1. The performance of skeletal muscle during repetitive stimulation may be limited by the development of an intracellular acidosis due to lactic acid accumulation. To study this, we have measured the intracellular pH (pH(i)) with the fluorescent indicator BCECF (2',7'-bis(carboxyethyl)-5(6)-carboxyfluorescein) during fatigue produced by repeated, short tetani in intact, single fibres isolated from the mouse flexor brevis muscle. 2. The pH(i) at rest was 7.33 +/- 0.02 (mean +/- S.E.M., n = 29, 22-degrees-C). During fatiguing stimulation pH(i) initially went alkaline by about 0.03 units (maximum alkalinization after about ten tetani). Thereafter pH(i) declined slowly and at the end of fatiguing stimulation (tetanic tension reduced to 30% of the original; 0.3 P(o), pH(i) was only 0.063 +/- 0.011 units (n = 14) more acid than in control. 3. We considered three possible causes of acidosis being so small in fatigue: (i) a high oxidative capacity so that fatigue occurs without marked production of lactic acid; (ii) an effective transport of H+ equivalents out of the fibres; (iii) a high intracellular buffer power. 4. The oxidative metabolism was inhibited by 2 mM-cyanide in three fibres. After being exposed to cyanide for 5 min without stimulation, the tetanic tension was reduced to about 0.9 P(o) and pH(i) was alkaline by about 0.1 units. The fibres fatigued faster in cyanide and the pH(i) decline in fatigue was more than twice as large as that under control conditions. 5. Inhibition of Na+-H+ exchange with amiloride resulted in a slow acidification of rested fibres; resting pH(i) was not affected by either inhibition of HCO3--Cl- exchange with DIDS (4,4'-diisothiocyanatostilbene-2,2'-disulphonic acid) or inhibition of the lactate transporter with cinnamate. 6. Fibres fatigued in cinnamate displayed a markedly larger acidification (approximately 0.4 pH units) and tension fell more rapidly than under control conditions; inhibition of Na+-H+ and HCO3--Cl- exchange did not have any significant effect on fatigue. 7. The intracellular buffer power, assessed by exposing fibres to the weak base trimethylamine, was about 15 mM (pH unit)-1 in a HEPES-buffered solution (non-CO2 or intrinsic buffer power) and about 33 mM (pH unit)-1 in a bicarbonate-buffered solution. Somewhat higher values of the intrinsic buffer power was obtained from changes of the partial pressure of CO2 (P(CO2)) of the bath solution. Application of lactate or butyrate frequently gave an infinite buffer power, which indicates that powerful pH-regulating mechanisms operate in these cases. 8. The dependence of tetanic tension on pH(i) of rested fibres was studied by changing the P(CO2) of the bath solution. The tension was found to change with a slope of about 0.33 P(o) (pH unit)-1). 9. The slowing of relaxation which occurs in fatigue is often attributed to acidosis. In agreement with this, acidification of rested fibres resulted in a marked slowing of relaxation. 10. Relaxation was slower after about ten fatiguing tetani, i.e. at a time when pH(i) was increased. Thereafter relaxation slowed further and the greatest slowing was observed in the most acidified fibres. Thus, the slowing of relaxation in fatigue was due to both pH-independent and pH-dependent processes. 11. In conclusion, the acidosis in fatigue was small and this was mainly because of an effective extrusion of H+ ions by the lactate transporter. Thus neither the tension decline nor the slowing of relaxation in fatigue could be explained by an intracellular acidification. If the lactate transporter was inhibited, the acidification during fatiguing stimulation became larger and probably contributed to fatigue.