The technique of combining light from multiple telescopes is called interferometry. Radio telescopes have employed it for years, but it is just emerging as a force in optical astronomy.

It's all about making a big deal of small things.

"When you want to see a small thing in the sky, you need a big mirror," says Mark Swain, an astronomer and technician at NASA's Jet Propulsion Laboratory who works on the Keck telescope. Keck is also an interferometer but is devoted to hunting for extrasolar planets.

"Interferometers are a cheap way to build a big mirror," he says. "A mirror the size of a football field would be tremendously expensive, if even possible."

CHARA's original budget forecast put the project at $11.5 million -- a fraction of what it costs to build and launch a space-based telescope.

"No one person understands it all," says Lu Rarogiewicz, who worked on a predecessor to CHARA that was dismantled three years ago. "It takes a team of fairly specialized people in many disciplines to put it together and get it to work."

McAlister, while giving a quick overview to a group of journalists visiting the observatory, called the technique a "magical process" that "involves a lot of plumbing."

The magic is in the mixing.

Imagine starlight coming from a point at the left edge of a star's disk. The light will travel a slightly longer path to reach one of two telescopes. The trick in interferometry is getting those two incoming light sources to meet at one location with near-perfect precision -- to well within the length of a single light wave.

McAlister and his colleagues do this by sending the light through vacuum tubes to a central building, a long, cramped bunker where the magic mixing is done. Inside, each beam of light is bounced between mirrors and delayed as needed until they are all exactly cued up to be combined.

As the light waves are put together, they interact and produce a series of wave patterns, called fringes, that are either built up or canceled out depending on the telescope's baseline.

Instead of conventional pictures, an interferometer produces multiple fringes, mere squiggles on a computer screen, that can be combined to determine a star's size and shape. But these can be powerful squiggles.

CHARA's ability will be akin to looking from New York, across America and the Pacific Ocean, and spotting a nickel in the middle of Siberia. In astronomy-speak, that's 200 micro-arcseconds of resolution.

The array will eventually consist of six telescopes, each with a light-collecting mirror that is 1 meter in diameter (3 feet). Since 1999, two of the telescopes have been working, but useful observations require a third, which is expected to come online later this month. All six should be operational by early next year, refining accuracy and allowing for quicker observations.

On a clear, shirtsleeve-warm evening in early June this year, a full Moon outshines most stars, even high atop Mount Wilson. But those that remain exhibit a characteristic prized by astronomers: They don't twinkle.

Which is precisely why the site was chosen for the CHARA array, even though it is less than 20 miles from the bright lights and heavy smog of downtown Los Angeles. It's the same reason the mountain got its first telescope in 1889.

"The prevailing winds bring in air that has been flowing across the Pacific Ocean and, except during storms, the temperature is very steady and the air becomes very uniform and flows very smoothly," McAlister explains.

"The air is not disrupted until it flows inland from Mount Wilson," he said. "When you observe stars from a sight like this, the stars don't twinkle much. The twinkling is the result of turbulent air distorting starlight."

The smooth flow of air also traps industrial and auto emissions in an inversion layer, creating the smog that snuggles against the mountains like dirty cotton.

Mount Wilson's elevation is just over 5,700 feet (1,742 meters). "And the inversion layer is typically at 3,000 to 4,000 feet," McAlister said. "So all the crud is down there."

In addition to binary stars, CHARA will study some of the most massive stars, hot young objects known as O- and B-type stars. It will also take a look at some lower-mass, cooler stars that have proved particularly difficult to study by other means.

What's known about all these stars is based heavily on theory, not on observation, McAlister said.

"These stars blow away a lot of their mass," he said. "They have very active winds of material that they expel out into the environment around them. And so we expect to see features of those winds as well."

"Most of the stars in the universe are not singles, so we shouldn't be ruling out binaries ... as places for planets," McAlister said.

CHARA is funded by the National Science Foundation, the W.M. Keck Foundation, the David and Lucile Packard Foundation, and Georgia State University, which will operate it.