Under the sunlit arch of a plastic greenhouse, an artificial swamp of cattails and elephant grass sprouts from hydroponic troughs, soaking sustenance from a city's sewage.

This could be the waste treatment plant of the future.Picture it: A natural system that uses microbes and marsh plants to transform raw sewage into sparkling clean water. Unlike conventional sewage facilities, which cost a fortune to run, this one makes money. It leaves no nasty sludge. It creates energy. It's so pretty it can be used as a park.

It sounds too good to be true.

But it's already becoming a reality. Simple plant-based sewage treatment systems are being used on a small scale in parts of the South. And researchers are developing more refined and efficient systems they say could replace aging and inefficient conventional plants on a large scale within the next decade.

"We're looking at a whole new generation of waste treatment that's cheaper and much more efficient than anything that's available now, and produces useful resources," said William Jewell, a Cornell University agricultural engineer.

"Our goal is to eliminate pollution at a profit," Jewell said.

Beside Ithaca's new $40 million conventional treatment plant, which settles, aerates and chemically treats about 5 million gallons a day, he has set up a pilot model of his system with the capacity to purify 10,000 gallons of sewage a day - about what 100 people would produce.

The demonstration is funded by the state energy office and the Gas Research Institute, a utility-sponsored group trying to find new sources of natural gas.

"It's leaps and bounds ahead of anything that's been done before, both in terms of energy used and quality of water discharged," said Ron Isaacson, manager of renewable resources for the Chicago-based institute.

In Jewell's two-step system, raw sewage is partially cleaned in a tankful of bacteria, then piped into long, low troughs in a greenhouse, where it provides a "nutrient film" on which plants thrive.

The plants are grown hydroponically - with no soil. Their roots form a dense mat as they soak up the pollutants.

"It's really a solar-powered filter," Jewell said on a recent tour.

In the first 4-foot-wide tank of cattails, the water is murky and smelly. By the time it reaches the end of the 250-foot-long series of tanks, in a few hours, it's clear as a brook and teeming with tiny snails and other creatures.

"These are copepods," said Jewell, scooping up a few flitting, flea-size crustaceans. "Whenever you see them, you have tremendously clean water, like in a stream."

To make the water safe to drink, however, would take further treatment that could be done with a more extensive outdoor hydroponic system, which he plans to build this fall, or with a conventional filtration and chlorination plant.

A bonus to plant-based sewage treatment is that methane, sold as natural gas, is produced by the first-stage bacterial process. More can be made by harvesting the plants and processing them using the same bacteria.

Jewell figures a plant with the capacity to handle the sewage of 10,000 people, about a third the population of Ithaca, could generate $400,000 worth of methane a year. The system could also be used to grow flowers and tree seedlings for city parks or commercial nurseries.

But even without those products, Jewell said, his system would cost far less to build and operate than the plants now in use. He estimates that for a community of 10,000, his system would cost less than $1 million to build, compared to at least $5 million for a conventional plant.

"And the quality of water produced by our system far exceeds that achieved by conventional facilities," said Jewell.

The National Aeronautics and Space Administration has sponsored pioneering plant-based waste treatment projects in towns in California, Louisiana, Mississippi and Alabama.

"These are the wastewater treatment system of the future for small towns, especially since the federal grant money for sewage treatment has dried up in the last few years," said B.C. Wolverton, at NASA's National Space Technology Laboratory in Bay St. Louis, Miss.

Wolverton's artificial marshes, which use floating plants like water hyacinths or rooted ones like bulrushes, have treated all the chemical waste and sewage from the NASA lab for more than 12 years. "We've saved many millions of dollars by using these systems."

Some towns are using Wolverton's sewage marshes as parks, he said. "The settling tanks are underground, and they're growing canna lilies and calla lilies in the channels."

Most of the systems that Wolverton has designed are for towns of fewer than 3,000 people. But an experimental system in operation in San Diego, using water hyacinths to make potable water from sewage, is expected to eventually handle a million gallons a day, he said.

"There's a desperate need to come up with a new source of water in the Southwest," said Wolverton. "This could be an answer."

Jewell's system in central New York differs in several ways from Wolverton's. For one thing, it uses greenhouses and hardy plants, making it feasible for use in northern climes.

"A main drawback to plant-based systems has been the climate limitation," said Jim Basilico, chief of the water and pesticide research division of the Environmental Protection Agency, which funded early stages of Jewell's project. "But if it works in Ithaca, it can work anywhere."

Another advantage of Jewell's system is that because of the type of bacteria it uses it doesn't produce sludge like other hydroponic systems and conventional plants do. In most systems, organic pollutants are broken down by aerobic bacteria, which need oxygen. It costs the nation half a billion dollars a year in electricity to fill that oxygen need by aerating sewage, said Jewell.

"We spend $200 million in New York state alone just to blow bubbles through settling tanks," he said. "The anaerobic alternative doesn't use any energy at all. Instead, it produces energy."

Because they are far more efficient at digesting waste matter, turning 90 percent of it into methane, anaerobes don't produce sludge like the fast-growing and inefficient aerobic bacteria do.

"I've begun to look at methane production as almost a magic process," said Jewell, who is also developing digesters to make methane from garbage, agricultural wastes and crops like sorghum.

Other researchers also are looking into systems that use anaerobic bacteria to turn pollutants into profitable products.