Myosin ATPase activity is usually considered to reflect the contractile capacity of a given muscle since it correlates with the maximum initial speed of shortening of the unloaded muscle (V max). There are several exceptions to this scheme, and it was the goal of this study to determine if the Mg 2+-ATPase activity of the covalently bound actomyosin SI is a more physiological index of contractility. On polyacrylamide gels, the complex obtained after condensation of fast skeletal myosin SI to skeletal actin is identical to that obtained with myosin SI from the ventricles of different species, including rat, guinea pig, and human, cross-linked to cardiac or skeletal actin. In every condition, the ATPase activity of the complex is 700-fold higher than that of myosin SI. It correlates linearly with the V mra both in phylogeny and in conditions in which an isomyosin shift has been reported, such as hypothyroidism and chronic cardiac overload. Such a relation indicates that, in species that already have a low V raax , a small change in myosin ATPase may induce dramatic consequences in the shortening velocity. Cardiac hypertrophy in humans, where the drop in V raax is not associated with a myosin change, does not fit into this scheme. The enzymatic activity of the complex is also unmodified in this condition, which shows that, in humans, the myosin ATPase is not a determinant of \ mm and suggests that other mechanisms may be involved. Measurement of this type of ATPase activity provides a new tool to explore contractility biochemically, which is more reproducible and, from a technical point of view, easier to perform than a kinetic assay. It also correlates better with mechanical data obtained with skinned fibers than with those measured on fresh papillary muscles. (Circulation Research 1989;64:1106-1115) I t is now well established that the maximum shortening velocity of an unloaded skeletal 1 or cardiac 2 muscle (V max) correlates linearly to both Ca 2+-activated and actin-activated myosin ATPase activity measured in the steady state 12 and also to the activity measured during the initial phosphate burst. 3 This relation is one of the basics of muscle physiology and pathology since, in response to chronic overload in some animal species such as rats, it has been shown that the drop in V m ax correlates with a diminution in myosin ATPase and to a corresponding isomyosin shift from the fast isoform V, to the slow isoform V 3. 4-7 From INSERM, Hdpital Lariboisiere,