Bada stumbled across the original experiment by accident when a colleague of Miller's mentioned having seen a box of experimental samples in Miller's office. Bada, who inherited Miller's scientific possessions after his death in 2007, found the box — literally labeled "1953-1954 experiments" — in his own office.
Inside it were samples taken by Miller from a device that spewed a concentrated stream of primordial gases over an electrical spark. It was a high-powered variation on the steady-steam apparatus that earned him fame — but unlike that device, it appeared to have produced few amino acids, and was unmentioned in his landmark 1953 Science study, "A Production of Amino Acids Under Possible Primitive Earth Conditions."
But Miller didn't have access to high-performance liquid chromatography, which lets chemists break down and classify samples with once-unthinkable levels of precision. And when Bada's team reanalyzed the disregarded samples, they found no fewer than 22 amino acids, several of which were never seen by Miller in a lifetime of primordial modeling.
Perhaps amino acids first formed when the gases in Miller's device accumulated around active volcanoes, said Bada. "Instead of having global synthesis of organic molecules, you had a lot of little localized factories in the form of these volcanic islands," he said.
Would the molecules formed by Miller in the 1950's, but only recognized in 2008, be prior art as of the 1950's? LBE has raised similar timing issues with the case of buckminsterfullerene, C60.