Abstract
It has been widely assumed that the rate-limiting step in actomyosin ATPase activity during contraction is a single-stage attachment of the myosin head to a site on the thin filament. This assumption leads to a rate of energy liberation that increases monotonically with speed of shortening, in agreement with pre-war estimates of the output of work and heat; it does not explain the fall in energy release at high speeds that has been found more recently. This fall can be explained by assuming that the attachment is a two-stage process.