Paul Drude's prediction of nonreciprocal mutual inductance for Tesla transformers

Inductors, transmission lines, and Tesla transformers have been modeled with lumped-element equivalent circuits for over a century. In a well-known paper from 1904, Paul Drude predicts that the mutual inductance for an unloaded Tesla transformer should be nonreciprocal. This historical curiosity is mostly forgotten today, perhaps because it appears incorrect. However, Drude's prediction is shown to be correct for the conditions treated, demonstrating the importance of constraints in deriving equivalent circuits for distributed systems. The predicted nonreciprocity is not fundamental, but instead is an artifact of the misrepresentation of energy by an equivalent circuit. The application to modern equivalent circuits is discussed.