ASHINGTON, Sept. 19 — Seeking to counter assertions that the nation's nuclear plants are vulnerable to attacks like the one on the World Trade Center, 19 prominent nuclear experts have concluded that a reactor containment building could easily withstand the force of a jetliner crash.

But the federal laboratory that conducted a major test cited by the experts says its experiment was not meant to demonstrate anything about reactors' structural soundness.

The 19 experts, many of them retired, work or worked at universities or companies that build or operate reactors. In an article on Friday in the journal Science, they dismiss fears voiced by opponents of nuclear power that the nation's reactors are vulnerable to a terrorist attack.

"We read that airplanes can fly through the reinforced, steel-lined 1.5-meter-thick concrete walls surrounding a nuclear reactor," the article says, "and inevitably cause a meltdown resulting in `tens of thousands of deaths' and `make a huge area uninhabitable for centuries,' to quote some recent stories." But, they add, "no airplane regardless of size, can fly through such a wall."

The article says the scenario "was actually tested in 1988 by mounting an unmanned plane on rails and `flying' it at 215 meters per second (about 480 m.p.h.) into a test wall." The engines penetrated only about two inches and the fuselage even less, according to the article.

But the relevance of the test, conducted at Sandia National Laboratories, has long been in dispute. People who opposed nuclear power before Sept. 11 pointed out that the test wall moved several feet; the movement reduced the damage by absorbing some of the force of impact.

At Sandia, a spokesman, John German, said the point of the test was to move the wall, as a way to measure the impact forces. The test was sponsored by the Muto Institute of Structural Mechanics Inc., of Tokyo, as a preliminary step in building a computer model of such impacts, but the Japanese decided not to sponsor the next step, Mr. German said.

Asked if it showed that a plane could not penetrate a dome, he said, "We've been trying like heck to shoot down this rumor."

Mr. German said: "That test was designed to measure the impact force of a fighter jet. But the wall was not being tested. No structure was being tested."

The nuclear experts contend that the test makes their point nevertheless. The opponents of nuclear power have argued that the plane in the Sandia test, an F-4 Phantom, weighs far less than a jumbo jet.

But James Muckerheide, a nuclear engineer who is the co-director of the Center for Nuclear Technology and Society at Worcester Polytechnic Institute, on whose work the authors relied, said in an e-mail response to a reporter's question that penetrating a reactor containment building would take far more than an airliner. Compared with the F-4, Mr. Muckerheide said, "a large passenger aircraft is a slow, empty, tin can."

"The mass of the aircraft can put a heavy compression load on the containment structure," he said, "but it has negligible penetrating ability."

The containment building can withstand huge compression loads, he argued. The fact that the block in the Sandia test moved had a trivial effect, Mr. Muckerheide said.

Whether a containment building is the soft spot of a nuclear plant is also not clear. Most of the radioactivity in a power plant is in the spent fuel pool, which, critics note, is usually in a building that is far less sturdy.

The Nuclear Regulatory Commission is conducting an engineering analysis of the vulnerability of power plants to aircraft attack, Sue Gagner, an agency spokeswoman, said. "If warranted by the ongoing detailed analysis, we will consider changes," Ms. Gagner said.

Articles in Science, like those in many scientific journals, are reviewed before publication by experts not connected with the authors. But the magazine's editor in chief, Donald Kennedy, said that if there was a difference between the authors and the group that performed the experiment, "they're going to thrash it out in our letters column, and we'll let them do it."

The magazine is published by the American Association for the Advancement of Science.