We compared the effects of pulsatile vs. nonpulsatile flow (Q) on pulmonary arterial pressure (Ppa)-Q relationships in zone 3 over wide ranges of pulse rate, stroke volume (SV), and Q. Excised left lungs of rabbits (n = 15) were perfused with tris(hydroxymethyl)aminomethane-buffered Tyrode solution containing 4% dextran, 1% albumin, and 10 mg/l of indomethacin and were ventilated with room air. Pulsatile Q was generated by a diaphragm pump delivering SV of 0.5, 1, or 2 ml (representing approximately 0.3, 0.6, and 1.2 times, respectively, the normal resting SV for rabbit left lung) and adjusting the pump frequency. Nonpulsatile Q was generated by raising an arterial reservoir to the required height. Mean pulmonary arterial (Ppa) and left atrial pressures were measured at end exhalation (positive end-expiratory pressure = 2.5 cmH2O) near the tips of the perfusion cannulas and were referenced to the lung base. Left atrial pressure was held constant at 7 cmH2O.Q was alternated between pulsatile and nonpulsatile, increasing Q stepwise from 100 to 600 ml/min (Q from approximately 0.3 to 2 times the normal resting Q for rabbit left lung), after which Q was reduced stepwise back to initial values. For the smallest SV there were no differences between Ppa-Q curves under pulsatile and nonpulsatile conditions. At the largest SV, Ppa was greater during pulsatile than nonpulsatile Q at Q > 100 ml/min. The slopes of the Ppa-Q curves were greater during pulsatile Q at the two larger SV values. These results can be explained by increasing Q turbulence and less ideal velocity profiles at higher peak Q resulting from the effects of rapidly changing inertial forces.