The North German Conductivity Anomaly was detected already in the early years of electromagnetic deep sounding. It refers to the reversal of induction arrows (as a graphical representation of the tipper transfer function, the ratio of vertical to horizontal magnetic field variations) at the northern and southern margins of the North German Basin. In spite of the many experiments carried out so far, its origin has remained ambiguous; explanations encompass a deep-crustal or even mantle source as well as the simple response to deepening of sedimentary sequences in the centre of the basin. Here, we report on modelling of new long-period magnetotelluric data collected along two profiles in NE Germany and S Sweden, with one transect crossing the Sorgenfrei–Tornquist Zone as the main boundary between Precambrian Baltica and the Palaeozoic belts of central Europe. With the exception of a few sites probably influenced by 3-D salt domes, the data allow a 2-D analysis. Resolution is reduced for large depths due to a well-conducting, saline aquifer, extending across the entire basin. The main result is that the reversal of induction arrows can largely be explained by the resistivity contrast between crystalline basement and sedimentary basin fill. Beneath Rügen island, a southward dipping conductor is interpreted to reflect an alum shale layer in Middle Cambrian–Lower Ordovician sediments. Beneath the southern basin, a moderately conductive upper mantle is modelled (although not very well resolved) which may reflect the thinning of the lithosphere as implied by seismic tomography. As the main anomalously inductive effect is primarily explained by the basin edges and numerous other anomalies exist inside the basin, we suggest not using the term ‘North German Conductivity Anomaly’ any longer.