1. The chronotropic and inotropic effects of amrinone, carbazeran and 3-isobutyl-1-methyl xanthine (IBMX) were examined in isolated preparations of papillary muscle and right atria from rabbit heart. The effects of the drugs on cardiac phosphodiesterase and cyclic nucleotide content were also examined. 2. Amrinone (2.4 x 10(-4)M-2 x 10(-3) M), carbazeran (9.1 x 10(-6) M-1.2 x 10(-3) M), and IBMX (1.8 x 10(-5) M-4.5 x 10(-4) M) produced concentration-dependent positive inotropic responses of papillary muscle preparations, the rank order of potency being carbazeran = IBMX greater than amrinone. Sub-threshold positive inotropic concentrations of all three compounds potentiated the positive inotropic effects of isoprenaline; leftward shifts in the concentration-effect curves were 5 fold (IBMX), 11 fold (amrinone) and 46 fold (carbazeran). 3. Amrinone and IBMX produced concentration-dependent positive chronotropic responses in isolated right atria and showed a similar rate selectivity to isoprenaline, but carbazeran elicited a decrease in beating frequency. None of these drugs potentiated the positive chronotropic effects of isoprenaline. 4. Concentrations of amrinone, carbazeran and IBMX that produced similar positive inotropic responses were associated with different increases in papillary muscle cyclic AMP and cyclic GMP concentrations. 5. All three compounds inhibited right atrial and ventricular phosphodiesterase, with amrinone being the least potent. There was, however, a marked difference between the IC50 and EC50 values for phosphodiesterase inhibition and positive inotropy. In contrast the positive chronotropic effects of amrinone and IBMX were observed in the same concentration ranges that produced phosphodiestrease inhibition. 6. The results indicate that amrinone possesses a similar rate/force selectivity to isoprenaline and IBMX. In contrast, carbazeran exerts both positive inotropic and negative chronotropic effects. Phosphodiesterase inhibition and elevation of intracellular cyclic AMP concentration may be involved, at least in part, in the cardiac effects of these drugs.