We performed the first analysis of heart rate variability (HRV) at rest and during exercise under full autonomic blockade on the same subjects, to test the conjecture that vagal tone withdrawal occurs at exercise onset. We hypothesized that between rest and exercise there would be 1) no differences in total power (P_TOT) under parasympathetic blockade, 2) a P_TOT fall under β₁-sympathetic blockade, and 3) no differences in P_TOT under blockade of both autonomic nervous system branches. Seven men [24 (3) yr, mean (SD)] performed 5-min cycling (80 W) supine, preceded by 5-min rest during control and with administration of atropine, metoprolol, and atropine + metoprolol (double blockade). Heart rate and arterial blood pressure were continuously recorded. HRV and blood pressure variability were determined by power spectral analysis, and baroreflex sensitivity was determined by the sequence method. At rest, P_TOT and the powers of low- and high-frequency components of HRV (LF and HF, respectively) were dramatically decreased with atropine and double blockade compared with control and metoprolol, with no effects on LF-to-HF ratio and on the normalized LF (LFnu) and HF (HFnu). During exercise, patterns were the same as at rest. Comparing exercise with rest, P_TOT varied as hypothesized. For systolic and diastolic blood pressure, resting P_TOT was the same in all conditions. During exercise, in all conditions, P_TOT was lower than in control. Baroreflex sensitivity decreased under atropine and double blockade at rest and under control and metoprolol during exercise. The results support the hypothesis that vagal suppression determined disappearance of HRV during exercise. NEW & NOTEWORTHY This study provides the first demonstration, by systematic analysis of heart rate variability at rest and during exercise under full autonomic blockade on the same subjects, that suppression of vagal activity is responsible for the disappearance of spontaneous heart rate variability during exercise. This finding supports previous hypotheses on the role of vagal withdrawal in the control of the rapid cardiovascular response at exercise onset.