Although right ventricular function may be examined by following the passage (first pass) of a bolus of radionuclide through the right heart before it reaches the left heart, the counts detected with conventional gamma cameras in such a short time interval are low; moreover, repeated determinations would result in an unacceptable radiation burden to the patient. We have modified the gated equilibrium blood pool method to allow repeated assessment of the right ventricular ejection fraction (RVEF) and have compared the results with the first-pass method in 43 patients. Good agreement was obtained between the two methods (r = 0.91, p less than 0.001). The mean difference between the two methods was 0.04 with an intra-observer variation for the equilibrium studies of 0.03 and an inter-observer difference of 0.04. The mean difference in RVEF for seven patients studied on two separate occasions 30 minutes apart was only 0.02. In four patients the mean RVEF measured at rest was 0.44 +/- 0.05 (SEM) and after exercise it was 0.48 +/- 0.06. After infusion of isoprenaline at 1 microgram/min the mean rose to 0.64 +/- 0.04 (p less than 0.02) and after infusion of a new beta 1-sympathomimetic agent, prenalterol, at doses of 1 and 2 mg it was 0.56 +/- 0.02 (p less than 0.02) and 0.59 +/- 0.03 (p less than 0.01) respectively, where the significance levels are relative to the resting values. In nine patients with good ventricular function the vasodilator nifedipine caused right and left ventricular ejection fractions to increase by the same amount; while in six patients with severe impairment of left ventricular function due to ischaemic heart disease the RVEF increased from 0.58 +/- 0.03 to 0.73 +/- 0.03 (p less than 0.01) after 2 mg of prenalterol, but the left ventricular ejection fraction increased only from 0.22 +/- 0.04 to 0.26 +/- 0.04. We conclude that repeated estimation of right ventricular performance is possible by equilibrium radionuclide ventriculography.