Cellular pH control is important in muscle physiology, and for interpretation of (31)P magnetic resonance spectroscopy (MRS) data. Cellular acidification in exercise results from coupled glycolytic ATP production mitigated by cytosolic buffering, 'consumption' of H(+) by phosphocreatine (PCr) breakdown, and membrane transport processes. Ex vivo methods for cytosolic buffer capacity are vulnerable to artefact, and MRS methods often require assumptions. (31)P MRS of early exercise, when pH increases unopposed by glycolysis, is conceptually simple, but limited in normal muscle by time resolution and signal-to-noise. A therapeutic trial (Martinuzzi A et al. Musc Nerve 37: 350-357, 2007) in McArdle's disease (glycogen phosphorylase deficiency), where pH does not decrease with exercise, offered the opportunity to test (31)P MRS data obtained throughout incremental plantar flexion exercise and recovery in ten McArdle's patients against the simple model of cellular pH control. Changes in pH, [Pi] and [PCr] throughout exercise and recovery were quantitatively consistent with mean +/- SEM buffer capacity of 10 +/- 1 mM/(pH unit), which was not significantly different from the control subjects under the initial-exercise conditions where the comparison could be made. The simple model of cellular acid-base balance therefore gives an adequate account of cellular pH changes during both exercise and recovery in McArdle's disease.