Twinkles in the Dark

By MICHAEL D. LEMONICK

The last thing physicist Kim Griest expected was to find what he was looking for. Griest and colleagues at the University of California, Berkeley had been scanning the skies for more than a year in search of the mysterious and elusive material called dark matter. The scientists couldn't see it and couldn't say what it was, but they knew it was out there because of its undeniable effect on stars and planets. What could the invisible stuff be made of? The Berkeley group was checking out a theory that dark matter takes the form of large planets or small, dim stars -- a plausible idea, but one that Griest suspected was wrong. A little over two weeks ago, he was prepared to say confidently that dark matter was almost certainly something else.

It didn't work out that way. At a meeting in Italy last week, Griest's boss, Bernard Sadoulet, announced that the team had discovered what appeared to be a tiny star or a huge planet, lurking out beyond the visible stars of the Milky Way. It may be just one of trillions of similar objects, whose combined mass far outweighs all the known stars. At the same time, a French group doing its own search disclosed that it had found two more of the dark bodies, making it unlikely that either team had made a mistake. If the discoveries pan out, they may solve a puzzle that has baffled scientists for nearly 60 years.

The hunt for dark matter is not just an academic exercise. It is a quest to know the fate of the universe. The unseen material makes up at least 90% of the mass in the cosmos, generates most of the gravity and thus controls the universe's evolution. If there's enough dark matter producing a sufficient amount of gravitational force, the universe will eventually stop expanding and then collapse in an apocalyptic Big Crunch. If there's not, the expansion will go on forever.

How do scientists know dark matter exists? Most galaxies rotate so quickly ) that they would disintegrate if they were not surrounded by a massive halo of invisible matter. Similarly, pairs and groups of galaxies revolve around one another faster than they should, unless there is more mass, and thus more gravity, than there appears to be.

Over the years theorists have come up with many competing notions about the identity of dark matter. The candidates have included various kinds of subatomic particles, many of which aren't even known to exist; black holes; and even long, thin strings of pure energy left over from the Big Bang. Large planets or dim stars -- known as MACHOs (massive compact halo objects) -- are by far the most mundane of the solutions to the puzzle. They're also the least popular: theorists think there should be just enough dark matter to stop the universe's expansion without reversing it, and MACHOs can't be numerous enough to do the job.

Still, they were worth looking for, if only to prove they weren't there, and Princeton astronomer Bohdan Paczynski had proposed an ingenious way to conduct the search. Albert Einstein showed in his general theory of relativity that the gravity from a star will bend rays of light that pass nearby. In principle, he said, a star could act as a lens, focusing and brightening the light of another star directly behind it. If a cloud of small stars or big planets really is orbiting the Milky Way, some of them should occasionally pass in front of stars in the next galaxy over -- the Large Magellanic Cloud. If you watched this galaxy very carefully for a year or two, you might sometimes see a star getting inexplicably brighter, then dimmer again. If you saw nothing, then there were no MACHOs worth talking about.

That was the strategy used by the American and the French groups, as well as by Paczynski and several Polish astronomers. Scanning the stars was only the beginning; the astronomers then had to put thousands of megabytes of data from their telescopes through a computer. The computer's job was to identify the unusual flickers of light caused by MACHOs amid the flashes from thousands of naturally pulsating stars that regularly switch from dim to bright and back again. After nearly 2 million individual observations that yielded just one dubious MACHO, Griest's group was ready to give up. Then, unexpectedly, the computer spit out what he calls "a beautiful event." After Griest and his colleagues had raised and ruled out phenomena that might be tricking them, they were ready to unveil their MACHO.

There is still a chance that what Griest found was some bizarre kind of variable star, but the fact that the French saw the same flicker in an entirely different type of star argues against that possibility. The next step is to comb through another million observations already stored in the computer. If nothing more shows up, it means MACHOs alone can't account for all the dark matter. Attention would then shift to the hunt for undiscovered subatomic particles. But if several more MACHOs -- Griest won't say exactly how many -- pop out of the computer, then they could probably account for the entire dark-matter halo. And scientists could be more confident that they have at last found the main fabric of the universe.