The kinetics of the increase in protein fluorescence following the addition of ATP to subfragment-1 (SF-1) and acto-SF-1 have been reinvestigated. The concentration dependence of the rate obtained with SF-1 did not fit a hyperbola and at high ATP concentration, approximately 40% of the signal amplitude was lost due to a fast phase at the beginning of the transient (20 degrees C). At lower temperature (less than or equal to 10 degrees C) the fluorescence transient was biphasic, with a fast phase observed at high ATP concentration. These results indicate that there are two steps in the SF-1 pathway in which there is a change in protein fluorescence. Measurements of ATP binding and hydrolysis by chemical quench-flow methods indicate that the rate of ATP binding is correlated with the fast fluorescence step and hydrolysis is correlated with the slow fluorescence change. The SF-1 mechanism can thus be described as: (formula: see text) where M represents SF-1 and states of enhanced fluorescence are given by M (16%) and M (36% enhancement, relative to SF-1). Step 1 is a rapid equilibrium with K1 approximately 10(3) M-1. Tight binding of ATP occurs in step 2 and the loss of signal amplitude requires k2 greater than or approximately 1500--2000 s-1. The maximum observed fluorescence rate defines the rate of hydrolysis, k3 + k-3 = 125 s-1 (20 degrees C, 0.1 M KCl, pH 7.0). The steps in the mechanism correspond to the Bagshaw--Trentham scheme, with the important difference that the assignment of rate constant is altered. Formation of the acto-SF-1 complex gave a fluorescence enhancement of approximately 14% relative to SF-1. Dissociation of acto-SF-1 by ATP produced a 20--22% enhancement in fluorescence. There was no detectable fluorescence change during dissociation as evidenced by a lag in the fluorescence transient which corresponded to the kinetics of dissociation. The fluorescence change occurred at the same maximum rate as for SF-1 but there was no loss in signal amplitude at high ATP concentration. The kinetics of the fluorescence change corresponded to the rate of ATP hydrolysis, whereas tight ATP binding occurred at a much faster rate in approximate agreement with the rate of dissociation. Thus the fluorescence change in the acto-SF-1 pathway corresponds to step 3 in the SF-1 mechanism. The complete scheme can be described as follows: (formula: see text) where AM represents acto-SF-1. The tight binding step in the SF-1 pathway (k2) is sufficiently fast so that a similar step (k2') in the acto-SF-1 pathway could precede dissociation but the AM-ATP intermediate has not been detected. Following hydrolysis on the free SF-1, actin recombines with M.ADP.Pi or possibly with a second SF-1 product intermediate as proposed by Chock et al. (1976) and the fluorescence returns to the original AM level with product release.