution of
electronic works in formats readable by the widest variety of
computers including obsolete, old, middle-aged and new
computers. It exists because of the efforts of hundreds of
volunteers and donations from people in all walks of life.

Volunteers and financial support to provide volunteers with the

assistance they need are critical to reaching Project
Gutenberg™’s goals and ensuring that the Project Gutenberg™
collection will remain freely available for generations to come. In
2001, the Project Gutenberg Literary Archive Foundation was
created to provide a secure and permanent future for Project
Gutenberg™ and future generations. To learn more about the
Project Gutenberg Literary Archive Foundation and how your
efforts and donations can help, see Sections 3 and 4 and the
Foundation information page at www.gutenberg.org.

Section 3. Information about the Project

Gutenberg Literary Archive Foundation
The Project Gutenberg Literary Archive Foundation is a non-
profit 501(c)(3) educational corporation organized under the
laws of the state of Mississippi and granted tax exempt status
by the Internal Revenue Service. The Foundation’s EIN or
federal tax identification number is 64-6221541. Contributions
to the Project Gutenberg Literary Archive Foundation are tax
deductible to the full extent permitted by U.S. federal laws and
your state’s laws.

The Foundation’s business office is located at 809 North 1500

West, Salt Lake City, UT 84116, (801) 596-1887. Email contact
links and up to date contact information can be found at the
Foundation’s website and official page at
www.gutenberg.org/contact
 Section 4. Information about Donations to
the Project Gutenberg Literary Archive
Foundation
Project Gutenberg™ depends upon and cannot survive without
widespread public support and donations to carry out its mission
of increasing the number of public domain and licensed works
that can be freely distributed in machine-readable form
accessible by the widest array of equipment including outdated
equipment. Many small donations ($1 to $5,000) are particularly
important to maintaining tax exempt status with the IRS.

The Foundation is committed to complying with the laws

regulating charities and charitable donations in all 50 states of
the United States. Compliance requirements are not uniform
and it takes a considerable effort, much paperwork and many
fees to meet and keep up with these requirements. We do not
solicit donations in locations where we have not received written
confirmation of compliance. To SEND DONATIONS or determine
the status of compliance for any particular state visit
www.gutenberg.org/donate.

While we cannot and do not solicit contributions from states

where we have not met the solicitation requirements, we know
of no prohibition against accepting unsolicited donations from
donors in such states who approach us with offers to donate.

International donations are gratefully accepted, but we cannot

make any statements concerning tax treatment of donations
received from outside the United States. U.S. laws alone swamp
our small staff.

Please check the Project Gutenberg web pages for current

donation methods and addresses. Donations are accepted in a
number of other ways including checks, online payments and
credit card donations. To donate, please visit:
www.gutenberg.org/donate.

Section 5. General Information About

Project Gutenberg™ electronic works
Professor Michael S. Hart was the originator of the Project
Gutenberg™ concept of a library of electronic works that could
be freely shared with anyone. For forty years, he produced and
distributed Project Gutenberg™ eBooks with only a loose
network of volunteer support.

Project Gutenberg™ eBooks are often created from several

printed editions, all of which are confirmed as not protected by
copyright in the U.S. unless a copyright notice is included. Thus,
we do not necessarily keep eBooks in compliance with any
particular paper edition.

Most people start at our website which has the main PG search
facility: www.gutenberg.org.

This website includes information about Project Gutenberg™,

including how to make donations to the Project Gutenberg
Literary Archive Foundation, how to help produce our new
eBooks, and how to subscribe to our email newsletter to hear
about new eBooks.
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com