Because the membrane-bound ATPase and its substrate ADP are both electronegative, the effect of the ionic concentration on the kinetic parameters of photophosphorylation Vmax and Km (ADP) was examined in lettuce thylakoids in various conditions. The initial rate of ATP synthesis was measured by the "scalar" H+ consumption in the external medium, in conditions where the "vectorial" electrochemical proton gradient (restricted here to ΔpH) was maintained constant; ΔpH was estimated from the 9-aminoacridine fluorescence quenching. This signal was calibrated by experiments where the "phosphate potential" ΔGp was in equilibrium with ΔpH. With a fixed H+/P stoichiometry of 3, a strong dependence of the probe response on the ionic conditions was found. On the other hand, at a given ΔpH, the Km for ADP was unexpectedly found insensitive to the ionic conditions. Especially, lowering the Mg2+ concentration from 5 mM to 0.5 mM diminished Vmax considerably but did not affect Km. Attempts to measure the membrane surface electrical potential ψs by salt-induced release of 9-aminoacridine and by particle electrophoresis showed important discrepancies between the two methods. This seems not to be due to the difference between the potential measured at the membrane surface, more precisely at the Stern or Gouy levels, and that at the plane of shear only. It is more likely a consequence of side effects with 9-aminoacridine. Taking the electrokinetic potential ζ as representative of ψs, it is then shown that the membrane surface potential is already small in the minimum salt medium (ca. -15 mV) and, consequently, that the ADP concentration near the membrane, and hence Km, can barely change with the ionic concentration increase. This makes that the Km determinations, based on ADP concentrations in the bulk phase, are close to the true Km, which must take into account the actual ADP concentration at the membrane level. © 1990.