After a bout of intense exercise, especially in untrained persons recovery of muscle force is often slow. Force depression is much more marked at low frequencies of stimulation than at high frequencies ("low-frequency fatigue') and recovery can take more than 1 day. Delayed force recovery is also seen in single muscle fibres from frog and mouse after fatigue induced by repeated, brief contractions. Evidence from our own and other laboratories indicates that the impairment is unlikely to result from metabolic changes and points to a defect in excitation-contraction coupling. We demonstrate that the likely site of failure is in the coupling between t-tubule depolarization and release of Ca2+ from the SR. The causative agent appears to be a localized increase in cytoplasmic Ca2+ which initiates some disruptive process, which can, however, be fully reversed, albeit slowly. Our experimental evidence does not support the involvement of Ca(2+)-activated proteases. Attempts to clarify the possible role of Ca(2+)-activated lipases (phospholipase A2) and Ca2+/calmodulin have been hampered by side-effects of available inhibitors. Efforts to clarify how Ca2+ exerts its effects are continuing.