London, July 21 : A BBC investigation has found that there are serious question marks over a key drug test just two weeks before the start of the Beijing Olympics.
The BBC has seen indications that labs are classing positive tests for the blood-boosting drug EPO as negatives.
Some samples have been described as suspicious - giving rise to fears that no action will be taken against cheats.
One sport drug expert told the BBC that many of the finalists in Olympic endurance events would be using EPO.
"Copycat" versions of the drug are available on the internet for as little as - and according to experts are often undetectable.
Although a test was introduced to detect recombinant EPO (erythropoietin) at the Sydney Olympic Games in 2000, a growing number of athletes were soon challenging the results in the courts.
Several, like US sprinter Marion Jones, had their first sample test positive, but were cleared on the second or B test.
In response, in 2004, the World Anti-Doping Agency (WADA) tightened the criteria by which an EPO positive could be declared.
As a result, the number of legal challenges fell.
But according to experts that BBC has spoken to, athletes continued to abuse EPO. As they became more adept at medicating themselves, so the number of positives declined.
Many scientists blame the anti-doping laboratories for a loss of nerve in the face of continuing EPO use.
According to Dr Rasmus Damsgaard, who runs the anti-doping programme for the International Ski Federation and for the Astana Cycling team, he has clear evidence that positive EPO tests are being declared as negative or suspicious.
Earlier this year, he sent five samples from skiers to a WADA lab for analysis. They all came back negative.
But when Dr Damsgaard demanded the gels or electronic printouts on which the determination of guilt or innocence was made, he was astonished to see what he believed was clear evidence of EPO use.
"It was very obvious that the gels were very un-natural or very different from natural distributions. But I also saw that they were declared negative because they didn't fulfil the WADA criteria of a positive test; although they looked suspicious and had no natural bands at all, they were still declared negative," he said.
Dr Damsgaard believes that there are many more such samples in WADA labs.
"From a little work with a lot of blood profiles, I found maybe five positives. I wonder that maybe hundreds, maybe even thousands of EPO positive samples are lying around in WADA-accredited labs," he said.