Debate continues on whether left ventricular (LV) systolic function during exercise is abnormal in young subjects with mild hypertension and on whether the abnormal blood pressure (BP) trend observed in hypertensives during prolonged exercise is due to impaired LV function. LV function was measured by means of M-mode echocardiography during prolonged exercise in 13 physically trained, young, mild hypertensives and 12 age-matched, trained normotensives with similar working capacity. Systolic BP/end-systolic volume (SBP/ESV) and end-systolic stress/ESV at rest were greater in the hypertensives (P < 0.0001 and P = 0.034), while LV filling was impaired (P = 0.05). BP changes during the first 20 min of exercise were similar in the two groups, but thereafter the between-group BP difference tended to decline progressively. LV diastolic dimension was similar at rest. During exercise it slightly increased in the normotensives and slightly decreased in the hypertensives (P = 0.032). Exercise ejection fraction (P = 0.018), SBP/ESV (P < 0.0001) and stress/ESV (P = 0.027) were greater in the hypertensives throughout the test. SBP/ESV normalized for LV wall thickness (P < 0.0001) and the changes in SBP/ESV from rest to exercise were also greater in the hypertensives (P = 0.002). Stroke volume increased to a lower extent in the hypertensives, but the between-group difference was not statistically significant. The increase in SBP/ESV from rest to exercise was related to the concentric remodelling of the ventricle in the hypertensives (P < 0.0001) and the subjects grouped together (P < 0.0001), but not in the normotensives. In conclusion, increased LV systolic performance is present early in hypertension not only at rest but also during vigorous exercise. It is partly due to concentric remodelling of the left ventricle and partly to enhanced inotropic state.