The FoF1 ATP synthase is a molecular motor critical to the metabolism of virtually all life forms, and it acts in the manner of a hydroelectric generator. The F1 complex contains an (αβ)3 (hexamer) ring in which catalysis occurs, as well as a rotor comprised by subunit-ε in addition to the coiled-coil and globular foot domains of subunit-γ. The F1 complex can hydrolyze ATP in vitro in a manner that drives counterclockwise (CCW) rotation, in 120° power strokes, as viewed from the positive side of the membrane. The power strokes that occur in ≈ 300 μsec are separated by catalytic dwells that occur on a msec time scale. A single-molecule rotation assay that uses the intensity of polarized light, scattered from a 75 × 35 nm gold nanorod, determined the average rotational velocity of the power stroke (ω, in degrees/ms) as a function of the rotational position of the rotor (θ, in degrees, measured in reference to the catalytic dwell). The velocity is not constant but rather accelerates and decelerates in two Phases. Phase-1 (0° - 60°) is believed to derive power from elastic energy in the protein. At concentrations of ATP that limit the rate of ATP hydrolysis, the rotor can stop for an ATP-binding dwell during Phase-1. Although the most probable position that the ATP-binding dwell occurs is 40° after the catalytic dwell, the ATP-binding dwell can occur at any rotational position during Phase-1 of the power stroke. Phase-2 of the power stroke (60° - 120°) is believed to be powered by the ATP-binding induced closure of the lever domain of a β-subunit (as it acts as a cam shaft against the γ-subunit). Algorithms were written, to sort and analyze F1-ATPase power strokes, to determine the average rotational velocity profile of power strokes as a function of the rotational position at which the ATP-binding dwell occurs (θATP-bd), and when the ATP-binding dwell is absent. Sorting individual ω(θ) curves, as a function of θATP-bd, revealed that a dependence of ω on
θATP-bd exists. The ATP-binding dwell can occur even at saturating ATP concentrations. We report that ω follows a distinct pattern in the vicinity of the ATP-binding dwell, and that the ω(θ) curve contains the same oscillations within it regardless of θATP-bd. We observed that an acceleration/deceleration dependence before and after the ATP-binding dwell, respectively, remained for increasing time intervals as the dwell occurred later in Phase-1, to a maximum of ≈ 40°. The results were interpreted in terms of a model in which the ATP-binding dwell results from internal drag at a variable position on the γε rotor.