Basic researchers at Leipzig University have solved a puzzle in the evolution of bacterial enzymes. By reconstructing a candidate for a special RNA polymerase as it existed about two billion years ago, they were able to explain a hitherto puzzling property of the corresponding modern enzymes. Unlike their ancestors, they do not work continuously and are thus significantly more effective – these pauses in activity constitute evolutionary progress. The reconstruction of the protein from prehistoric times was made possible thanks to interdisciplinary cooperation between molecular biochemistry and bioinformatics.