The effect of molsidomine-induced venodilation on cardiac preload was studied in conscious resting dogs, instrumented to analyze left ventricular function and myocardial perfusion. Direct effects on veins were studied during chloralose anesthesia by measuring regional venous capacitance changes with an induction angiometer. Kinetics of molsidomine-induced effects were compared to those induced by nitroglycerin and isosorbide dinitrate. This comparison was restricted to low i.v. dosages, causing only transient threshold effects on peripheral resistance and heart rate. During molsidomine-induced venous pooling, neither any direct effect on the coronary circulation nor any direct cardiac depressant activity of the drug was detected. 100 microgram/kg molsidomine caused a reduction of left ventricular preload by 5 mm Hg, lasting at least 4 hours. This effect was significantly more pronounced than that induced by 1 microgram/kg nitroglycerin or by 25 microgram/kg isosorbide dinitrate, lasting 2 min or 20 min, respectively. However, in raising regional venous capacitance, these nitrate dosages were equi-effective to 100 microgram/kg molsidomine, the effect of which was persistent and with a greater delay in onset. These results indicate that the lasting persistance of venodilation is a decisive factor for the amount of volume pooled in the capacitance system and, consequently, for the extent of preload reduction obtained. It is concluded, that lasting vasodilation, restricted to the veins, is beneficial for ventricular performance in ischemic heart disease.