H+ ions are generated rapidly when muscles are maximally activated. This results in an intracellular proton load. Typical proton loads in active muscles reach a level of 20-25 mumol X g-1, resulting in a fall in intracellular pH of 0.3-0.5 units in mammalian muscle and 0.6-0.8 units in frog muscle. In isolated frog muscles stimulated to fatigue a proton load of this magnitude is developed, and at the same time maximum isometric force is suppressed by 70-80%. Proton loss is slowed when external pH is kept low. This is paralleled by a slow recovery of contractile tension and seems to support the idea that suppression results from intracellular acidosis. Nonfatigued muscles subjected to similar intracellular proton loads by high CO2 levels show a suppression of maximal tension by only about 30%. This indicates that only a part of the suppression during fatigue is normally due to the direct effect of intracellular acidosis. Further evidence for a component of fatigue that is not due to intracellular acidosis is provided by the fact that some muscle preparations (rat diaphragm) can be fatigued with very little lactate accumulation and very low proton loads. Even under these conditions, a low external pH (6.2) can slow recovery of tension development 10-fold compared with normal pH (7.4). We must conclude that there are at least two components to fatigue. One, due to a direct effect of intracellular acidosis, acting directly on the myofibrils, accounts for a part of the suppression of contractile force. A second, which in many cases may be the major component, is not dependent on intracellular acidosis. This component seems to be due to a change of state in one or more of the steps of the excitation-contraction coupling process. Reversal of this state is sensitive to external pH which suggests that this component is accessible from the outside of the cell.