The surprising heart revisited an early history of the funny current with modern lessons.

40 years ago a single experiment upset a decade of painstaking research on the mechanisms of rhythm generation in the heart. It did so by turning a theory of pacemaker activity upside down. Instead of attributing rhythm to decaying potassium current carrying outward current, it attributed the pacemaker depolarization to the slow activation of a channel conducting sodium ions into cardiac cells. But this was no standard upset of a theory. Like the replacement of Newtonian mechanics by relativity theory, the new theory explained every minute detail of the experimental observations that had established the theory it was replacing. Computational modelling of the heart achieved one of the major successes, the complete mapping of one theory onto another without challenging the correctness of any of the experimental findings. This review details the way in which this transition occurred and draws some important lessons for modelling of biological processes today.