The pH dependence of the rate constant of dark adaptation (thermal isomerization from all-trans- to 13-cis-bR) drastically changes when Arg82 of bacteriorhodopsin is replaced by an alanine. In the wild type (WT) the rate decreases sharply between pH 2.5 and pH 5. In R82A the sharp decrease is shifted to pH > 7. This correlates with the shift in the pK of the purple-to-blue transition from pH 2.6 in the wild type to pH 7.2 in the mutant (in 150 mM KCl). We propose that the same group that controls the purple-to-blue transition, namely, Asp85, catalyzes dark adaptation. The rate of dark adaptation in the R82A mutant is proportional to the fraction of protonated Asp85, indicating that dark adaptation occurs when Asp85 is transiently protonated. Thermal isomerization is at least 2 x 10(3) times more likely when Asp85 is protonated (blue membrane) than when it is deprotonated (purple membrane). The pH dependence of dark adaptation in the WT can be explained by a model in which the rate of dark adaptation in the WT is also proportional to the fraction of protonated Asp85 and that the pK of Asp85 depends on some other group, X, which deprotonates (or moves away from Asp85) with pK9 and causes the shift in the pK of Asp85 from 2.6 to 7.2. The quantum yield of light adaptation is at least an order of magnitude less in R82A as compared to the WT. The rise time of M formation is very fast in R82A and, unlike the WT, pH independent (1 microsecond versus 85 and 6 microseconds in the WT at pH 7 and 10, respectively). The activation energy of the L to M transition is 6.9 kcal/mol versus 13.5 kcal/mol in the WT. Thus the loss of a positive charge in the active site greatly increases the rate of light-induced deprotonation of the Schiff base. In the R82A mutant, the M decay at pH > 8.8 is much faster than the recovery of initial bR, which suggests a decrease in the rate of back-reaction from N to M. In a suspension of R82A membranes the rate of proton release as measured by the pH-sensitive dye pyranine is delayed by at least 20-fold (in 2 M KCl), while the uptake of protons did not change much (12 ms in the WT versus 8 ms in R82A).(ABSTRACT TRUNCATED AT 400 WORDS)