The primary force of muscle contraction is thought to involve a change in the myosin head whilst attached to actin, the energy coming from ATP hydrolysis. This change in attached state could either be a conformational change in the head or an alteration in the binding angle made with actin. A considerable amount is known about one bound state, the so-called strongly attached state, which occurs in the presence of ADP or in the absence of nucleotide. In this state, which probably corresponds to the last attached state of the force-producing cycle, the angle between the long axis myosin head and the actin filament is roughly 45°. Details of other attached states before and during power production have been difficult to obtain because, even at very high protein concentration, the complex is almost completely dissociated by ATP. Electron micrographs of the complex in the presence of ATP have therefore been obtained only after chemically cross-linking myosin subfragment-1 (S1) to actin filaments to prevent dissociation. But it is unclear then whether the variability in attachment angle observed is due merely to the cross-link acting as a hinge.We have recently found low ionic-strength conditions under which, without resorting to cross-linking, a high fraction of S1 is bound to actin during steady state ATP hydrolysis. The structure of this complex is being studied by cryo-electron microscopy of hydrated specimens. Most advantages of frozen specimens over ambient temperature methods such as negative staining have already been documented. These include improved preservation and fixation rates and the ability to observe protein directly rather than a surrounding stain envelope. In the present experiments, hydrated specimens have the additional benefit that it is feasible to use protein concentrations roughly two orders of magnitude higher than in conventional specimens, thereby reducing dissociation of weakly bound complexes.