While satellites are adept at spotting early signs of activity, geologists still need strong boots and the courage to use them on the rough and dangerous terrain around volcanoes that are acting up. Sensors stuck on the mountainside are still the best way to predict an imminent eruption.

But volcano forecasting has proved notoriously difficult. And people die, even in modern times. Mt. St. Helens killed 57 people in 1980. In 1991, after 600 years of dormancy, Mount Pinatubo in the Philippines rumbled for days before erupting and killing about 750 people, including journalists who had been stationed at a supposedly safe distance.

Emergency officials need better information in order to make sensible decisions about evacuations. Help seems to have arrived, via the geologists who study Mt. Etna.

Gene Ulmer, a geology professor at Temple University, was in the village of Nicolosi, on the flanks of Mt. Etna, on June 18. He and 20 other geologists had just visited the mountain and were eating dinner at a restaurant.

The next day, Ulmer and the other visiting geologists hunkered down at a spot three miles from the volcano, recommended as safe but with a clear view.

"At 1:33 p.m., the mountaintop blew out," Ulmer said. "Within 10 minutes the summit was completely obscured in a cloud of dust and black plumes rising as much as 10,000 feet into the air. There was a tremendous noise."

Heavy ash fell to the ground. Within a couple of hours, finer ash began to settle back to the surface. The air, choked with the tiny particles, looked like milk, Ulmer recalls. "As it came down, it was like someone was painting every ravine with white paint," he said.

What he witnessed was the start of what has developed into a major set of eruptions that are still continuing, Etna's biggest temper tantrum since 1992.

But what Ulmer marveled at most was the accuracy of the prediction.

"All of us in volcanology are excited that maybe we have finally found out the right monitoring instruments to put into the mountain so we can make some life-saving predictions about possible eruptions," he said.

One of those instruments is a spin-off from the Apollo era. Reflecting devices similar to those left on the Moon are stationed at various points on the mountain. From miles away, researchers bounce lasers off the mirrors. If they have moved, the laser beam shifts.

Other devices on the mountain sense changes in temperature and the output of telltale gases.

Ironically, while Mt. Etna has been well studied by the satellite radar interferometry method, the current eruption will be mostly missed.

Paul Lundgren, a JPL geophysicist, has used the technique to model how the ground swells on Mt. Etna, studying how plumes of magma inside the volcano behave. In research published in 1998, Lundgren and his colleagues showed Etna acts like a giant, slow lung, inflating for months prior to eruptions and then deflating afterward.

But Lundgren, who returned last Thursday from a field study of Mt. Etna, said the ERS 2 satellite that gathers the data is malfunctioning. No NASA satellites, flying or planned, can do the work, he said. This fall, the European Space Agency will launch ENVISAT, which will have the same capability is ERS 2 but will produce data Lundgren said "will not be suitable for combining with previous ERS data to make interferograms spanning the current eruption."

Ulmer, the Temple geologist, was forced to retern from Italy earlier than he had planned. He was interviewed by telephone last week from his home in Philadelphia, where he is recovering from a volcano-induced staph infection he contracted in Italy.

During his trip, Ulmer had visited several volcanoes. And no matter how good a volcanologist's boots are, they fill with volcanic ash that he describes as having the consistency of flour. But the ash is made up of small glass shards that, once inside his boots, were ground into his skin. His feet blistered and then became infected.

An expert like Ulmer knows the many ways a volcano can kill. He had a first aid kit with antibacterial salves and pills. But the kit had not been updated in three years, he later realized, and the medicines were not effective.

After a month of missed work and a steady dose of antibiotics, Ulmer expects to survive. But humans throughout history have not had as much luck in their interactions with volcanoes. Some scientists say we're lucky to even be here.

Some 74,000 years ago, in what is now Sumatra, a volcano called Toba erupted with a force estimated to have been 10,000 times that of Mt. St. Helens. The sky darkened around the globe as ash blocked out the Sun. Temperatures plummeted by as much as 21 degrees at higher latitudes around the planet, said Michael Rampino, a biologist and geologist at New York University.

Rampino has estimated that three-quarters of the plants in the Northern Hemisphere may have died.

Stanley Ambrose, an anthropologist at the University of Illinois, suggested in 1998 that Rampino's work might explain a curious bottleneck in human evolution, a phenomenon observed by other researchers who study DNA: The blueprints of life for all humans are remarkably similar given an evolutionary timeline known to stretch back more than 2 million years.

Ambrose thinks that early humans, struggling as always against the elements, were pushed to the edge of extinction after the Toba eruption. Perhaps only a few thousand survived, Ambrose says. Humans today would all be descended from these few, and in terms of the genetic code, not a whole lot of evolution occurs in 74,000 years.

At the least, however, we evolved enough to gain the capacity to invent satellites and employ them to warn us of the next Toba, if it is to come.