MALIBU, Calif. - A lab technician testified yesterday that she taped a do-not-cross line across the floor of her workspace to keep Floyd Landis' observers from interfering with tests of the Tour de France champion's urine samples.

Cynthia Mongongu of the Chatenay-Malabry lab in France talked about her makeshift barricade at Landis' arbitration hearing and said she used it to keep two observers at bay.

She also talked about being "accosted" by an expert observing the testing of Landis' backup "B" samples.

"I needed to be able to concentrate on my work," she said.

Landis' attorneys raised the issue during cross-examination to try to puncture Mongongu's credibility and prove she was biased in the case. They asked why she filed a sworn statement about interference from the Landis observers, but not from J. Thomas Brenna, who was present to observe for the U.S. Anti-Doping Agency.

"It wasn't necessarily Dr. Brenna," said Mongongu, an analytical chemist at the lab. "It's just the whole . . . any kind of group of people who are around me."

The attorneys followed those questions with more about whether Mongongu was the source of leaks to the French newspaper L'Equipe about any positive tests.

"Absolutely not," Mongongu said. She also denied knowing the source of the leaks.

Both sides have decried the number of leaks from the French lab. The Landis camp is trying to use those as part of a larger effort to show a pattern of incompetence, and possible malfeasance, at the lab outside Paris.

Landis is accused of using banned synthetic testosterone during his Tour de France victory last year.

A three-man arbitration panel hearing 9 days of testimony will decide whether to uphold Landis' positive doping test. If it does, he could face a 2-year ban from cycling and become the first person in the 104-year history of the Tour to have his title stripped.

During the Tour, Landis was tested eight times. Of those, one tested positive.

Much of the testimony yesterday was about the backup "B" samples of Landis' seven negative tests that were, at USADA's request, subjected to carbon-isotope ratio testing to look for synthetic testosterone. Four of those seven returned "abnormal testosterone profiles," and the Landis camp is trying to prove that's a result of mishandled tests. *