Studies performed by NASA engineers during the 1990s raised the possibility that tiny pinholes on the space shuttle orbiters' wings could be enlarged by hot gasses during re-entry, but concluded that the problem was unlikely to endanger the spacecraft or their crews.

Yet with increasing focus on the leading edge of Columbia's left wing in the days since the spacecraft broke up over Texas, interest in the pinholes may be renewed.

NASA officials declined to say Thursday whether they considered pinholes a possible cause of the Columbia accident. Engineers who have studied the pinholes for NASA and its contractors said they could not discuss the shuttle.

The microscopic pits were first discovered in July 1992, during inspections performed after Columbia's 12th flight. Inspections of the other orbiters also found the holes developing in the silicon carbide coating that protects the heat-resistant material on the leading edge of each orbiter's wings.

The holes generally developed after 10 to 15 flights, but were "worse" on Columbia's wings, according to a July 1999 presentation made by Donald M. Curry of NASA and David W. Johnson of Lockheed Martin.

Though no larger in diameter than a human hair, the holes were of concern because of the 3,000-degree temperatures that part of the shuttle must withstand during atmospheric re-entry.

"The chemical reactions which lead to pinhole formation are expected to occur when the shuttle wing leading edges are hottest — during launch and re-entry," NASA engineer Nathan S. Jacobson wrote in a 1999 paper.

According to NASA technical reports, the troublesome pinholes appear to have been caused in part by tiny flakes of paint peeling off the tower that holds the shuttle and its booster rocket before launch. Experiments showed that zinc in the paint could corrode the silicon carbide coating that protects the heat-shielding material on the leading edge of the shuttle's wing.

The report also concluded that the corrosive salt air at Cape Canaveral may also contribute to pinhole formation. Columbia sat on the pad for 39 days before its final launch, not an atypically long time.

Columbia's last flight was its second since a refurbishment performed in late 1999. Between missions, crews routinely inspect and repair any damage to the shuttle's heat-resistant skin.

Columbia came apart at the point during its descent through the atmosphere when the wings' leading edges can approach 3,000 degrees Fahrenheit, the highest temperatures a shuttle must endure in flight.

As NASA's investigation into the Feb. 1 accident proceeds, attention has centered on the leading edge of the shuttle's left wing. Sensors embedded in the wing failed about eight minutes before the shuttle broke up. Soon afterward, the craft's automatic pilot system began compensating for increased drag on the left side. About that time, a camera at Kirtland Air Force Base in New Mexico snapped a photograph of the shuttle that appears to show damage to the left wing's leading edge.

"I can't really make a judgment on what the image actually shows," NASA Deputy Associate Administrator Mike Kostelnik said Friday. "But it is during the time period when we're getting these anomalies."