Motile forces in muscle are generated by the so-called ''power stroke,'' a series of structural changes in the actomyosin crossbridge driven by hydrolysis of ATP. The initiation of this power stroke is closely related to phosphate release after ATP cleavage and to the change of the myosin head from weak, nonstereospecific actin attachment to strong, stereospecific binding. The exact sequence of events, however, is highly controversial but crucial for the mechanism of how ATP hydrolysis drives structural changes in the head domain of myosins and related NTPases like kinesins and small G proteins. Here, we show that the phosphate analogue AlF4 can form two ADP⅐phosphate analog states, one with weak binding of myosin to actin and the other with strong binding of myosin to actin. Thus, change from weak to strong binding (i.e., the initiation of the power stroke) can occur before phosphate is released from the active site.actomyosin cross-bridge ͉ phosphate release ͉ weak binding cross-bridge states ͉ strong binding cross-bridge states ͉ force generation M uscle contraction results from cyclic interactions of the myosin head with actin filaments as ATP is hydrolyzed. Within each cycle, a multistep power stroke drives actin filaments several nanometers past the myosin filaments or generates motile forces when filament sliding is prevented. According to the concept of Lymn and Taylor (1), the power stroke is driven by the release of the ATP-hydrolysis products ADP and phosphate from the active site. From subsequent work, it was proposed that the initiation of the power stroke is closely related to the release of phosphate (2) and to the change of the myosin head from a conformation of weak, nonstereospecific actin binding to a conformation of strong, stereospecific actin attachment (3, 4). The exact sequence of events [i.e., how structural changes in the actomyosin cross-bridge are coupled to changes in actin affinity and to the release of phosphate from the active site] is still unclear. This coupling, however, is crucial for the mechanism of how ATP hydrolysis drives structural changes in the head domain not only of myosins but also of kinesins and small G proteins. Although it is clear that intermediates with MgADP and phosphate in the active site (ADP⅐P i intermediates) are central to the initiation of the power stroke, it is still controversial whether, for example, phosphate release has to occur before the transition to the strong binding conformation (5, 6) or whether a strong binding ADP⅐P i intermediate is formed before phosphate release. A second ADP⅐P i intermediate has been postulated before (7-11), but its properties in terms of weak or strong actin-binding conformation remained unclear. One difficulty in characterizing the ADP⅐P i intermediates is their transient nature during active cross-bridge cycling, specifically of an ADP⅐P i intermediate postulated to bind strongly to actin before the release of phosphate (8-10). To elucidate events around phosphate release, we searched for approaches to accumulate...