A University of California professor has accused the Interior Department of suppressing data suggesting that toxic contamination at a California wildlife refuge is moving up the food chain and could pose a threat to endangered species.

Interior officials have declined to make public the results of last year's sampling at the Kesterson National Wildlife Refuge near Los Banos, contending that the information has not been reviewed and is "too tenuous to release for open forum."

The information has found its way into court documents, however, and Berkeley ecology professor Arnold Schultz says the department's refusal to discuss it publicly amounts to a "gag order" blocking access to data on the environmental impact of agricultural drainage.

Schultz is the organizer of a series of scientific gatherings at Berkeley designed to examine the implications of contamination by agricultural drainage in California's irrigated valleys. For the past two years, Interior scientists have attended the symposiums to present the latest information on Kesterson, which is laced with toxic levels of selenium washed from the soil by agriculture waste water.

This year, Interior scientists refused to discuss their most recent findings. "We got basically a rehash of earlier work," Schultz complained in a letter to U.S. Fish and Wildlife Service research director David L. Trauger.

In his response, Trauger told Schultz that Interior withheld the results under a "longstanding policy" that requires "evaluation by two staff scientists, a biometrician, the immediate supervisor, the Fish and Wildlife Service editorial office and the center director."

However, a senior U.S. Fish and Wildlife Service official outlined the results in an affidavit filed last month in federal district court. The affidavit, signed by deputy regional director Joseph R. Blum, said that the latest samples found selenium concentrations from 10 to 1,000 times higher than normal in small rodents captured near Kesterson.

The results are significant because they indicate that the toxic element, originally detected in vegetation and aquatic organisms in the refuge, is working its way up the food chain to terrestrial fauna. According to Blum's affidavit, the findings suggest an increasing risk to birds of prey and higher mammals, including the endangered San Joaquin kit fox, which inhabits the area.

Interior officials were forced to close the Kesterson refuge last year after confirming that selenium was devastating waterfowl and other wildlife. Fish, frogs and other aquatic animals virtually disappeared, and hundreds of birds died or produced seriously deformed offspring.

The situation has thrust the department into an uncomfortable political position as it attempts to reconcile its dual roles as wildlife protector and chief promoter of western irrigation projects.

In an effort to minimize the threat to wildlife while it searches for an alternative solution to irrigation drainage, the Fish and Wildlife Service has conducted a "hazing" program to frighten birds away from the poisoned refuge.

In his letter to Schultz, research director Trauger said that the department had not conducted any studies to determine the effectiveness of the hazing program.

The statement is contradicted in Blum's affidavit, which states that the service has monitored the hazing program at Kesterson and found it "largely unsuccessful in keeping American coots, water birds and tricolored blackbirds a candidate species for the federal endangered species list from using the reservoir during any season."

In the last year, Blum said, 40 percent of all nests at Kesterson contained dead or abnormal embryos or chicks. "In contrast, no dead or abnormal embryos or chicks were observed" at an uncontaminated refuge eight miles southwest of Kesterson, he said.

Reseachers from San Francisco State University found that bird embryo abnormalities are increasing and affecting a broader range of species. Abnormalities in avocets, a long-legged wading bird, were detected for the first time, and the reseachers did not find a single stilt or avocet that lived long enough to fly.