This study was designed to investigate whether atrial natriuretic factor (ANF) administered over the physiological, pathological and pharmacological range has a negative inotropic action on the heart. Anesthetized rabbits were infused with increasing doses of ANF (0.05, 0.25 and 0.5 μg kg-1min-1), while measuring hemodynamic variables including the maximum rate of change of left ventricular pressure (d P/d tmax) as an index of inotropic state. Plasma levels of immunoreactive ANF (iANF) were measured to relate the hemodynamic changes to actual plasma levels of the peptide. Administration of ANF was associated with decreases in blood pressure, left ventricular pressure and d P/d tmax so that after 0.5 μg kg-1 min-1 infusion, these variables had decreased by 21±2 mmHg, 21±5.3 mmHg and 925±175 mmHg/s, respectively (P<0.01). There were no significant changes in right atrial pressure, left ventricular end-diastolic pressure or heart rate. Since d P/d tmax can be influenced by changing hemodynamic variables and baroreflex changes, a second group of rabbits was studied in which afterload and heart rate were held artificially constant. Again, in this group of rabbits, infusions of ANF led to decreasing inotropic state, so that at the highest infusion rate, a 14% decrease in d P/d tmax was observed (P<0.05). By comparison, hydralazine, a drug which causes active vasodilatation but no direct inotropic action, significantly (P<0.01) decreased blood pressure, left ventricular pressure and d P/d tmax when infused at a rate of 10 μg kg-1 min-1. However, in animals in which afterload was controlled, hydralazine did not affect any of the variables measured. The results indicate that ANF does have a negative inotropic action in the anesthetized rabbit. © 1990 Springer-Verlag.