We reanalyzed experiments in in situ hearts of 16 open-chest anesthetized dogs, in which two different loading interventions were performed, i.e., an occlusion of the descending aorta (InP) and a rapid volume infusion (InV). Previous studies had demonstrated that the end-systolic elastance (Ees) of the InP was substantially larger than the Ees of the InV suggesting either a load dependency of Ees as such, or an increase in contractility during InP. The data were reanalyzed in the light of the muscular pump concept by plotting peak normalized velocity of circumferential shortening versus a global representative force approximating the left ventricle by a sphere. In all but one experiment the points of the two interventions are located on a single relationship over a very broad range of forces (from 397 to 2,461 g between the control states of experiments and from 602 to 3,278 g difference between control and highest load within experiments). The virtual independence of the force-velocity relation (FVR) and the dependence of the end-systolic pressure-volume relation (ESPVR) on the type of loading intervention can be ascribed to the fact that the former is assessed early during ejection and is therefore less influenced by shortening deactivation and internal resistance than the ESPVR. We conclude that the FVR offers a more consistent characterization of intrinsic LV function than the ESPVR.