This study explored the effects of gender and aging on the complexity of cardiac pacemaker activity. Electrocardiogram signals were studied in normal women (n = 240) and men (n = 240) ranging in age from 40 to 79 yr. Nonlinear analysis of short-term resting R-R intervals was performed using the correlation dimension (CD), approximate entropy (ApEn), and largest Lyapunov exponent (LLE). Evidence of nonlinear structure was obtained by the surrogate data test. CD, ApEn, and LLE were negatively correlated with age. Despite similar means and SDs of the R-R intervals, women had a significantly higher CD, ApEn, and LLE compared with men in the age strata of 40-44 and 45-49 yr. CD and ApEn were strongly (r > 0.71) correlated with low- and high-frequency components. We conclude that the resting cardiac pacemaker activity of women is more complex than that of men in middle age, and the gender-related difference diminishes after the age of 50 yr. The higher complexity implies a more comprehensive neural modulation.