¿ìè¶ÌÊÓƵ

There is more than meets the eye to our Galaxy, the Milky Way. Most
of its 100 billion bright stars can be found within a relatively thin disc
about 100 000 light years across and only some 2000 light years thick. These
stars are not static: they orbit the centre of the Galaxy, and as they do
they oscillate up and down within the confines of the disc rather as the
needle of a sewing machine bobs up and down through the cloth. This motion
is constrained by the gravitational pull of the matter in the disc. The
more matter there is, the more tightly it grips the stars and the smaller
the amplitude of the bobbing. Studies of many stars over the past sixty
years all agree that something is missing: the motion of the stars indicates
that there must be a great deal more matter in the disc than we can see
in the form of bright stars.

Now researchers think they have identified where and what this dark
matter is. Late last year, three teams of researchers independently reported
that they had found evidence that the visible disc of the Milky Way may
be surrounded by a halo that contains five to ten times as much matter
in the form of dark objects as all its bright stars put together. These
objects are called massive astronomical compact halo objects, or MACHOs
for short.

The first hint that the bright stars account for only a small proportion
of the mass of galaxies like our own came from studies of the way these
stars rotate. Across the entire disc of a galaxy like the Milky Way, the
rotation speed is constant. This can only mean that the disc of bright stars
is embedded in a much bigger disc of dark material. If most of the galactic
mass were concentrated in the centre, outer stars in the disc would move
more slowly, just as the outer planets of the Solar System rotate more slowly
than the inner ones under the gravitational influence of the Sun. But if
mass is spread evenly through the visible disc and beyond, then even outer
stars will be strongly influenced.

By the middle of the 1980s, it was clear that our Galaxy contains up
to ten times as much dark matter as the matter we can see in the form of
stars. All this dark stuff is likely to be made of the same kind of ‘baryonic’
matter as the atoms that make up your body, or planet Earth, or our Sun.
This differs from the cosmic dark matter that is sometimes invoked to explain
what holds the Universe together. Such dark matter cannot be baryonic, because
astrophysicists know how much baryonic matter was formed in the big bang
and this amount is already used up in our Galaxy. Candidates for cosmic
dark matter include neutrinos and weakly interacting massive particles,
or WIMPS.

The amount of baryonic material in the Universe was determined by the
conditions in the cosmic fireball of the big bang. It can be calculated
from studies of two characteristics of the Universe as it is today. The
first is the cosmic background of microwave radiation, which can be thought
of as the afterglow of the big bang. The second is the composition of stars
that were formed when the Universe was young; measurements of the spectra
of light from these stars indicate the amount of helium they contain. Together,
these measurements put tight constraints on the original composition of
the Universe and hence on the amount of baryonic material that could have
been produced. Overall, there could be up to ten times as much baryonic
material as we see in the form of bright stars.

Originally there were two theories about what sort of bodies the MACHOs
could be. The first, and most spectacular, was that they are big black holes,
formed when our Galaxy was very young. Large clouds of material might have
collapsed to form very heavy ‘super stars’ which ran through their life
cycles very quickly and then exploded, leaving massive black holes behind.
(Such black objects would count as baryonic material because they had been
made from stars, unlike the hypothetical mini black holes that might have
been made in the big bang before baryons formed.)

Each baryonic black hole would have a mass millions of times that of
the Sun. Such bodies would be expected to exert a detectable influence
on the structure of galaxies by attracting ordinary stars, which would be
accelerated and ‘boil off’ from the Milky Way. But nobody has observed the
Galaxy evaporating in this way, so astrophysicists now believe that the
matter cannot be in the form of massive black holes (see ¿ìè¶ÌÊÓƵ,
Science, 8 January).

Too light to shine

That leaves the rather less dramatic possibility that MACHOs are stars
that have failed to light up. Stars shine because energy is released by
nuclear fusion reactions in their interiors, which typically convert hydrogen,
the lightest of all the elements, into helium. These fusion reactions can
only be triggered if a star has enough mass for its gravity to squeeze the
hydrogen in its heart so tightly that it overcomes the electrostatic forces
that normally keep the positively charged particles apart.

The critical mass for a star to form out of the Universe’s primeval
gas lies somewhere between the mass of our Sun and that of the planet Jupiter,
the largest planet in the Solar System, which has about 0.4 per cent of
the Sun’s mass. But stars with a mass less than about 8 per cent that of
the Sun – that is 20 times as massive as Jupiter – never get hot enough
to trigger nuclear burning. Their surface temperature never rises above
2000 K and their brightness never exceeds one-millionth that of the Sun.
Such stars, called brown dwarfs, eventually fade into complete oblivion
as old, dead stars called black dwarfs.

In 1985 Bohdan Paczynski, an astrophysicist at Princeton University
in New Jersey, came up with an idea for how MACHOs might be detected: he
realised that a MACHO in our Galaxy should act as a gravitational lens.
If light from a very distant star passes near a dark object on its way to
the Earth, the otherwise invisible dark object will reveal its presence
by bending the light gravitationally and focusing it. If the dark object
is in the right position in the sky relative to the distant star, this will
produce a brightening of the distant star.

So when a MACHO passes in front of a distant star, that star appears
to flash briefly. Such gravitational lensing is known to happen on a much
larger scale: light from bright, very distant quasars, way beyond our own
galaxy, is focused in this way by the gravitational lensing effect of a
whole galaxy. What Paczynski had in mind is now known as microlensing –
a smaller-scale version of the same effect. Light from the distant star
would be bent around the intervening object and focused, so that from Earth
the distant star would brighten and fade as the MACHO passed in front of
it (see Figure below).

It was straightforward to calculate the size of this effect and to predict
the shape of the ‘light curve’ that would be obtained by plotting the variation
of brightness with time associated with such an event (see Figure opposite).
Seen through telescopes on Earth, the light curve of the distant star should
rise, and then fall, in a smooth, symmetrical curve.

Rare event

Unfortunately, there appeared to be no practical way of putting Paczynski’s
idea into operation, as the chances of actually seeing such a microlensing
event seemed tiny. For example, Paczynski calculated that a MACHO with a
mass 1 per cent that of our Sun, passing in front of the rich backdrop of
stars provided by the Large Magellanic Cloud, a near neighbour to our own
Galaxy, would on average produce a microlensing effect once every 50 000
years or so. Each event would last just over six days. The chance of a telescope
just happening to be pointed at that particular star at that particular
time is tiny.

The only hope of detecting microlensing events would be to monitor millions
of stars for as long as possible on as many nights as possible. To complicate
things further, it would be best to monitor them at two different wavelengths
simultaneously. When a star flares up, it usually changes in brightness
and colour. By contrast, a microlensing event simply acts as a cosmic magnifying
glass, and increases the brightness of the background star equally across
the entire spectrum.

Remarkably, something that less than ten years ago seemed impossible
has since become reality. In that time astronomy has been transformed by
two key developments. The first is the widespread replacement of photographic
plates by solid-state charge-coupled devices. CCD cameras are now used as
standard by many observatories. They can have more than a million picture
elements in an array 1024 elements square. Hundreds of thousands of stars
can be monitored using a single detector of this kind.

The second development is the increase in the speed and memory of computers.
A microlensing survey may produce 5000 megabytes of data for analysis each
night (or several hundred thousand times as much data as the text for this
article), and the analysis may involve measuring the brightnesses of as
many as 10 million stars and comparing them with previous measurements.
This really would be impossible if the observations consisted of photographic
plates that had to be compared with one another by eye. But computers can
now handle the task in real time, identifying microlensing events as they
happen. This has allowed astronomers to turn their telescopes towards the
phenomenon before it fades away.

Last autumn the first microlensing events were reported simultaneously
by a team of American and Australian astronomers and by a French team. A
third collaboration, involving Polish and American astronomers and headed
by Paczynski, was slightly slower off the mark with its publicity, but found
a microlensing event at about the same time.

The US-Australia team, headed by Charles Alcock of the Lawrence Livermore
Laboratory in California, found one flashing star with the characteristics
of a microlensing event using a CCD system at the 1.2-metre telescope of
the Mount Stromlo and Siding Spring Observatories in Australia. The French
team, led by Michel Spiro of the DAPNIA laboratory at Saclay, reported
two events. Their observations were made at the European Southern Observatory
in Chile. Paczynski’s group, also using a telescope in Chile, at the Las
Campaas Observatory, reported a single observation.

This handful of observations is being taken seriously as evidence of
microlensing because they have all the right characteristics. First, the
light curves are very smooth and symmetrical, just like Paczynski’s theoretical
predictions (see Figure below). And the event looks exactly the same in
the red and blue sections of the visible spectrum – strong evidence that
it has been caused by microlensing and not by a flare on the distant star.
In addition, both the amount of brightening, by factors ranging from 2.5
to 7, and the duration of the events, between 17 and 33 days, match the
theoretical predictions. The object responsible for the US-Australia team’s
event could be as massive as the Sun. The astronomers say the ‘most probable’
mass is about one-tenth of this, but the other two candidates have masses
in the range from 3 per cent to 30 per cent of the mass of the Sun.

While most astronomers accept that these are genuine microlensing events,
the exact nature of the MACHOs remains something of a mystery. The masses
suggested by the first observations, from about 10 per cent to 30 per cent
of the mass of the Sun, are a little too high for them to be true brown
dwarfs. Nevertheless, it is tempting to conclude that they are indeed very
faint ‘real’ stars. Alternatively, they could be a previously unknown kind
of object. Whatever they are, there would have to be about five thousand
billion of them to account for the missing mass of our Galaxy.

Astronomers are hoping that further searches of the heavens will not
only resolve these questions, but also teach us something about how the
Galaxy itself formed. When stars form out of clouds of gas and dust today,
only a small proportion of material gets turned into very low-mass stars.
If these new observations are representative of MACHOs in general, this
means that when the Galaxy was young, low-mass objects must have been favoured.
Meanwhile, astronomers have plenty more work to do to understand the nature
of MACHOs and their relationship to the bright stars that they pass in
front of.

Further reading: In the Beginning by John Gribbin (Viking). The Hidden
Universe by Roger Tayler (Ellis Horwood).