WASHINGTON — A vast pool of molten rock in the continental crust that underlies southwestern Washington state could supply magma to three active volcanoes in the Cascade Mountains — Mount St. Helens, Mount Rainier and Mount Adams — according to a study that's causing a stir among scientists.

The study, published Sunday in the magazine Nature Geoscience, concluded that the magma pool among the three mountains could be the "most widespread magma-bearing area of continental crust discovered so far."

Other scientists dismiss the existence of an underground vat of magma covering potentially hundreds of square miles as "far-fetched" and "highly unlikely." Rather than magma heated to 1,300 to 1,400 degrees, some think it could be water.

They also discount speculation that a so-called "supervolcano" such as the one under the Yellowstone National Park area might be beneath the region.

Even so, the study is another piece of the puzzle as scientists try to understand the deep plumbing of volcanoes and, perhaps eventually, learn how to predict their eruptions better.

In the late 1980s, scientists discovered a massive underground electromagnetic anomaly known as the Southern Washington Cascades Conductor. But the two-year study is the first to suggest that it may be the source of magma for Mounts St. Helens, Rainier and Adams.

The three volcanoes are along the edges of the Southern Washington Cascades Conductor, in a rough triangle with Mount St. Helens to the south, Rainier to the north and Adams to the east.

Scientists think that each volcano has its own small magma chamber. A large pool of magma 12 to 15 miles under the region's surface supplies each of the shallower chambers, the new study theorizes.

Among the scientists who study volcanoes, the study has, in fact, touched off a major debate.

Seth Moran, a volcano seismologist with the USGS Cascades Volcano Observatory in Vancouver, Wash., said the most telling evidence that the theory was wrong was the lack of any surface evidence, such as geothermal vents or hot springs, among the mountains that would indicate the presence of a super-heated underground magma pool.

"Such hypotheses sometimes just fade away," Steve Malone, a professor emeritus at the University of Washington's earth and space science department, said. "It absolutely is not a reason at this point to re-evaluate hazard mitigation efforts."