This time of year, a lush canopy of green spills down the steep embankments near the headwaters of the South River. Here, where it emerges from a marsh of tall grasses, the river is a shadow of the broad channel that spills into the Chesapeake Bay downstream.

From this seemingly pristine stretch of the South River has emerged something of a mystery, a puzzle that may speak to the health of the waterway itself. Researchers have pulled from these waters brown bullhead catfish with lesions around their mouths. In at least one case, the abnormality was so extreme that it resulted in what appear to be cartoonishly oversize lips.

The U.S. Fish and Wildlife Service has sent 33 catfish -- not all of which showed the lesions -- to a lab for analysis. "Intuitively, you see something like that and you just think something's wrong," said Drew Koslow, a marine scientist who was hired as river keeper last year by the nonprofit South River Federation, which works to protect the health of the river. "There's something that's causing it, and we're at the point of trying to find out what that is."

Officials from the federal agency said little will be known until the testing is complete. "Right now all we can actually say is we caught fish from the river and they have some kind of lip thing," said Valerie Fellows, a spokeswoman for the Fish and Wildlife Service. "We can't even say that it's a tumor . . . we can't say anything until we have the analysis back."

The state Department of Natural Resources collected more than two dozen catfish from the area, too. Test results determined that minor lesions on five of those fish were not cancerous.

Bullhead catfish have been analyzed by scientists for decades as an indicator of the health of river systems, mainly because the fish are bottom-feeders and burrow into the mud during the winter, increasing their exposure to sediment pollution, and because they tend not to travel widely.

It was a study of brown bullheads that led scientists to conclude last year that the rate of cancerous tumors in fish in the Anacostia River was as high as ever documented in an American river, and that the tumors were linked to pollution caused by vehicle emissions and runoff.

Fifty to 68 percent of mature brown bullhead catfish collected from three parts of the river in Washington had liver tumors, most of which were cancerous, the study found. In addition, 13 to 23 percent of the bullheads had skin tumors. The study linked the liver tumors to contaminants that often come from fossil fuels, most commonly in the form of vehicle emissions and runoff.

On the South River, nets used by Fish and Wildlife workers conducting a yellow perch survey earlier this year trapped large numbers of catfish. Some of those catfish had what Steve Minkkinen, a scientist involved in the survey, called "anomalies."

Because a sample of brown bullheads was readily available, Minkkinen called a colleague, Fred Pinkney. Pinkney, who also organized the study on the Anacostia, has a long-standing interest in catfish.

A senior biologist in the federal agency's environmental contaminants program, Pinkney said he and his team went out on the river when the perch nets were being emptied in March. They randomly selected 30 mature brown bullheads to submit for further study, he said. On some of the fish, he said, he noticed "lumps and bumps as they were coming out of the net."

"I said, 'If anyone sees anything that is just extraordinary, put it in a separate bucket and I will have those diagnosed.'"

Three bullheads fit that description, Pinkney said -- all from the same net, about a quarter-mile downriver from the Route 50 bridge.

Mark Matsche, a scientist with the state Department of Natural Resources, headed out to the same spot a couple of weeks after learning of the find. Matsche said that while some of the more than two dozen catfish he pulled from the river had lesions, none showed abnormalities as extreme as Pinkney's three.

Tests later showed that "two slightly raised lesions were a thickened, normal area of skin" that are considered "very minor," Matsche said. Tests on three other fish showed that ulcers on those fish were "minor and incidental."

Matsche noted that benign tumors appear in some fish throughout the bay. "If we find one or two around the bay, it's nothing to be worried about," he said. "It's only when you come across a larger-scale event at one location that you need to take a look at it more closely."

He said that depending on the type of lesion, possible causes could include pollutants, bacterial or viral infection, or parasites, some of which might occur naturally.

Beth McGee, senior water quality scientist at the Chesapeake Bay Foundation, said the lesions are "clearly an indication that the system is not in balance."

"But what the pollution, what the stressors responsible are, it's too soon to tell," she said.

The state Department of the Environment recommends that women and children not eat more than one serving each month of catfish from any of the western shore rivers south of Baltimore Harbor, a category that includes the South River.

Richard McIntire, a spokesman for the department, said headwaters generally are notorious for containing concentrated amounts of various contaminants. "We are interested, but one of the overwhelming thoughts is that you're going to find these types of conditions, or the potential for these types of conditions to exist, in any urban river," he said. "It's just at what level."

Koslow, the river keeper, said he was heartened by the results of the tests on the fish collected by the Department of Natural Resources, but would await results from Fish and Wildlife before setting his concern aside. He considers the bottom sediment at the headwater, where toxins would accumulate, to be pretty much dead. Basic life-forms, such as clams and worms, couldn't survive in it, he said.

On a recent morning, not far from where the bullheads were collected, Koslow, out on his skiff, conducted a regular battery of water-quality tests. The water is only four feet deep, but when he lowers a monitor to the bottom, it detects almost no dissolved oxygen -- one-eighth of the amount most fish require, he says.

The sky above is blue, and wind rustles through a blanket of trees that cover the banks of the river. Koslow's forehead scrunches up as he takes the reading. "Nothing can live in that," he says.

An angler pulls a catfish from the Anacostia River in Washington, where scientists last year reported a high rate of cancerous tumors among the catfish population. The bottom-feeding fish are considered an indicator of a river's health. Doug Koslow, keeper of the South River, sets up monitoring equipment to check the river's headwaters. This is one of the monitoring devices Koslow uses.