The influence of celiprolol (beta1- and beta2-adrenoceptor partial agonist), propranolol (beta1- and beta2-adrenoceptor antagonist), and atenolol (beta1-adrenoceptor antagonist) on heart-rate variability (HRV) was assessed from Holter records in 12 normal volunteers. A combination of summary statistics and nonlinear procedures was used to assess HRV and autonomic balance. Under double-blind and randomised conditions (Latin-square design), subjects received placebo, celiprolol (200 and 800 mg), propranolol (160 mg), atenolol (50 mg), and combinations of these agents. Single oral doses of medication (at weekly intervals) were administered at 22:30 h with sleeping heart rates (HRs) recorded overnight. Compared with placebo, celiprolol (200 and 800 mg) increased the sleeping HR, the HR effect of celiprolol was different from the bradycardia after propranolol, 160 mg, and atenolol, 50 mg. Dose-response effects on HR with celiprolol were evident in the presence of atenolol, unlike those with propranolol that abolished the HR increase between celiprolol, 200 mg and 800 mg. These data were consistent with beta1-selective adrenoceptor agonism with 200 mg but agonism at both the beta1- and beta2-adrenoceptor with celiprolol, 800 mg. The action of the drugs on short-term HRV indices (rMSSD and pNN50) closely followed their effects on HR. The longer-term HRV indices (global SD, SDANN) were reduced by celiprolol but increased by propranolol and atenolol. At a fixed HR, the data dispersion (SDNN5) was higher with propranolol compared with celiprolol; however, the dispersion was not merely an HR-dependent phenomenon. A novel nonlinear approach (quadrant analysis) revealed the sequencing of cardiac accelerations and decelerations after the high correlation between adjacent intervals had been removed. Celiprolol increased the frequency of consecutive cardiac accelerations; the duration between and variance of these beat-to-beat differences shortened after celiprolol but lengthened with increased variance after propranolol and atenolol. These results demonstrated reduced HRV indices and a shift toward sympathetic dominance after the beta-adrenoceptor agonist celiprolol contrasting with increased HRV indices and parasympathetic dominance after the beta-adrenoceptor antagonists propranolol and atenolol. The implications of these findings for the treatment of patients with cardiovascular disease warrant further study.