Missouri biologist Frederick vom Saal and his team exposed the dangers of bisphenol A — and earned the wrath of the plastic industry.

Five EPA panels have met to advise the EPA on chemical screening. Nagel was invited to be part of the second EPA panel for the Endocrine Disruptor Screening Program from 2001-03. She listened in frustration as fellow panelists dismissed the importance of measuring low-dose effects of hormones such as bisphenol A and failed to recognize the distinction between testing the chemicals on fetal animals versus adult animals.

Nagel says that on such panels, industry representatives are able to dilute the process because of their tenacity in insisting that there is controversy in science where none really exists. The panels also try not to include particularly outspoken scientists. "They deselect people like Fred and me," she says. "They want people who are trying to bend over backwards not to have an opinion. But industry is never not trying to further their cause."

Nagel was not invited to join later panels. The screening program's participants have since decided to conduct hormone testing on a breed of rat that many endocrinologists consider a bad candidate for such study. The panelists also chose to feed the rats a type of chow that is high in soy, which contains enough natural estrogens to disrupt the study findings. Perhaps worst of all, the panel concluded that it was open to allowing chemical companies to tailor the tests to their liking.

"That's not uncommon at all," Nagel says. "Traditionally, these panels will say, 'Here's the guidelines for the tests, but you can choose one of these four ways [to carry them out]."

The flaws in this effort were widely reported. For instance, The Dallas Morning News revealed that the EPA had solicited advice on what breed of rat to use from a toxicologist who works for a company contracted by the chemical industry.

Last month, a House committee opened an investigation on rumored conflicts of interest in the scientific panels that advise the EPA.

But the EPA isn't the only federal agency battling conflict-of-interest accusations. Another federal agency, the Center for the Evaluation of Risks to Human Reproduction, which is part of the U.S. National Toxicology Program, convened an expert panel on bisphenol A in 1997. Naturally, vom Saal wanted to be part of the process, but he was barred. No scientists were allowed to be part of the CERHR panel if they'd published a study on bisphenol A.

The American Chemistry Council, on its Web site, explains why it agrees with this decision: "Scientists who have conducted significant amounts of research or have otherwise taken a position on the chemical of interest, either favorable or unfavorable, are generally excluded from participation on the panel to avoid conflicts of interest or bias."

Vom Saal calls this position "absolutely based on complete ignorance of the way science works. What we love about what we do is it's absolutely self-correcting. Unlike practically any other field, if you publish something important and it's wrong ... it's critical that other scientists point that out."

Because more than 200 studies have confirmed vom Saal's initial hypothesis on bisphenol A, "it's not a debated hypothesis anymore," he says. "And the idea that we proposed that and published that makes us biased — when 200 independent groups have confirmed it — there's something very, very seriously wrong with that message."

So vom Saal flew to Virginia on his own dime to attend March 2007 meetings of the CERHR panel on bisphenol A, joining other scientists and members of the press in the audience. Like Nagel, he was disgusted by the basic scientific misunderstandings he was hearing from members of the appointed panel.

"It was like listening to a high school debate or something," vom Saal says of the panel's arguments. "All of the critical issues that Wade, Susan and I and other scientists working in the field have been raising ... they weren't discussing that. They were saying, 'Humans are exposed to bisphenol A orally, and it's completely metabolized.' What they were sitting there saying is completely contradicted by a large scientific literature."

The panel threw out studies by vom Saal and many others because of vom Saal's methodology. Panelists disagreed with any study that used injection as a method to introduce bisphenol A to mice, claiming that because humans absorb bisphenol A through drinking water and food, the only acceptable test method was to feed it to mice. But bisphenol A is most dangerous to fetuses, which absorb the chemical through their mothers' bodies rather than by eating it, vom Saal and other scientists argued in letters they submitted as public comments on the process. And because bisphenol A is present in air and water, digestion isn't the only means by which people come into contact with it. Bisphenol A can enter through the lungs and even through the skin (the same way that birth control does in patch form).