We performed in vivo 31P NMR spectroscopic studies of human brain on a 4.1 T whole-body NMR system. Based on a control group of 20 healthy volunteers, the normal pH(i) was 7.05 (SD, 0.06; SEM, 0.01) in the left temporal lobe and 7.04 (SD, 0.04; SEM, 0.01) in the right temporal lobe. We also studied a patient group consisting of 13 individuals with unilateral temporal lobe epilepsy. The mean pH(i) was 7.02 (SD, 0.04; SEM, 0.01) in the ipsilateral lobe and 7.02 (SD, 0.05; SEM, 0.01) in the contralateral lobe. These results clearly show that no statistically significant difference in pH(i) is observed between the two lobes, either in normal controls or in patients. Also, no significant pH(i) difference exists between the control group and the patient group. Lateralization in each of the 13 patients with unilateral epilepsy, based on their individual pH(i) difference between the ipsilateral lobe and contralateral lobe (ΔpH(i)), showed, that three patients were nondiagnostic cases because their ΔpH(i)s were not significantly different from zero (≤0.02), five patients showed small ΔpH(i)s consistent with their clinical lateralization, whereas the remaining five patients showed ΔpH(i)-based lateralization opposite to the clinical findings. These results seem to indicate an essentially random distribution around ΔpH(i) = 0 within a very small experimental error of ±0.02 pH units. pH(i) obtained from eight different areas in each of the 13 unilateral patients also did not show any significantly nonzero ΔpH(i) values. These results led to the conclusion that even at the excellent spectral resolution and reproducibility of the 4.1 T machine typical SD of 0.05 pH units), no significant pH(i) effect, induced by temporal lobe epilepsy, could be detected. Therefore, in this study, ΔpH(i) does not appear to be a clinically useful tool for the lateralization of epileptic foci in patients with temporal lobe epilepsy.