WASHINGTON -- If a tree falls in the forest, physicist Mark Coles and his colleagues do not want to hear it.

They have begun a pioneering search for the first direct evidence of gravity waves using a sensitive new detector nestled in the woodlands of Livingston Parish, Louisiana.

To be successful, their instrument, and a companion detector in southeastern Washington state, must be isolated from the bumps and shakes of the natural world and human activities such as the logging now under way near the Louisiana detector.

The detectors, called jointly the Laser Interferometer Gravitational-Wave Observatory (or LIGO), are designed to pick up a murmur from the cosmos so slight as to be almost laughable.

A typical gravity wave, a ripple in the fabric of space and time caused by a cataclysmic event in deep space, is estimated to be so weak when it reaches Earth that the entire planet briefly expands and contracts less than the width of an atom in .response.

So the LIGO teams have their work cut out for them, not least ensuring that all extraneous vibrations are eliminated from the test apparatus. The detectors then are designed to pick up movements as minuscule as 1/10,000 of the diameter of a proton.

"It's just an exquisitely small number that we are trying to pull out,” said Coles, director of the Louisiana LIGO .facility.

While the task is daunting, the payoff would be immense.

A successful detection would be big news, a confirmation of one of the fundamental predictions of Einstein's general theory of relativity. Detection of gravity waves would open a new window on the universe for astronomers, says Gary Sanders, a physicist at the California Institute of Technology and deputy director of the LIGO project.

"Every time we have looked at the sky in a new way, we have seen a new universe,” Sanders said.

For centuries, astronomers had to rely on the naked eye or optical telescopes to scan the heavens. But since the 1940s, new types of telescopes -- radio, infrared, gamma ray -- have allowed scientists to "see” other wavelengths of electromagnetic radiation being emitted by distant objects.

Gravity waves, which are not a part of the electromagnetic spectrum, would be an entirely new medium, Sanders said. They could bring information about phenomena, such as black holes or dark matter, that are not amenable to direct study by other instruments.

"This is a different messenger,” .Sanders said.

And capturing that messenger is a difficult technical challenge. The detectors still are 100 to 1,000 times more jittery than plans call for, officials said. The teams had been fine tuning the equipment for nearly three years in a series of engineering runs before launching the first collection of science data during a 17-day run that ended last week.

"It's going about as well as we could want at this stage,” said Barry Barish, a Cal Tech physicist and director of the project. "The sensitivity is just what we thought it would be.”

Although the odds are long that LIGO will be able to detect gravity waves in its initial configuration, scientists are impressed with the progress so far.

"They have achieved technical goals which seemed impossible a decade ago and which even today defy what we scientists have come to think was feasible,” said John Bahcall, an astrophysicist at the Institute for Advanced Study in Princeton, N.J.

Nor is the United States alone in the quest for gravity waves. New interferometer projects are under way in Germany, Italy, Japan and Australia.

The sensitivity of the LIGO detectors comes from the clever use of mirrors and lasers as well as a determined effort to screen out all extraneous influences -- everything from the long-distance ground murmur created by heavy ocean waves to the collective motion of gas molecules in the air.

At the most basic level, designers of LIGO want to detect the slight stretching of space in one direction and shrinkage in another as a gravity wave passes by.

Each detector is L-shaped, with 2.5-mile-long tubes set at right angles to each other. A powerful laser beam is split and sent down each tube. The beams bounce off finely polished mirrors hanging at the ends of the tubes and return to the corner of the L. They return in phase and when combined in an interference pattern, the peaks and valleys of the light waves cancel each other out.

But if a gravity wave were to pass through the detector, it should stretch one of the tubes slightly and shrink the other, disturbing the hanging mirrors ever so little and throwing the laser beams out of phase. The interference pattern would reveal the form of the passing gravitational wave. That, in turn, might give clues about the cosmic object or objects that caused the wave.

In building LIGO, which was funded by the National Science Foundation for $365 million, designers knew that they needed at least two detectors located at widely separated sites in order to have unequivocal evidence for a gravity wave.

The sites in Washington and Louisiana were selected because they are seismically quiet. The 2.5-mile-long vacuum tubes in which the LIGO mirrors hang are mounted on hefty shock absorbers to screen out ground motion.

But the LIGO detectors, particularly the one in Louisiana, continue to prove more jittery than planned.

"The ground motion is really higher than we had anticipated from the original site surveys,” Coles said. Those were done when the area was largely free of logging activity, he said. In recent weeks, he said, loggers have been working a tract just 500 yards away from the corner station of the detector. "Even if they were a mile away, we couldn't run during the day,” Coles said.

The science run was conducted largely at night, when logging stopped. Even then, the passing of nearby freight trains at midnight and 6 a.m. set limits on how long the device can run for now without interruption.

"We want to be insensitive to vehicles going by,” Coles said. "When a tree falls, we don't want to know about it. We want to be immune to everything” but gravity waves. "It is decoupling from the rest of the world that we want.”

To that end, the Louisiana team is planning to install a vibration-damping system this winter that will actively adjust the detector mounts from moment to moment to further prevent unwanted motion.

When it is running, Coles said, the Louisiana interferometer has been quite reliable. The lasers remain "locked” in focus for hours at a time and other technical issues have been solved, including the challenge of keeping a tight vacuum in 5 miles of steel tubing at each LIGO site.

While not having to contend with nearby logging, the Hanford team has some headaches of its own, including the rumble of cement trucks nearby where the Department of Energy is building a plant to process nuclear waste.

"These are still works in progress,” said Fred Raab, head of the LIGO fa.cility at Hanford, Wash.

Plans are well along for upgrade of both detectors with more powerful lasers and better optics as well as more sophisticated vibration protection. If all continues to go well with the science runs now getting under way, officials said, they may request money in the fiscal year 2004 budget to start the upgrade, with an eye to finishing LIGO II by 2007.

Many scientists say the upgraded LIGO will have the best shot at detecting gravity waves, although they say the teams will gain essential experience with the detectors as they are now .configured.

Part of the uncertainty about results comes simply from the novelty of the whole enterprise and the limited data on which to base predictions. The first indirect evidence for the existence of gravity waves came in 1974, when two astronomers found a pair of neutron stars, the superdense remnants of collapsed stars, that were spiraling toward each other. For more than 20 years, Russell Hulse and Joseph Taylor Jr. observed the decreased orbital period as the neutron stars moved closer and closer to each other. Their studies, for which they won a Nobel Prize in 1993, confirmed that the ever-tightening death spiral of the stars could be explained as a loss of .gravitational energy by the emission of gravity waves.

But fewer than a half dozen such bi.nary neutron star systems have been observed, making it difficult for scientists to say much about the character of the presumed gravity waves they may be sending our way.

Predictions of whether the current LIGO detectors will see anything "are based on small amounts of evidence and lots of inference,” said Sanders of Cal Tech. The detectors could pick up something in their early runs, he said, but the chances "are not high.”

But he adds, "What we'll see will be different from the predictions,” just as every new instrument in astronomy has provided surprises. "Why should looking at the sky with a messenger as different as gravity waves be any different?” he asks.

For his part, astrophysicist Bahcall has no doubt that gravity waves are real. Will LIGO eventually detect them? "Yes, if we have the courage and the conviction and the financial ability to hang in there for the long term,” he said.