The ability of exercise training to block the generation of hypertension produced by chronic stress in the borderline hypertensive rat was tested. Twenty-three male borderline hypertensive rats, F1 offspring of spontaneously hypertensive and Wistar-Kyoto rats, were divided into three groups. Two groups(8 rats per group) were subjected to 2 hours of daily, predictable, uncontrolled tail shock for 12 weeks. One of these groups was also given 2 hours of daily swim stress (exercise trained). A third group served as a maturation control and received neither intervention (n = 7). After 12 weeks of stress, direct recording of blood pressure verified the pattern observed with tail cuff; shock only group, 180/118 .+-. 3/3 mmHg; exercise-trained and shocked group, 166/108 .+-. 4/2 mmHg; and control group, 160/98 .+-. 6/4 mmHg (mean .+-. SEM). Systolic and diastolic blood pressures in the shock only group were significantly higher than in both the other groups (p < 0.05). The control group differed from the exercise-trained and shocked group only in diastolic BP (p < 0.05). During a short-term stress session plasma norepinephrine levels in the exercise-trained and shocked group were significantly lower than those in the shock only group (555 .+-. 56 vs 776 .+-. 84 pg/ml; p < 0.05). These results indicate that an alteration of autonomic function resulted from the exercise training, but its contribution to the resistance of the exercise-trained and shocked rats to stress-induced hypertension is unclear.