The company that lost the contract to build the Hubble Space Telescope's light-gathering mirrors included a $10 million test in its bid that would have caught the gross flaw that now cripples the orbiting observatory, according to congressional testimony yesterday.

Officials of the National Aeronautics and Space Administration have said such a test was not done because it would have cost "hundreds of millions of dollars."

Sen. Albert Gore Jr. (D-Tenn.) said Eastman Kodak Co., which offered to build and test the Hubble mirrors, lost the contract to Perkin-Elmer Corp., which manufactured but did not test Hubble's two mirrors as a complete unit. Perkin-Elmer instead tested the mirrors individually, but never assembled them into a fully functioning telescope in an "end-to-end" or "all-up" test.

The military's spy satellites, which resemble the $1.5 billion Hubble telescope and have been built by Perkin-Elmer and Eastman Kodak, routinely are given end-to-end tests before launch.

Such an examination should have been done and would have provided "a reality check," Gore said, chairing a hearing of the Senate Commerce, Science and Transportation subcommittee on science, technology and space. He cited Eastman Kodak's 1977 bid as evidence, but did not make the document public.

"In retrospect, perhaps an all-up test could have been done, should have been done," agreed NASA deputy administrator James R. Thompson Jr. He testified before Gore cited the Eastman Kodak bid.

NASA associate administrator Lennard A. Fisk told Gore, "We would have had to build the world's greatest testing facility to test the world's greatest mirrors."

Other optical experts said a less expensive end-to-end test could have been performed and would have detected any major flaw in either of Hubble's two mirrors, which were designed to gather and focus starlight on the back of the telescope at a tiny spot where cameras and sensors sort out the images and send them back to Earth.

Because of the flaw, the images are slightly out of focus and the telescope's scientific usefulness has been curbed.

"The initial idea not to test end-to-end probably was not a good idea. However, it was an inexpensive idea at the time," said Bertran Bulkin, director of scientific space programs at Lockheed Missiles and Space Co., which built the Hubble with Perkin-Elmer.