A Frankensteinesque contraption of glass bulbs and crackling electrodes has produced yet another revelation about the origin of life.

The results suggest that Earth's early atmosphere could have produced chemicals necessary for life—contradicting the view that life's building blocks had to come from comets and meteors. "Maybe we're over-optimistic, but I think this is a paradigm shift," says chemist Jeffrey Bada, whose team performed the experiment at the Scripps Institution of Oceanography in La Jolla, Calif.

Bada was revisiting the famous experiment first done by his mentor, chemist Stanley Miller, at the University of Chicago in 1953. Miller, along with his colleague Harold Urey, used a sparking device to mimic a lightning storm on early Earth. Their experiment produced a brown broth rich in amino acids, the building blocks of proteins. The disclosure made the pages of national magazines and showed that theories about the origin of life could actually be tested in the laboratory.

But the Miller-Urey results were later questioned: It turns out that the gases he used (a reactive mixture of methane and ammonia) did not exist in large amounts on early Earth. Scientists now believe the primeval atmosphere contained an inert mix of carbon dioxide and nitrogen—a change that made a world of difference.

When Miller repeated the experiment using the correct combo in 1983, the brown broth failed to materialize. Instead, the mix created a colorless brew, containing few amino acids. It seemed to refute a long-cherished icon of evolution—and creationists quickly seized on it as supposed evidence of evolution's wobbly foundations.

But Bada's repeat of the experiment—armed with a new insight—seems likely to turn the tables once again.

Bada discovered that the reactions were producing chemicals called nitrites, which destroy amino acids as quickly as they form. They were also turning the water acidic—which prevents amino acids from forming. Yet primitive Earth would have contained iron and carbonate minerals that neutralized nitrites and acids. So Bada added chemicals to the experiment to duplicate these functions. When he reran it, he still got the same watery liquid as Miller did in 1983, but this time it was chock-full of amino acids. Bada presented his results this week at the American Chemical Society annual meeting in Chicago.

"It's important work," says Christopher McKay, a planetary scientist at NASA Ames Research Center in Moffett Field, Calif. "This is a move toward more realism in terms of what the conditions were on early Earth."

Most researchers believe that the origin of life depended heavily on chemicals delivered to Earth by comets and meteorites. But if the new work holds up, it could tilt that equation, says Christopher Chyba, an astrobiologist at Princeton University. "That would be a terrific result for understanding the origin of life," he says, "and for understanding the prospects for life elsewhere."

But James Ferris, a prebiotic chemist at Rensselaer Polytechnic Institute in Troy, N.Y., doubts that atmospheric electricity could have been the only source of organic molecules. "You get a fair amount of amino acids," he says. "What you don't get are things like building blocks of nucleic acids." Meteors, comets or primordial ponds of hydrogen cyanide would still need to provide those molecules.

Bada's experiment could also have implications for life on Mars, because the Red Planet may have been swaddled in nitrogen and carbon dioxide early in its life. Bada intends to test this extrapolation by doing experiments with lower-pressure mixes of those gases.

Chyba is cautious: "We don't know," he says, "whether Mars really ever had that atmosphere." That's because Mars today has carbon dioxide, but hardly any nitrogen—which is also needed for making amino acids. Some scientists suspect that nitrogen gas existed on Mars, but was blasted away by asteroid impacts billions of years ago.