The purpose of this investigation was to study the effects of endurance exercise on male reproductive function (sex hormones and seminograms). Professional cyclists [n = 12; mean age 24 +/- 2 (SD) yr], elite triathletes (n = 9; 26 +/- 3 yr), recreational marathon runners (n = 10; 32 +/- 6 yr), and sedentary subjects (control group; n = 9; 30 +/- 4 yr) were selected as subjects. for each group, the following parameters were measured three times during the sports season (training period: winter; competition period: spring; resting period: fall): percentage of body fat, hormonal profile (resting levels of follicle-stimulating hormone, luteinizing hormone, total and free testosterone, and cortisol), and seminograms (quantitative parameters sperm volume and sperm count; qualitative parameters: sperm motality and morphology). The following comparisons were made in the measured parameters: 1) within groups (longitudinal design) and 2) between groups in each of the three periods (cross-sectional design) and over time (mixed design). In addition, both the volume and the intensity of training of each subject during the season (except for the control group) were quantified. Despite significant differences in training characteristics and in body fat percent, in general no significant differences (P > 0.05) were found in hormonal profiles or in semen characteristics between or within groups. A lower sperm motility (46.2 +/- 19.5%), however, was observed in the cyclists during the competition period when compared either with the other groups during this same period (P < 0.05) or with themselves during the other two periods of study (P < 0.01). In any case, the later phenomenon was attributed to physical factors associated with cycling, such as mechanical trauma to the testis and/or increased gonadal temperature. In conclusion, our findings suggest that endurance exercise does not adversely affect the hypothalamic-pituitary-testis axis.