WASHINGTON, Aug. 1 – Unions representing thousands of staff scientists at the Environmental Protection Agency say the agency is bending to political pressure and ignoring sound science in allowing a group of toxic chemicals to be used in agricultural pesticides.

Leaders of several federal employee unions say the chemicals pose serious risks for fetuses, pregnant women, young children and the elderly through food and exposure and should not be approved by Thursday, the Congressional deadline for completing an agency review of thousands of substances in pesticides.

“We are concerned that the agency has not, consistent with its principles of scientific integrity and sound science, adequately summarized or drawn conclusions” about the chemicals, union leaders told the agency administrator, Stephen L. Johnson, in a newly disclosed letter sent May 25.

The leaders also wrote that they believed that under priorities of E.P.A. management, “the concerns of agriculture and the pesticide industry come before our responsibility to protect the health of our nation’s citizens.”

Nine union leaders representing 9,000 agency scientists and other personnel around the country signed the letter. It was given to The New York Times on Tuesday by environmental advocacy organizations working on their behalf in the hope that it would arouse public outcry and increase pressure on the agency to withdraw the chemicals from use.

The chemicals at issue are organophosphates and carbamates, long a matter of controversy over their environmental and health risks. They are in such pesticides as chlorpyrifos, methyl parathion and diazinon.

The advocacy organizations that released the letter, Public Employees for Environmental Responsibility and the Pesticide Action Network, also provided the agency’s response, on June 27, from Susan B. Hazen, acting assistant administrator. Ms. Hazen assured the scientists that her agency was applying proper scientific review for the use of all chemicals in pesticides.

Ms. Hazen did not deny the accusation that industry positions were taken into account. She welcomed information “from all interested parties,” she said.

In an interview on Tuesday, Jim Jones, director of the E.P.A.’s pesticide office, described the scientists’ accusations as inaccurate, saying the agency examines the effects of various chemicals and adjusts recommendations for public use according to what the science dictates.

Risk assessments of the pesticides cited in the unions’ letter, Mr. Jones said, have been “aggressively regulated” through steady reviews of their use over the last six years.

The complaints from agency employees are the latest to come from within federal agencies that accuse the Bush administration of allowing politics or industry pressure to trump science on issues like climate change and stem cell research.

In this case, they also echo concerns raised by the E.P.A. inspector general in January in a report that suggested the agency had not done enough to protect children from exposure to pesticides, which can affect the development of the brain and the nervous system. That investigation was prompted, in part, by published reports of a Florida program in which parents would be paid for letting their children participate in an effort measuring the effects of pesticides in the home. The program was quickly shut down.

The inspector general’s report fueled a growing desire among union leaders to take a more active role in shedding light on what they say is a flawed system.

“More and more, the unions are coming together to confront the agency’s unwillingness to make the appropriate use of science to show risks to public health and the environment,” said William Hirzy, a senior scientist at the environmental agency and a union official.

Despite the agency’s insistence that pesticide regulations follow scientific guidelines, several agency scientists said industry determined how chemicals were regulated.

“It’s how the game is played,” said an E.P.A. specialist involved in the pesticide program who spoke on the condition of anonymity because, he said, critics within the agency often lose choice assignments.

“You go to a meeting, and word comes down that this is an important chemical, this is one we’ve got to save,” he said. “It’s all informal, of course. But it suggests that industry interests are governing the decisions of E.P.A. management. The pesticide program functions as a governmental cover for what is effectively a private industry licensing program.”

Another senior E.P.A. scientist who also spoke on condition of anonymity said the agency often ignored independent scientific studies that contradicted the industry-subsidized study that supported many regulations on pesticides.

She cited a North Carolina researcher who found that chlorpyrifos might have a more damaging effect on developing brains than other studies. “What we heard back from headquarters was, ‘No, he’s wrong,’ ” the scientist said.

“Chemicals like these can be harmful to children in ways we don’t understand yet,” the scientist said. “If there is disagreement, doesn’t that cry out for further research?”

Mr. Jones said the agency had addressed chlorpyrifos in complying with a 10-year Congressional mandate to review 9,741 pesticide ingredients by Thursday.

Work has been completed on 9,637 of them, or 99 percent, he said, and “all are protective of children.”