The sheer immensity of the Antarctic ice is almost unimaginable, even when seen from a plane flying five miles up en route to the South Pole. Under a sun that never sets, the relentless glare stretches in every direction from horizon to blazing horizon.

And when suddenly a few stony spots appear in this solid sea of white, it takes a few seconds to wrap the brain around the sight. Those tiny protusions are the uppermost visible tips of the Transantarctic Mountains, a 14,000-foot-high chain the size of the American Rockies, buried in trillions of tons of ancient ice. And that isn't even the thickest part of the sheet. In some places, it's three miles deep.

Indeed, Antarctica holds 90 percent of the planet's ice, and 70 percent of its fresh water. If it all turned to liquid, sea levels would rise 200 feet, submerging half of civilization. That's not going to happen. But if even a modest fraction of that titanic mass were to melt within a couple of decades--possibly in response to global warming--it would still raise sea levels from several feet to several yards, evicting many of the 3 billion people who live within 50 miles of the ocean.

Throughout this century, the sea has been rising steadily by a millimeter a year, about the thickness of a dime. So far, that's manageable.

But any effects of global warming would most likely show up first and most dramatically at high latitudes. And recent evidence suggests that much of the West Antarctic Ice Sheet may be unstable. Crossing some still-unknown temperature threshold could trigger catastrophic melting.

Nor is that the only threat. The churning of Antarctica's super-cold border waters powers much of the global sea circulation system that redistributes heat around the planet. And the shiny ice cover, about one-twelfth of Earth's total surface area in the coldest months, reflects a huge amount of sunlight, keeping the climate far cooler than it would be if that energy were absorbed by land or water. Thus, any appreciable change in the ice would derange climate patterns worldwide. Remote as it may seem, the Antarctic ice plays a crucial role in the life of every person everywhere.

That's why, on a fiercely bright day last month glaciologist Andrew Fountain and assistant Patrick Bardel were picking their way across the slippery top of Taylor Glacier, which creeps half an inch a day down from the mammoth East Antarctic ice sheet toward the Ross Sea in a 30-mile-long slab 1,000 to 2,000 feet thick.

Fountain and colleagues monitor 100 points on four glaciers twice a year to see how their mass is changing. "We really don't know whether the ice sheet is growing or shrinking," said Fountain, of Portland State University in Oregon, squinting into the frigid wind sluicing down the desolate valley.

The project is one of about 150 in Antarctica for which the National Science Foundation (NSF) spends more than $200 million a year to try to determine how the continent is evolving, how it might be responding to global warming, what effects it has on world weather patterns and, of course, whether the planet's seaside habitats are doomed.

It will take years to find answers. But about 1,500 scientists and support personnel spend the November-to-February local summer working in some of the harshest conditions on Earth in an effort to answer some of these questions, which may be among the most important facing humanity.

Like nearly everyone working in Antarctica, Fountain and Bardel arrived by military cargo plane at McMurdo Station, the year-round U.S. and NSF headquarters at the bottom of the world.

After a grueling eight-hour flight so noisy that thick earplugs are mandatory, arriving passengers disperse from McMurdo (summer population 1,250) to sites around the continent. A couple hundred fly to the snow-buried Amundsen-Scott South Pole Station 850 miles north, the second of NSF's three year-round facilities. (The third, Palmer Station off the Antarctic Peninsula, is serviced by ship via Chile.)

Others travel 800 miles west onto the West Antarctic Ice Sheet (WAIS), to camps such as Siple Dome, where drill crews are extracting ice cores to look for clues to the history of climate. Still others, including Fountain, go east to study the history and ecology of the baresided Dry Valleys--rocky scarps separated by glacier tendrils.

There are few comfortable assignments on the continent, and much of the work is done while burdened with 30 to 50 pounds of multiple layers of clothing and gear. Although January temperatures can be as high as 32 degrees Fahrenheit at the coast, they are minus 30 or colder in the interior--and there is almost nothing poking above the ice to impede the winds that come screaming off the high mid-continent and into the lower elevations. Lethal storms can freeze-dry a creature within minutes.

So every scientist or technician who comes into Antarctica through NSF is required to undergo rigorous physical and dental screening to make sure they can endure their stay. ("Winter-over" personnel, who brave the acute confinement and near total darkness from March to October, also get psychological tests.)

Once there, they are issued one or two large bags of extreme cold weather garments, typically including a heavy red goose down parka, nylon wind pants, bulky mittens and hats, thick rubber "bunny boots" that are over six inches wide across the arch, and a couple bushels of miscellaneous jackets, long underwear, Polartec liners, face protectors, neck gaiters, thermal socks and ski goggles. In addition, people working in the field carry a sizable water bottle: Dehydration is a persistent threat.

Finally, anyone heading for a protected ecosystem also carries an empty bottle for urine and plastic bags for feces. So far, the largest creature found living in the Dry Valleys is a tiny worm called a nematode. But because of NSF's commitment to the principle of "pack it in, pack it out," all human waste must be carried from sensitive field sites.

For scientists trying to devise a prognosis for the ice, each site has its own peculiarities.

"In terms of climate response, there are actually three Antarcticas," said ice veteran Robert Bindschadler of NASA's Goddard Space Flight Center in Greenbelt.

One is the East Antarctic ice mass, which experiences change "on a time scale of tens of thousands to hundreds of thousands of years," he said. Another is the Antarctic Peninsula, popularly known as the banana belt, where many of "the ice shelves are disintegrating, seemingly in direct correlation" with recent warming.

Bindschadler works on the third and possibly most intriguing section--the WAIS, which is separated from East Antarctica by the Transantarctic mountains. Formed 20 million years ago, it has disappeared completely and reformed itself, perhaps more than once. Now rarely over a half-mile thick, it has been shrinking for 20,000 years since the beginning of the end of the last ice age. "It has lost two-thirds of its mass since then," Bindschadler said, "raising world sea levels about 10 meters [33 feet]. It has another five meters left in it. But is it through? Will it continue to shrink? We just don't know."

What researchers do know is that the WAIS is oozing toward the Ross Ice Shelf by way of at least five parallel interior currents called ice streams--gelid equivalents of the way the Gulf Stream flows through the larger Atlantic currents. Some are speeding up; others are slowing down. Bindschadler's team takes measurements using a satellite network known as the Global Positioning System to track how each is moving. The GPS units are so sensitive, Bindschadler said, that "we can now get readings in a week that would have taken us months before."

One major problem: No one knows how the ice streams get started in the first place. At some points, the ice is motionless; a few hundered yards away, it is moving. Knowing the physics of these "onset points" may explain a lot about the future of the WAIS. Meanwhile, Bindschadler said, "I'm more certain than ever about the uncertainties."

Few experts expect an appreciable decline in the WAIS any time soon. Michael Oppenheimer of the Environmental Defense Fund estimates that, even in the worst case, almost all the sea-level rise from WAIS melting would occur "beyond the 21st century." Ice-core scientist Kendrick Taylor of Nevada's Desert Research Institute believes "there's not much [human greenhouse gas emissions] can do to directly influence" the WAIS, and that "it would probably take at least several hundred years, perhaps even more, before the ice sheet responds" to climate changes.

So far, at least, rising air temperature is not menacing. Except for an estimated 5 degrees of warming over the past 50 years around the peninsula, "there has been no significant [temperature] change in other areas," said Charles Stearns of the University of Wisconsin, whose McMurdo operation monitors 50 automatic weather stations on the continent and provides forecasts for the U.S. Antarctic Program.

Nevertheless, many uncertainties remain.

The center for WAIS research is the camp at Siple Dome, a 30-by-50-mile hump in the ice sheet just inland from the west edge of the Ross Ice Shelf. The camp, home to three dozen people in mid-January, ranks about halfway in personal comfort between McMurdo and the numerous remote research camps that are often little more than a tent or hut with a propane heater and stove.

All the dwellings at Siple are Jamesways: modular military structures with plywood walls and padded-canvas roofs, designed to hold 40 or 50 soldiers. At the more luxurious South Pole Station, the Jamesway interiors are divided into personal cubicles by hanging gray wool blankets, giving each person about 6 by 6 feet, most containing a bed, chair and table; and the outbuildings have indoor plumbing. At Siple, each structure is often an open room filled with cots and sleeping bags, with a 10-yard walk to the outhouses.

In its spartan efficiency, Siple has the general aura of an 1840s mining camp run by Ralph Nader. The whole camp, including its fuel station (known as "Babe the Blue Box"), galley ("Cafe Bubba") and stacks of carefully labeled cartons, would fit comfortably into a football field were it not for the airstrip needed to land the 100-foot-long, 65-ton LC-130 Hercules cargo planes on their three 20-foot-long skis--and, of course, the core-drilling rig two carefully marked miles away.

Residents of outposts like Siple--or the Pole, with its 200 summer residents--almost uniformly talk about McMurdo Station the way a rural Nebraskan might regard New York City: overcrowded and rather decadent. For example, the bathing rule at South Pole Station and Siple is a two-minute shower every other day. At McMurdo--where accomodations are like staying two to a room at a Motel Six--showers have been more frequent since 1994, when NSF installed a new desalination system. At Siple, there's one unisex shower enclosure and one washing machine for three dozen residents, about one-third of them women.

It may be inconvenient, but the activity at Siple is essential to understanding the future of the WAIS--a task made more difficult because theories are changing rapidly. As recently as 1993, few experts thought any near-term alternation in the ice was possible. The consensus was that the Antarctic ice mass might fluctuate over hundreds or thousands of years, but that at present it was stable and perhaps even growing.

Then scientists on the other side of the world boring into the Greenland ice sheet--which conatins in its slowly accumulated crystals a detailed chemical diary of weather events over the past 100,000 years--extracted some astonishing evidence. The ice cores revealed that at various times, temperatures there had plunged as much as 35 degrees within a couple of decades. Maybe less. Precipitation apparently doubled in only a few years.

So far, no one fully understands why. But the Greenland findings strikingly showed the possibility of vast and very rapid rearrangements in the interaction of ocean and atmosphere system.

"It's a totally new paradigm we never even thought about six years ago," said Joan Fitzpatrick, technical director of the National Ice Core Laboratory in Colorado. "So the question we have to ask ourselves is: Knowing that the climate has been that unstable...are we, in our ignorance, doing anything" that could cause abrupt change?

Kendrick Taylor of the Desert Research Institute and colleagues are exploring yet another method of determining the Antarctic ice mass's uncertain future by understanding its definite past--a history written in the gases and dust trapped in the ice over hundereds of thousands of years.

The researchers are drilling where the ice is nearly 100,000 years old. That makes it comparable in age to the cores extracted from Greenland. Scientists are eager to see how the two compare. They hope the results will shed light on the "vulnerability of the WAIS to sea-level rise--including whether, during previous warm periods, the WAIS has disappeared completely or not," Fitzgerald said.

One such meltdown may have occurred as recently as 125,000 years ago. If the drilling gets to that depth "and they don't find ice that old, it probably was beacuse there was none," said Bindschadler. "And we're already warmer than during that period."

As of last month, the drilling crews came a lot closer to answering that ominous question. The 110-foot drill rig at Siple Dome finally hit bedrock at 1,004 meters. At that depth, the pressure on the ice exceeds 1,000 pounds per square inch, and extracting pieces of it is a delicate business. As the ice is compressed over millennia, it traps "jillions and jillions of bubbles containing gases in the air at the time that the snow fell," Fitzpatrick said. As the cores are brought up to the thin air of the Antarctic surface, "all they want to do is explode, and they will spontaneously shatter into pieces."

So when the drill descends in pursuit of 3-foot-long cylinders or "cores" about 5-1/4 inches thick, the drillers keep the cores from exploding by filling the hole with hundreds of gallons of butyl acetate, a solvent that smells like a combination of nail-polish remover and molasses and keeps the hole open as they drill. Its density is nearly the same as that of ice, yet it evaporates off the cores rapidly when they are extracted.

The dripping cores are handled as if they were Ming vases, carefully wiped clean, boxed, labeled and then gently placed in walk-in, fan-equipped refrigerators called evaporators.

It seems preposterous to have a refirgerator on the world's coldest continent. But in mid-summer, temperatures at Siple can reach the 20s or even the 30s--hot enough to allow some of the trapped gases to escape from the cores. "Minus 15 Celcius [about 5 F] is what people who work with the core gases think is as warm as it should get," said Gregg Lamorey, who with Taylor oversees the drilling program and coordinates science activities.

The cores are kept in the refrigerator for three days while the last of the solvent evaporates, and then carefully transferred to a storage cellar 25 feet below the ice surface, where the temperature stays dependably around minus 24 C [minus 10 F]. They are stored there to "relax" over the winter--to be retrieved and shipped the next season for analysis in the United States.

"There are so many pieces of the puzzle being examined at once," Lamorey said. "Eventually we should get a really comprehensive idea of what's going on."

But the cost of each new bit of knowledge is a logistical nightmare. There are no roads or vehicles except tractors and snowmobiles on the continent. So anyone or anything that has to move more than a couple of miles goes by helicopter or ski-equipped airplane (usually the Hercules--formerly flown by the U.S. Navy and now by the Air National Guard--whose cargo hold is big enough to handle 27,000 pounds of material, the equivalent of about eight family sedans). People ride in whatever space is available amid the crates and pallets in seats rigged from nylon straps.

Flights to, from and around the continent are always iffy. Fogs or whiteouts appear with surprising speed (one Hercules pilot described it as "like flying inside a Ping-Pong ball"), and many trips from New Zealand to McMurdo are aborted midway--an infuriating process called "boomeranging" by passengers who have managed to retain a sense of humor after eight uncomfortable hours in a cargo bay louder than the front row at a heavy-metal concert.

It takes hundreds of Hercules flights a year to support the science program and provision the skeleton "winter-over" crews that stay at McMurdo (about 200) and the Pole (30) from March to October.

In addition, almost all of what goes into Antarctica comes out again, with 70 percent of it boxed and shipped to Washington state for recycling. There are nearly 20 categories of waste, each of which must be packaged separately, and even the most environmentally devoted staff member can be baffled. For example, product containers, food-contaminated containers, aluminum metal and aluminum foil have different cans. "I got a call one time from someone who wanted to know which waste bin to use for dental floss." said Joyce Jatko, NSF's environmental officer for polar programs.

The ancient and indifferent ice gives up its secrets slowly, and it may be years before scientists begin to solve Antarctica's climate puzzles. But one thing is certain: There will be no shortage of volunteers for the effort. At least one-third of the total staff returns each year. At the Pole, "we have a 75 to 80 percent retention rate" every summer season, said area manager David Fischer. One-fourth to one-third are women.

Why do they come back? Surveying 10 people will produce 10 different answers. Veteran Antarctic types don't raise this kind of question very often, if ever. It's like one cowboy asking another why he works around cattle. Even a weathered nine-season veteran such as Siple Dome camp manager and de facto mayor Sarah Gundlach was hard-pressed to explain why she undertakes a very tough job: "We arrive in October every year, when it's around minus 34, and we have to construct the whole camp from scratch" while living in tents.

"I dunno," she said, wrinkling her brow. "I guess it's the challenge of the work, of building your own little town. They just drop you off there with the minimal amount of equipment. And then when they leave, there's just this...silence."