A reanalysis of Stanley Miller's classic 1953 experiment into the origins of life's biochemical building blocks shows that his results were even better than he thought they were. Karen Hopkin reports