To assess the effects of exercise on left ventricular volumes we studied 10 normal men, 15 patients with coronary disease who developed angina pectoris during exercise, and 10 patients with known coronary disease who did not develop angina during exercise. Each subject performed supine bicycle exercise under a mobile, single-crystal scintillation camera until angina or fatigue occurred. Technetium-99m bound to human serum albumin was the imaging agent. Data were collected at rest and during the last 2 minutes of each 3-minute stage of exercise and for 10 minutes after exercise. Volumes were calculated by a new radionuclide technique that correlates well with cineangiography and is expressed in nondimensional units. In normal subjects, the end-diastolic volume (EDV) at rest was not different from that at peak exercise: 15.8 ± 6.1 (SD) vs 15.2 ± 7 (NS). The end-systolic volume (ESV) decreased from 5.6 ± 2.9 to 2.6 ± 1.6 at peak exercise (p < 0.01). ESV decreased progressively in all but two of 30 exercise periods. Angina patients had a larger EDV (p < 0.01) at rest and during chest pain (p < 0.01) than normals. Angina patients increased their ESV from 10.9 ± 6.7 to 14.2 ± 7.5 during chest pain (p < 0.001), resulting in a decreased ejection fraction (EF) (0.57 ± 0.15 to 0.45 ± 0.16, p < 0.001). All angina patients had a higher ESV during chest pain than during the exercise stage before chest pain. As a group, patients who did not develop angina had a lower EDV at rest and peak exercise than those who did develop angina: 14.8 vs 23.5 (p < 0.01) and 13.7 vs 23.5 (p < 0.01), respectively. The ESV went from 8.1 ± 4.1 at rest to 7.3 ± 5.3 at peak exercise, and the EF from 0.45 ± 0.12 to 0.50 ± 0.16 (all NS). We conclude: that the EF increases during exercise due to a decrease in ESV; that the EF in patients with angina decreases because of an increase in ESV; and that the EF in coronary disease patients without angina shows no change because there is no significant change in the ESV. Radionuclide equilibrium angiography may prove useful for assessing EF and volume changes in patients with coronary artery disease.