In the poorest neighborhoods of the villages flanking Mexico's Popocatpetl volcano, cardboard roofs are being hastily torn off and replaced with sturdier asbestos protection.
Church bells are ringing and warning sirens are wailing, calling villagers to participate in mock evacuations to lower, safer sites.
The reason for all this activity is that El Popocatpetl (popo-kah-TEH-peh-tul), an eternally snowcapped peak visible from Mexico City on rare clear days, is telling its neighbors that it is getting ready for what could be a major and very costly eruption.
Mexican and American scientists studying the 3.4-mile-high volcano, which awoke from a long-dormant state in January 1994, agree all the signs point to an explosion. The riddle they can't yet answer is when El Popo, as the volcano is known by millions of Mexicans, is likely to pop, and how and with what force.
"We could have a large explosion that could expel hot stones and in fact large boulders reaching populations several kilometers away," says Roberto Meli, director general of the Mexican Center for Disaster Prevention in Mexico City. "We could receive a large amount of ash that would cause much trouble for water and water delivery, for power lines, traffic, and crops.
"But while that is possible, the probability is not very high," he adds. "What is more probable is a small explosion that allows the pressure building inside [the volcano] to decrease and the lava flow to stop." A dome of crusty lava is now building inside El Popo's crater. But even such an anticlimactic denouement may not occur for months or years, he says.
Other scientists agree that Mr. Meli is probably right. But the devastation and terror that volcanoes have wreaked in this century alone, plus the scientific age's push to move from probability to prediction, have kept the world's volcanologists searching for ways to better predict when a volcano that is acting up is likely to blow.
One of those volcano specialists is Stanley Williams, a professor of geology at Arizona State University in Tempe. Mr. Williams achieved a bit of fame after being chewed up and spit out by a volcano in Colombia in 1993, and he keeps his name in the public eye by speaking willingly and at length with the press about volcanoes.
He is also perfecting an instrument that from a car or an airplane can measure the gases a volcano is emitting, and thus help in foreseeing what the volcano is likely to do next.
Williams's instrument, which he describes as a "little box with something looking like a telescope sticking out of it," is particularly notable in that it is able to measure a volcano's gases from 20 to 25 miles away. Today, gas-measuring instruments operate from within a volcano's crater, making installation and subsequent monitoring very risky - as Williams experienced in Colombia.
Williams knows firsthand the attraction of the kind of remote-reading instrument he is developing. He was one of seven volcanologists who descended into the crater of the Galeras volcano in the Colombian Andes in January 1993 - and the only one of the seven to come out alive. A sudden explosion of gases, lava, ash, and searing rocks killed the other scientists and left Williams with serious injuries.
Now back tromping around volcanoes - albeit at a safer distance - Williams spent a week in late May snooping and sniffing around El Popo, his gas-reading box in hand. "We now know that the instrument we were there to test works," says Williams, speaking from his office in Tempe by telephone. "We were able to drive around and measure from 30 to 40 kilometers [20 to 25 miles] downwind how much sulfur dioxide and carbon dioxide is coming out of Popo."
El Popo counts nine monitoring stations. Some are located on the volcano's slopes to measure microseismological activity, while others are in the crater to measure changes in the sulfuric and carbon gases the volcano is emitting.
Measuring El Popo's carbon dioxide production is particularly tricky because of all the carbon gases already swirling around in the air from the notorious air pollution of Mexico City, about 45 miles northwest.
Interest in studying volcanic gases is relatively recent - but then volcanology is a young field. The study of volcanoes developed on the heels of the explosion in 1902 of the volcano Pelee on the Caribbean island of Martinique. The volcano was building a dome - like the one now growing inside El Popo's crater - that had become visible from the nearby city of Saint-Pierre. Martinique's leaders told the people not to worry, but the volcano exploded, and more than 25,000 people were killed within minutes by a burning, ferocious, fast-descending cloud of fire, gases, ash, and rock.
Now scientists know that the appearance of a dome of rocky lava means magma is rising within the volcano, and that some kind of explosion is almost sure to occur. They also know that the hundreds of micro-quakes a volcano can register every day say something about how much energy is being released, how much magma is rising, and where it is. And they are learning more about what a volcano's gases can tell them.
As magma expands in a volcano, a tremendous drop in pressure releases gases from the magma. "It's like the bubbles that fizz up when a bottle of soda is opened," Williams says.
As magma first rises in a volcano, light gases like carbon dioxide are likely to predominate. If the magma then sits and ages, a heavier gas like sulfur dioxide becomes more prevalent - usually indicating a cataclysmic explosion is less likely. But if the rise in heavier gases abruptly falls off and is replaced by carbon dioxide, this can mean fresh magma is rising again, that pressure is building against the old lava dome like a cork in a bottle, and a big bang may be imminent.
Scientists got a chance to test volcanic gas readings in the Philippines in 1991 - and warnings based on their findings very likely saved many lives.
The Pinatubo volcano overlooking what was then the United States' Clarke Air Force Base suddenly went from producing very little sulfur dioxide to belching 500 tons a day and then 5,000 tons a day. Then the sulfur dioxide fell off abruptly, replaced by lighter gases. Volcanologists warned Clarke's commanders to evacuate - and 36 hours after 30,000 soldiers left the base, Mt. Pinatubo blew.
Over the few days he recently spent studying El Popo with his box, Williams read the volcano's sulfuric emissions and noted a "pulsating" ebb and flow in the discharges. But he was wasn't there long enough nor did he have access to enough other information, he says, to really decipher what El Popo is up to.
His lack of access to Mexican information reflects a clash of styles between some Mexican disaster-prevention officials, who take a serious but sanguine approach to Popo and want to avoid causing a public panic, and Williams, who believes that more public information and preparation is the best way to reduce fears about volcanoes.
Williams says that prior to his recent trip to El Popo's environs, he was assured of full cooperation from Mexican authorities. Then The New York Times published a story that featured Williams and his research and stated that the volcano was "threatening one of the most populous regions on earth." Even though that assessment was not a quote from Williams, "I suddenly found all the doors down there being slammed in my face," he says.
Mexican officials are frank in their rejection of what they consider Williams's sensationalism. "I'm directly responsible for what he calls closed doors," says Mexico's Meli. "I know him to be one of the best [volcanologists], but I don't favor the way he likes to give shocking news to the press without a proper knowledge of what's going on.
"It's [Williams's] view that we are trying to hide how dangerous this volcano is," he adds, "which is why I told him we would not like to participate in anything he would be doing in Mexico."
An unrepentant Williams says he still believes Mexican officials are not getting enough data to carry out the kind of studies that were done on Mt. Pinatubo, and that the Mexican people are not getting the information they should to prepare for a disaster.
"Am I saying Popo is going to produce a Saint-Pierre? No," he says. "But volcanic ash that silently accumulates on a roof is a real danger. Mexicans need to know to sweep it off promptly just as New Yorkers know to shovel snow or people in Los Angeles know what to do in an earthquake."
At 45 miles' distance from Popo, Mexico City could receive ash, as it has historically, but one of the world's largest cities is generally considered to face only a slight risk. Still, more than a half-million people live in areas of great or moderate risk.
The agency Meli runs did place El Popo's environs on a "yellow alert" after a series of small explosions in December 1994, and since then public money has helped to replace weak roofs, and practice evacuations have been stepped up.
Popo's highest reaches were also declared closed to the public - although five climbers entered the restricted area and were killed by a relatively small explosion April 30. Despite that explosion, formation of the dome continues.
In Puebla state, taking in El Popo's eastern slopes, an evacuation zone has been delineated to include 307 communities and 400,000 people.
"Our goal is to establish a [disaster relief] committee in each of those 307 towns," says Alejandro Rivera Dominguez, director of the Center for Disaster Prevention at the Autonomous University of Puebla, a leading force in Puebla state's disaster response organization.
Mr. Rivera says that threatened communities should be closely involved in disaster-planning efforts, and he strongly supports the idea that the public is better off the better informed it is. "People need to understand that in the case of a major eruption, they have very little time to make the decision to leave," he says.
Still, many people around El Popo resist the idea that a cataclysm could befall them. During the small eruptions of December 1994, the government did evacuate more than 50,000 people from Popo's slopes. When nothing more threatening occurred, many of the evacuees responded angrily that their lives had been disrupted "for nothing."
The small town of Santiago Xalitzintla, sitting inside El Popo's high-risk zone, is a case in point. Rivera notes that in the event of a "pyroclastic flow" of hot lava and rocks, hot ash, gases, and rocks could sweep over the town in about 20 minutes.
Still, when an evacuation drill was organized in late May, it was mostly Santiago Xalitzintla's schoolchildren, who had recently studied volcanoes in school, who participated. Most of the town's older residents only watched.
Said one to a visiting television reporter: "I've lived here a long time and El Popo has never hurt me, so I don't believe it's going to now."