Intracellular pH of in vitro diaphragm preparations was determined following low- (5 Hz, 1.5 min) and high- (75 Hz, 1 min) frequency stimulation, using glass microelectrodes of the liquid membrane type (pH(m)). Results were compared with values obtained by the standard homogenate technique (pH(h)). High- and low-frequency stimulation reduced peak tetanic tension to 21 ± 1 (SE) and 71 ± 2% of initial values, respectively. Peak tetanic tension returned to resting values after 10- to 15-min recovery from high- or low-frequency stimulation. Resting pH(m) was 7.063 ± 0.011 (n = 72), and after fatiguing stimulation declined to values as low as 6.33. During recovery pH(m) significantly increased and by 10 min had returned to prefatigue values. No difference was observed in the recovery of pH(m) between the low- and high-frequency stimulation groups (analysis of variance test, ANOVA), and in both groups pH(m) recovery was highly correlated to the recovery of peak tetanic tension (r = 0.94, P < 0.001). Resting pH(h) was 7.219 ± 0.023 (n = 13), which was significantly higher than the pH(m) value. In contrast to pH(m), intracellular pH(h) was significantly higher during recovery from 75- vs. 5-Hz stimulation (P < 0.05). For both groups pH(h) increased significantly with time and by 10 min returned to prestimulation values. The ANOVA test demonstrated that pH(h) values were significantly higher than pH(m) values during recovery from fatigue. The results from this study support our hypothesis that fatigue from both high- and low-frequency stimulation is at least partially due to the deleterious effects of intracellular acidosis on excitation-contraction coupling.