A four-engine Boeing jetliner in a radio-controlled test crashed into the desert today, causing a spectacular burst of flames and smoke and severely singeing the hopes of aviation safety experts that they had found a way to eliminate lethal fires from aircraft fuel.

A huge fireball engulfed three-fourths of the plane immediately after it crashed in a test performed by two government agencies. The plane carried no passengers and no crew.

The fireball lasted about six seconds, but was followed by an outpouring of black smoke that often accompanies landing crashes. Flames from what officials called a "secondary fire" were visible through the smoke.

A radio transmission from the crash site 19 minutes after impact told it all: "We're having some problems putting this fire out," a voice said, repeating what was obvious from a mountaintop four miles away.

Smoke was still rising from the wreckage on Rogers Dry Lake here an hour after the 12:23 p.m. (EST) impact. Two large holes had been burned in the top of the fuselage and the interior of the cabin was described as charred.

"There may not have been much flame, but there was one hell of a fire," said a federal source who visited the wreckage shortly after the crash.

James Woodall, executive assistant to the chief of the Federal Aviation Administration's technical center in Atlantic City, said he was "perplexed" by the secondary fire. Woodall is the man generally regarded as a prime mover of the concept behind the test crash that was supposed to prevent such fires.

"We don't have all the answers, obviously," he said, but they will be developed later because, "even if everything was destroyed out there, we have the data" from telemetry and many onboard experiments that were expected to survive. FAA and National Aeronautics and Space Administration officials said they expect to be analyzing that data for months.

The FAA and NASA sponsored this $11.8 million crash primarily to test an aircraft fuel mixture called "antimisting kerosene." If it had worked as projected, the fuel-fed fireball that almost always accompanies plane crashes would not have attached itself to the airplane, although small fires were expected.

The first release of fuel "did ignite into the giant fireball and went out," Woodall said. "Soon after that there was a secondary flash and it did stick with the airplane."

A day earlier, Transportation Secretary Elizabeth Hanford Dole, who witnessed the crash with about 200 VIPs, aerospace industry officials and 400 journalists, said in an interview that she hoped to issue by next spring a proposed federal rule to require antimisting kerosene.

Today, after the crash, Dole spokeman Mari Maseng said that Dole "has questions now, as we all do. We are waiting to hear the results of the analysis . . . . We all hoped there wouldn't be a fire; there was a fire."

FAA Administrator Donald D. Engen said during the post-crash briefing that it is still his goal to forward a rule to Dole, but that "a lot depends on the initial reduction of data" derived from the crash.

The plane was a Boeing 720 purchased new by the FAA in 1960 to train inspectors. Renamed NASA 833, it was the largest ever flown by remote control. The plane carried 75 dummies and more than 350 electronic sensors plus 12 high-speed cameras in the cabin.

Under ground control, it took off at 12:15 p.m. (EST) and climbed beautifully to 1,200 feet, headed south toward the San Gabriel Mountains and Los Angeles, then circled back toward the crash site. The flight appeared to be smooth until the final seconds, when the plane obviously experienced some alignment difficulties as it glided toward the desert floor in an approximation of a landing approach.

Fitz Fulton, a NASA test pilot who "flew" the plane through an electronic link from a remote control room, said "there was a little oscillation, but it did arrive at the center line" of the planned crash site.

The left wing hit the ground first. The plane canted somewhat to the left, then the right wing apparently struck one of the steel "can openers" placed there to approximate landing light standards and to guarantee that the tanks carrying the antimisting kerosene fuel would be ripped open.

Although FAA officials were clearly disappointed, not everyone was unhappy.

Tom Trip, a spokesman for the Air Transport Association, which represents many of the nation's airlines, said, "The product didn't work as advertised." The airlines, anticipating a regulatory requirement for antimisting kerosene, have been concerned about its potentially high cost because of aircraft modifications that would be required to use the fuel, and about its effectiveness.

James McAbee of ICI Americas Inc., the chemical company that manufactured the fuel additive, called the test "inconclusive." He agreed, however, that the massive fireball millions would see on their television screens makes it "potentially" more difficult to get a federal rule requiring his product.

The antimisting kerosene, he said, "performed as expected, but some other scenarios took over." Officials had hoped the plane would slide along the ground for a greater distance than it did, but plane crashes tend not to follow precise scenarios.

There were were many other experiments on board the aircraft that could provide valuable research information. Those experiments will take many weeks to evaluate. Woodall said he hoped some cameras survived, but those near the cabin's ceiling, he said, probably did not.