Since this spring, despite rife opposition from the medical community, the Environmental Protection Agency has quietly begun lifting a 1998 ban on accepting such research. Once the prohibition is gone, which will likely happen next year, chemical companies will have the full support of the federal government to dose healthy young men and women with the latest insecticides, rodenticides, and fungicides.

This marks the second round in a fiery debate over pesticide tests using people. In the late 1990s, a group of doctors and public health advocates noticed that pesticide companies were conducting a growing number of these trials as part of attempts to get government approval. The advocates railed against the EPA and balked at the agency's failure to enforce ethical standards. The "EPA does not routinely require companies who conduct human experiments to . . . follow any ethical protocol," noted a 1998 report from the Environmental Working Group.

Later that year, with criticism mounting, the agency prohibited its offices from using human data in new pesticide registrations. Some companies continued the testing, however, saying it was necessary to determine health risks. But they also preferred that method because they got more favorable readings from dosing people as opposed to lab rats.

The tests appear to defy the very essence of the Hippocratic oath, "First, do no harm." Unlike tests for exploratory vaccines and medicines, pesticide studies offer zero benefits for participants. They're designed to find the level at which concoctions of orange juice and bug spray won't send people crawling toward death, and are considered a glowing success only when nothing happens. Independent researchers say the tests' scientific value is highly suspect.

But there's big money at stake, especially with the EPA considering new restrictions or outright bans on a number of products. On March 31, the Office of Management and Budget, the White House's rule review board, signed off on a rough draft of a new policy that would again allow the EPA to accept the test results.

Doctors, environmentalists, and public health advocates have been fighting the change. When the EPA first took up the idea, medical experts began to pore over a stack of human tests. They found many of the studies were cloaked in claims of valid research but were dominated by practices that belonged in the annals of medical farce. "A reasonable person might conclude that they were specifically designed to fail to show effects of the pesticides," said Dr. Alan Lockwood, a member of Physicians for Social Responsibility and a neurology professor at the State University of New York at Buffalo.

Vermin killers have a nasty history. In 1934, Nazi Germany whipped up the first batch of pesticides—organophosphates, in scientific parlance—for use as a chemical weapon. Although the toxic soup never made it to the front lines, I.G. Farben, the company that manufactured it, found it could be marketed as bug sprays and rodent zappers.

Today, big chemical companies are fans of human research because it encourages less stringent standards. With data from lab animals, the EPA assumes the predicted hazards for humans would be greater by a factor of 10. It's called the "inter-species rule," adopted by Congress to account for potential differences between reactions in, say, a two-year-old child and a mature lab rat. Testing on humans lets a company duck the automatic increase.

That translates directly into several billion dollars for the pesticide industry, which annually sells nearly 4.5 billion pounds of chemicals—at a profit of more than $6 billion. Manufacturers have an outsized financial incentive to push for testing on humans, warned Dr. Lynn Goldman, the EPA's pesticide director under President Clinton and now a professor at the Johns Hopkins Bloomberg School of Public Health. "EPA must of course be mindful at all times of the test sponsors' interests in performing tests and, of course, of the almost overwhelming economic incentives that companies have to find ways to market more of their products," she said in January, testifying before the National Academy of Sciences.

Critics say the companies give sparse attention to decent testing procedures and that nearly every aspect of the testing seems driven by the need to get EPA approval.

Take, for example, a 1999 test conducted by the Lincoln, Nebraska-based MDS Harris Laboratory. A handful of subjects were administered Dow Chemical's chlorpyrifos, a direct descendant of Hitler's nerve agents. MDS Harris had recruited the group of healthy young men and women by assuring them their health would be preserved, and by handing out juicy compensation checks. They were told in consent agreements that low doses of chlorpyrifos "have been shown to improve performance on numerous tests of mental function," implying that the chemical could propel them into a new realm of genius. "The consent process was inadequate, deceptive, or both," Dr. Lockwood said. "This makes it sound like chlorpyrifos is good for you and may make you smarter—a clear deception."

Nevertheless, when none of them died, fainted, or delivered farewell speeches while clutching their hearts in agony, Dow submitted a glowing report of the pesticide to the EPA and eagerly awaited registration approval. Just one year later, on June 8, 2000, the EPA determined that chlorpyrifos, a widely employed pesticide, posed an "unreasonable threat" and said residential uses should be expeditiously restricted.

In another experiment, conducted in 1997 at the Central Toxicology Laboratory, researchers gave oral doses of dichlorvos, a common insecticide, to a group of six young men. When four of them suffered a dangerous drop in vital enzyme levels, they had to withdraw from the test. With only two subjects able to complete the doses, the Central Toxicology Laboratory announced that "no symptoms or adverse effects . . . were reported." They skirted the fact that two-thirds of the participants had to drop out and effectively asserted that the results derived from two people adequately reflected the potential harm to 266 million U.S. citizens.

And there is potential harm. Pesticides eat away at an enzyme called cholinesterase, which plays a key role in all physical movement. It sweeps away chemical debris between nerve cells, allowing those cells to fire up to 1,000 electric impulses to each other every second. Pesticides break down cholinesterase, leaving millions of chemical messages to clog the works. In mild cases, this leads to nausea, sweating, uncontrollable drooling, headaches, and vomiting. In severe cases, it causes muscular tremors, abnormally low blood pressure, loss of bowel functions, slowed heart rates, and even death.

But test groups rarely get that sick. And that's no surprise, considering their size and make-up. They're usually limited to between six and 50 people, typically young and healthy adults who are paid anywhere from $300 to $1,000. The studies are advertised in local newspapers or on college campuses, specifically targeted to attract people from low-income or minority communities.

Pesticide companies insist that trying out their wares on you and your neighbors allows idiosyncratic human reactions to surface. "These safety factors are necessary," said Ray McAllister, vice president for science and regulatory affairs for CropLife America, a lobbying group representing 41 corporations, including Dow, DuPont, and Monsanto. "If we don't know how humans react, then we can't be confident of safety."

The industry is lining the campaign coffers on Capitol Hill. In the five years since the EPA stopped looking at human research, the Center for Responsive Politics reports, companies providing agricultural services and products donated more than $20 million to political campaigns, almost 70 percent of which went to Republicans.