If Fischer was elated, though, Marcy was “frightened”; the discovery of a new solar system with multiple planets, like our own, was so unexpected that he worried people “might think we’re lunatics.” Only after 10 days of exhaustively rechecking his calculations did Marcy call his collaborator Robert Noyes of Harvard, who was observing the same star at the Whipple Observatory in Arizona. His first words were: “Bob, I hope you’re sitting down” and only after Noyes confirmed the pattern with his own data did the two teams announce their findings last week. It was, after all, only four years ago that the only planets known to orbit a normal star were the nine in Earth’s solar system. (A planetary system surrounding a neutron star–which emits X-rays rather than visible light–was found in 1992.) Since 1995 about 20 extrasolar planets have been discovered. But all have been solitary, raising the question of whether our nine-planet system is somehow unique. The discovery of Upsilon Andromadae’s three planets, says Fischer, “implies that our Milky Way is teeming with planetary systems.”

Of course, the search for other planets is also, in part, a search for extraterrestrial life. Unfortunately none of the three planets of Upsilon Andromedae, nor any of the others discovered so far, is a likely prospect. One, designated “B,” orbits only at a distance of about 5 million miles from its star (about one-twentieth the distance of the Earth from the Sun), suggesting surface temperatures far above the boiling point of water. The two others, “C” and “D” (there is no “A”), are in more temperate, highly elliptical orbits that would fit between Venus (67 million miles from the Sun) and Jupiter (484 million miles). But all of Upsilon’s planets are also huge. Their exact mass can’t be calculated, but even the smallest is at least 225 times Earth’s mass, while the largest is four or more times as big as Jupiter. Noyes believes that planets of this size are unlikely to be solid, because–based on our solar system–there aren’t sufficient quantities of heavy elements available to make them. Instead, he suspects that like Jupiter, they will be composed mostly of gas or frozen vapor, unlikely places for life to develop.

But there’s no reason to assume that terrestrial planets in Earth-like orbits don’t exist; big ones that orbit close to their stars are, obviously, the easiest to find. And even those require exquisitely sensitive instruments and calculations. Planets outside the solar system cannot actually be viewed from Earth. The light they reflect from their stars is many millions of times less bright than the starlight reaching us directly. Instead, their orbits and mass are deduced from their gravitational effects on the stars themselves. Upsilon Andromedae, 44 light-years (259 trillion miles) distant, is moving away from Earth at an average speed of around 1.8 kilometers a second (about 4,000 miles per hour). As the planets circle it, their gravitational fields tug on the star and either slow it down (when they’re between it and the Earth) or speed it up (when they’re on the far side). But the changes are tiny, on the order of a few tens of meters a second–not much faster than a man can run. After taking into account the Earth’s own rotation, turbulence in the atmosphere and other sources of possible error, researchers are now approaching the theoretical limits of detection with the equipment they now have. “If we were on an extrasolar planet observing our own solar system,” says Noyes’s colleague Timothy W. Brown, “we wouldn’t even find Jupiter, much less the Earth, by our method, because Jupiter is so far from the Sun.”

But better instruments are on the way. NASA’s Space Interferometry Mission, scheduled to be launched in 2005, will deploy much more sensitive motion detectors than Earthbound telescopes, able to detect a star’s side-to-side motion as well as changes in its speed relative to Earth. Six years later, NASA hopes to launch the Terrestrial Planet Finder, which will have instruments, still being developed, that can electronically “nullify” the light coming from a star, making it possible for the first time to directly view the reflected light of an orbiting planet. Such a device could detect Earth-sized planets in temperate orbits, and even record their surface temperatures and the composition of their atmospheres–strong indications of habitability, if not life itself.

And the existence of other multiplanet solar systems increases the likelihood of finding life–in particular, life that has been around long enough to evolve some complexity. As Marcy’s colleague R. Paul Butler points out, our very own existence owes a lot to other planets in our solar system, especially Jupiter. A single Earth-sized planet would be a sitting duck for meteors, comets and other space junk that can wipe out whole ecosystems in a flash–as the dinosaurs discovered 70 million years ago. But the immense gravity of Jupiter (and to a lesser extent the other giant outer planets) sweeps the neighboring skies almost clear of dangerous debris, like a celestial vacuum cleaner, so that “we hardly ever get pelted by anything large anymore.” Radio astronomer Alex Wolszczan, who discovered the first neutron-star solar system in 1992, calls the Upsilon findings “historic,” a confirmation that the variety of planetary systems is more impressive than anyone could have imagined. “Sooner or later,” he adds, “we’ll start to find other systems like our own. There’s no logical reason why we should be unique.”