SUMMARY1. Double-barrelled, neutral-carrier pH-sensitive micro-electrodes were used to measure the intracellular pH (pHi) and the pHi regulation of neuropile glial (n.g.) cells and of identified neurones of the leech Hirudo medicinalis.2. The distribution of H+ in the n.g. cells and in the neurones was found not to be in electrical equilibrium. The mean pHi ofthe n.g. cells was 6-87 + 0 13 (mean + S.D.of mean here and for all following data n = 27) in HEPES-buffered leech saline (pHo = 7 4) and 7418 + 0-19 (n = 13) in 2 % C02-11 mM-HC03--saline. The mean pHi was 7'28+0-1 (n = 20) in Retzius neurones and 7-32+0415 (n = 12) in noxious neurones in HEPES-buffered leech saline, and 7-20+0415 (n = 10) and 7-27 +016 (n = 6) in 2 % C02-1 1 mM-HC03--buffered saline in these two types of neurones, respectively.3. The cytoplasmic buffering power, as calculated by the change in pHi following the change from 2 % C02-11 mM-HC03-to 5 % C02-22 mM-HCO.-in the leech saline, was 20-30 mM/pH unit in the n.g. cells and between 12 and 33 mM/pH unit in the neurones.4. The recovery of pHi in n.g. cells from an experimentally induced acidification (addition and removal of 20 mm-NH4Cl) was dependent on the presence of external Na+. Independent of the buffer system used, pHi recovery was inhibited when external Na+ was exchanged by N-methyl-D-glucamine. Amiloride (2-3 mM) reduced the rate of pHi recovery by about 50 % in these n.g. cells.5. In CO2-HCO3--free saline, or in the presence of the anion exchange blocker 4-acetamido-4'-isothiocyanostilbene-2, 2'-disulphonic acid (SITS, 0 5 mM), pHi recovery from an acid load was often slowed by up to 50 % in n.g. cells. This suggests that there is a significant contribution of a HC03--dependent membrane transport to pHi regulation in n.g. cells.6. When a HEPES-buffered saline was exchanged by a 2 % C02-11 mM-HCO3-buffered saline, the pHi of n.g. cells increased by 0-31 pH units. This alkaline shift was reversible upon removal of the CO2-HC03-and was absent in the Na+-free saline. It was not inhibited by 1 mM-furosemide or by 0 5 mM-SITS. Successive addition of first Na+ and then CO2-HCO3-to an Na+-free HEPES-buffered saline produced two separate phases of intracellular alkalization, suggesting two different pHi regulation transport systems.J. W. DEITMER AND W.-R. SCHLUE 7. In sensory and Retzius neurones pHi recovery from acidification was also dependent on external Na+, both in the presence and in the absence of CO2-HC03-.The rate of pHi recovery was reversibly reduced by 2 mM-amiloride by up to 90 % in these neurones.8. In conclusion, bicarbonate buffer affects n.g. cell pHi differently from the way in which it affects neuronal pHi. The regulation of pHi depends on the presence of external Na+ in both n.g. cells and neurones. The experiments suggest the presence of an Na+-H+ exchange and a SITS-sensitive HCO3--and Na+-dependent acid extrusion mechanism across n.g. (and nerve cell) membranes and another, SITSinsensitive HC03-and Na+-dependent transport system only in n.g. cells.