WASHINGTON -- The leading theory of what doomed the Space Shuttle Columbia will be put to the test Thursday in San Antonio, Tex. when a chunk of foam is shot at the reinforced carbon carbon leading edge of a mock shuttle wing.

The test, to be conducted at the Southwest Research Institute, follows on the heels of a nearly identical test that caused much greater than expected damage to a stronger fiberglass wing panel.

That test, conducted May 29, shot a three-quarters of a kilogram chunk of insulating foam out of a nitrogen gas cannon at 850 kilometers per hour against the forward edge of the test wing, simulating the conditions seen during Columbia Jan. 16 launch.

When the foam hit the wing, the so-called T-seal between the two leading edge wing sections lifted and moved, leaving a 56-centimeter long gap that was as much as a half-centimeter wide in some spots.

Scott Hubbard, director of NASAs Ames Research Center and a member of the Columbia Accident Investigation Board, said the foam also wedged in the crack, permanently deforming the shape of the wing.

Investigators can only speculate at this point whether Columbias wing was deformed in much the same way. Hubbard said the board has not undertaken an analysis to determine whether the deformity observed on the test article May 29 was sufficient to alter the aerodynamics of the wing.

The wing hardware involved in the May 29 test was made of fiberglass, at least twice as resistant to impact damage than the more brittle reinforced carbon-carbon composite that is actually flown on the shuttle.

Hubbard told reporters June 4 that he was surprised by the amount of damage the foam inflicted on the fiberglass test article.

"I thought Oh my God, this is something. This isnt just a light bounce," Hubbard said, "Peoples intuitive sense of physics is sometimes way off I did think Oh my God, this is really an impact. This is really a significant effect."

Although the board has not issued any formal conclusions about the May 29 test, for Hubbard, at least, it gave additional credence to the theory that the foam impact doomed Columbia.

"I think this moves us a lot closer toward saying that foam can do this kind of damage," Hubbard said.

Hubbard said he would brief reporters shortly after Thursdays test on his initial observations of the test shot and impact. A fuller analysis of any damage that may occur will take three-to-five days, he said.

Video of the test firing will be shown on NASA TV and available to news media that same day.