Coordinated movements are initiated and controlled by trains of action potentials rapidly propagated along motor and sensory myelinated nerve fibres by saltatory conduction. In the mammalian node of Ranvier, action potentials are generated by Na+ channel kinetics, i.e. action potential upstroke and repolarization are solely due to Na+ channel activation and inactivation. Outward current needed for repolarization is provided by unspecific leakage current. K+ currents in rat and human myelinated nerve fibres serve the function of securing action potential propagation (reviewed by Vogel & Schwarz, 1995). A characteristic feature of the mammalian node of Ranvier is the distinct distribution of Na+ and K+ channels. Na+ channels (Nav1.6) are confined to the nodal membrane, whereas fast activating Kv1.1/Kv1.2 channels are located in the juxtaparanodal axolemma. In between both is the paranode representing a high-resistance pathway caused by the tight binding of Schwann cell loops to the axolemma. Due to this high-resistance pathway, juxtaparanodal Kv1 channels are activated in the wake of action potential repolarization, thereby limiting the depolarizing afterpotential and preventing reexcitation and spontaneous activity. Dampening of excitability is also provided by slowly activating KCNQ channels incorporated in the nodal membrane together with Na+ channels. KCNQ channels contribute to the maintenance of a negative resting potential, thereby facilitating recovery from Na+ channel inactivation and restoring excitability (Schwarz et al. 2006). In addition to Kv1 and KCNQ channels, KCa and KATP channels have been detected by recording their single channel currents with patch electrodes on acutely demyelinated rat and human nerve fibres (reviewed by Vogel & Schwarz, 1995). KATP channels are activated by ATP depletion and provide a link between axoplasmic metabolism and excitability. KCa (BK) channels are activated by an increase of the axoplasmic Ca2+ concentration ([Ca2+]i) in the vicinity of BK channels, thereby hyperpolarizing the axonal membrane. This effect would dampen axonal excitability. However, as yet an elevation of [Ca2+]i has never been reported. The missing link is now given by the paper by Zhang & David (2016) in this issue of The Journal of Physiology. The authors performed elegant Ca2+ imaging experiments and provide convincing evidence for the presence of Ca2+ channels in the node of Ranvier of myelinated nerve fibres of the mouse. Zhang and David injected a fluorescent Ca2+ indicator into myelinated motor nerve fibres and detected a transient increase in axoplasmic [Ca2+] around nodes of Ranvier upon stimulation with trains of action potentials at 50 Hz for 5 s. Using selective pharmacology they showed that in the intact node, Ca2+ influx is mediated through T-type Ca2+ channels and the reverse mode of the Na+/Ca2+ exchanger. Astonishingly, stimulation following acute paranodal demyelination led to an additional increase in [Ca2+]i induced by Ca2+ influx through L-type Ca2+ channels normally covered by myelin in the para- or juxtaparanodal axolemma. Although the nodal [Ca2+]i increase is relatively low (∼100 nm) it could influence axonal excitability as well as axo–glial communication. As mentioned above, BK channels are present in the rat and human para- and juxtaparanodal axolemma. These channels could be activated by an increase in the nodal [Ca2+]i, thereby dampening excitability by hyperpolarizing the axon. Since BK channels are normally located near Ca2+ channels, their activation could be facilitated by the larger [Ca2+] around the inner mouth of Ca2+ channels. Nodal [Ca2+]i elevation could also suppress the nodal M-current mediated by KCNQ channels. This effect would depolarize the resting membrane potential, an effect opposite to that induced by BK channels. Another important function of Ca2+ influx could be the stimulation of ATP production to fuel the Na+/K+ pump to restore Na+ and K+ transmembrane gradients. The larger [Ca2+]i elevation recorded in acutely demyelinated axons (∼500 nm after 50 Hz stimulation for 5 s) could be deleterious. [Ca2+]i increase accompanying chronic demyelination could activate intracellular proteases initiating axonal degeneration. In central axons as an early indication of neuronal damage, [Ca2+]i increase in demyelinating lesions has been described in in vivo experiments in a mouse model of experimental autoimmune encephalomyelitis (EAE) (Siffrin et al. 2010). The study by Zhang and David now provides a candidate mechanism, which might become an attractive target for neuroprotection. An axonal [Ca2+] increase could also induce neurodegeneration by activation of Ca2+-dependent ion channels normally not expressed in the axolemma. One example is the new expression of transient receptor potential melastastin 4 (TRPM4) cation channels in central axons damaged by chronic inflammation during EAE. The axoplasmic [Ca2+] increase facilitates TRPM4-dependent ion influx in CNS nerve fibres thereby inducing oncotic axonal swelling and degeneration (Schattling et al. 2014). However, TRPM4 expression has not yet been reported in axons of the peripheral nervous system. All of these and additional scenarios could be experimentally tested in the near future. None declared.