The pH dependence of the L-to-M transition in the photocycle of bacteriorhodopsin was studied in pump-probe resonance Raman (RR) flow experiments in the range pH 3.5-7.8 on a time scale of 0-700 micros. For pH < 5, following the initial decay of L to M, the two intermediates approach nearly constant levels. From a specially designed perturbation-relaxation experiment at pH 4.6, in which the composition of L and M is perturbed by photoreversal of M, it could be concluded that the incomplete decay of L is due to an intermediate equilibration between L and M. It was found, by both RR and optical transient spectroscopy, that the maximum level of M (approximately 500 micros) increases with pH according to a pKa of 5. 6 (150 mM Na+). Since the proton release from an internal group XH to the extracellular surface is determined by nearly the same pKa of 5.7 [Zimanyi, L., Varo, G., Chang, M., Ni, B., Needleman, R., and Lanyi, J. K. (1992) Biochemistry 31, 8535-8543], it is concluded that this increase is controlled by the dissociation of XH. From the analysis of the perturbation-relaxation experiments and the multiexponential rise of M, a kinetic scheme with two sequential L-M equilibria is proposed for the L-M transition. By comparison with the time behavior of proton release [Heberle, J., and Dencher, N. A. (1992) Proc. Natl. Acad. Sci. U.S.A. 89, 5996-6000], it is suggested that it is the second equilibrium which is further shifted toward the M state by the dissociation of XH. From the magnitude of this shift, it is concluded that the L-M transition and proton release are not as strongly coupled as is generally assumed. Instead, it is proposed that structural changes during the photocycle are the dominating factors which reduce the pKa of XH to approximately 5.7 so that proton release becomes possible under normal conditions.