Investigators have begun examining whether the left side of the doomed space shuttle Columbia might have been subjected to extra stress during liftoff because of a steering adjustment designed to counter changeable winds at high altitude.

In 1998 and again on its final, fatal flight, Columbia underwent unusually large steering adjustments, and both times the data showed that the left side was subjected to a sudden, quick increase in force while the right side was largely unaffected, Air Force Maj. Gen John Barry said at a briefing in Houston today.

Investigators have concluded that a breach, or burn through, somewhere in the shuttle's underside on or near the left wing is the most likely explanation for the Columbia's breakup and disintegration over east Texas on Feb. 1. But members of the board investigating the disaster said today that they are still weeks or months away from unraveling the mystery.

They are examining anything out of the ordinary, such as the steering incident and new information about the history of Columbia's giant external fuel tank, that may have contributed to the chain of events that led to the death of seven astronauts.

The steering adjustment incident -- actually a swiveling of the steering nozzles on the shuttle's solid fuel rocket boosters -- occurred 62 seconds after Columbia's Jan. 16 liftoff. About 20 seconds later, debris came off the external tank and, as shown in video of the launch, struck the underside of the left wing near the leading edge. This sequence is a prime focus of the investigation.

The steering maneuver at 62 seconds into the launch was not beyond the shuttle's acceptable design limits, Barry emphasized, but it was considerably more exaggerated than usual.

The investigators said they have commissioned metallurgical studies and spectral analyses of images of the debris impact in hopes of determining whether the material that struck the orbiter during liftoff, believed to be insulating foam, might have included ice or metal, and whether the spray seen coming off the orbiter at that time included heat shield material being jarred loose.

Barry said he and other investigators intend to "follow the foam," just as some of their colleagues said earlier they would "follow the heat," to solve the mystery.

Barry, who heads a team focusing on issues of the shuttle's maintenance and materials, said the shuttle's steep angle of ascent also "might have put added load on the orbiter" and "might have been a contributing factor" to the accident.

Investigators also disclosed that they are exploring whether NASA may have inadvertently damaged the external tank when it connected, disconnected and then reconnected the solid rocket boosters to the tank last year.

According to NASA records, the super lightweight tank was "de-mated" from a set of solid rocket boosters so they could be used in another mission. Accident investigators will determine whether NASA engineers thoroughly inspected the attachment hardware for possible damage during that process.

Another target of further study cited by the investigators will be the mechanics of how and why the foam broke off the tank, a phenomenon that had occurred in two of the past three missions.

As the shuttle investigation enters its sixth week and the investigation board brings on new members and pursues new avenues, prospects are dimming that the panel will be able to wrap up quickly, said its chairman, retired Adm. Harold W. Gehman Jr. While members "are moving along methodically," he said, the precise cause of the shuttle breakup only minutes before its scheduled landing in Florida continues to be "elusive."

Gehman and Barry said that, whatever the cause of the accident, more attention must be paid to the advancing age of shuttles, and to such matters as whether the tough carbon fiber material on the wing's leading edge had become more brittle with time and oxidation, and was therefore more vulnerable to damage from debris than engineers realized.

Although NASA officials note that the shuttles are periodically overhauled and updated, as was the Columbia last year, Gehman said, "We are finding places where original equipment is still on the shuttles."

Next week, the board will resume public hearings in Houston, and investigators will begin interviewing senior NASA officials in Washington to delve into the agency's management, decision-making, culture and budget policy, and what effects they may have had on shuttle safety.

Investigators are continuing to study the exchange of e-mails among shuttle managers, engineers and contractors throughout the Columbia's 16-day mission, members said today. They also want to learn more about NASA's decision to withdraw a request for the Defense Department to use powerful ground-based cameras to photograph the damaged shuttle wing days before the accident.

The board remains keenly interested in finding pieces that apparently fell early in the shuttle's descent over California, Utah and Nevada. The team has recovered 28,286 pieces, which that have been taken to a hangar at the Kennedy Space Center. A total of 1,038 pieces -- including landing gear, heat-shield tiles and an inboard elevon actuator -- have been laid out on a grid representing the underside of the shuttle. Overall, those pieces are 39,300 pounds, or about 19 percent of the shuttle's total weight.

About 4,000 searchers and a dozen aircraft are still looking for wreckage. A Navy team using sonar has identified more than 200 submerged targets in the Toledo Bend reservoir and Lake Nacogdoches, where the shuttle's main engines may have fallen, Gehman said.

Sheila Widnall, a member of the Columbia accident investigation board, displays a weather tracking map that shows the breakup of the shuttle.