Evaluation of heart rate variability (HRV) is based on analysis of consecutive R-R intervals and may provide quantitative information on the modulation of cardiac vagal and sympathetic efferent activities. In particular, power spectral analysis as well as the fractal dimension of HRV represent non-invasive measures that reflect brain-heart interaction in different physiopathological conditions. This study was performed to investigate the relationship between autonomic nervous cardiac influence and three phases (menses, follicular phase and luteal phase) of the menstrual cycle. Heart rate (HR) was recorded in 6 eumenorrheic young females, continuously for 20 min, at rest. From the tachograms, fractal dimension, FFT spectra and beta coefficient were evaluated. The components of two spectral bands were calculated: 0.040 Hz - 0.150 Hz (low frequency, LF), and 0.150 Hz - 0.80 Hz (high frequency, HF). The results indicate no correlation between HR and follicular as well as luteal phase of the menstrual cycle. Total power and HF spectral component increase, whereas LF spectral component decreases in the luteal phase compared to the follicular one. The fractal dimension does not show significant differences among the three phases. Beta coefficient decreases during luteal phase in respect of follicular one and menses. In conclusion, these results indicate in healthy young women a correlation between female sex hormones (17 beta-oestradiol, progesterone and pituitary gonadotrophins) concentrations and some HRV parameters with higher HR variability during the luteal phase.