[en] The highly selective permeation of ions through biological ion channels can be described and explained in terms of fluctuational dynamics under the influence of powerful electrostatic forces. Hence valence selectivity, e.g. between Ca²⁺ and Na⁺ in calcium and sodium channels, can be described in terms of ionic Coulomb blockade, which gives rise to distinct conduction bands and stop-bands as the fixed negative charge Q_f at the selectivity filter of the channel is varied. This picture accounts successfully for a wide range of conduction phenomena in a diversity of ion channels. A disturbing anomaly, however, is that what appears to be the same electrostatic charge and structure (the so-called EEEE motif) seems to select Na⁺ conduction in bacterial channels but Ca²⁺ conduction in mammalian channels. As a possible resolution of this paradox it is hypothesised that an additional charged protein residue on the permeation path of the mammalian channel increases $|Q_f|$ by e, thereby altering the selectivity from Na⁺ to Ca²⁺. Experiments are proposed that will enable the hypothesis to be tested. (special issue on unsolved problems of noise in physics, biology and technology)