1. The effects of prolonged intermittent fatiguing stimulation were studied on various steps of excitation‐contraction (E‐C) coupling in cut single frog muscle fibres using the triple Vaseline voltage clamp and the fluorescent Ca2+ indicator rhod‐2. 2. There were two phases of changes in amplitude of Ca2+ transients during fatiguing stimulation: first a 5‐10% increase, then a larger decrease. The decrease in amplitude of Ca2+ transients was accompanied by a slowing down of the rate of decay of the transients and by an increase in resting [Ca2+]. 3. A complete recovery of both amplitude and time course of Ca2+ transients as well as of the resting [Ca2+] occurred within 1‐3 min after cessation of fatiguing stimulation. 4. The changes in Ca2+ release signals during fatiguing stimulation were accompanied by decreases in the amplitude and the rate of decay of the action potentials as well as by a decrease in resting potential. However, these alterations are not likely to contribute to fatigue significantly, since fibres stimulated under voltage‐clamp conditions, when the T‐tubule voltage sensor is activated directly by applied voltage steps, showed similar fatiguability to fibres stimulated by action potentials under current‐clamp conditions. 5. Simultaneous measurements of intramembrane charge movement and [Ca2+] revealed that the decrease in sarcoplasmic reticulum (SR) Ca2+ release during fatiguing stimulation is not accompanied by any significant change in charge movement. 6. These results suggest that fatigue caused by repeated tetanic stimulation develops primarily at the level of SR Ca2+ release with only small possible additional effects at the level of membrane excitability and action potential propagation along the surface/T‐tubule membrane. The T‐tubule voltage sensor with this type of stimulation is virtually fatigue resistant.