The redox midpoint potential (Em) of the primary quinone of bacterial reaction centers, Q(A), in native membranes (chromatophores) measured by redox potentiometry is reported to be pH dependent (-60 mV/pH) up to a highly distinctive pKa (9.8 in Rba. sphaeroides) for the reduced state. In contrast, the Em of Q(A) in isolated RCs of Rba. sphaeroides, although more variable, has been found to be essentially pH-independent by both redox potentiometry and by delayed fluorescence, which determines the free energy (Delta G(PA)) of the P+Q(A) state relative to P. Delayed fluorescence was used here to determine the free energy of P+Q(A) in chromatophores. The emission intensity in chromatophores is two orders of magnitude greater than from isolated RCs largely due to the entropic effect of antenna pigments "drawing out" the excitation from the RC. The pH dependence of Delta G(P*A) was almost identical to that of isolated RCs, in stark contrast with potentiometric redox titrations of Q(A). We considered that Q(A) might be reduced by disproportionation with QH2 through the Q(B) site, so the titration actually reflects the quinone pool, giving the -60 mV/pH unit dependence expected for the Q/QH2 couple. However, the parameters necessary to achieve a strong pH-dependence are not in good agreement with expected properties of Q(A) and Q(B). We also consider the possibility that the time scale of potentiometric titrations allows the reduced state (Q(A)) to relax to a different conformation that is accompanied by stoichiometric H+ binding. Finally, we discuss the choice of parameters necessary for determining the free energy level of P+Q(A) from delayed fluorescence emission from chromatophores of Rba. sphaeroides.