The aim of this study was to investigate any difference in bone mass at different sites between female long-distance runners with amenorrhea and those with eumenorrhea. We compared 10 amenorrheic and 10 eumenorrheic athletes to determine whether athletes with amenorrhea have lower BMD in multiple skeletal regions, including weight-bearing lower limbs. The amenorrheic group had experienced menstrual dysfunction ranging from 3 to 43 months. As a further control group, 16 eumenorrheic soccer players were compared with the former two running groups regarding their BMD measurements. The two groups were matched for age, height, and amount of training. Areal bone mineral density (BMD) was measured and was found to be significantly lower in the total body, humerus, spine, lumbar spine, pelvis, femoral neck, trochanter, total femur, femur diaphysis, tibia diaphysis and in the nonweight-bearing head of the femur in the amenorrheic group. Body weight, BMI, fat mass, and body fat percent were significantly lower in the amenorrheic group. The differences in the BMD of the head, humerus, femoral neck, total femur, femur diaphysis, and tibia diaphysis disappeared when adjusted for body weight. Compared with the soccer group, the amenorrheic subjects had significantly lower BMD values at all sites except for the head, Ward's triangle, and femur diaphysis. Blood samples were obtained in the two running groups for analysis of osteocalcin, carboxy terminal telopeptide (ICTP), procollagen I (PICP), and estradiol. There were no significant differences between the groups but there was a strong tendency towards a lower estradiol level and a higher osteocalcin level in the amenorrheic group. A free estradiol index (FE2) was derived as the ratio of estradiol to sex hormone binding globulin (SHBG) and was significantly lower in the amenorrheic group. No difference in their daily intake of total energy, protein, carbohydrates, fiber, calcium, and vitamin D was observed. However, both groups showed a surprisingly low energy intake in relation to their training regimens. Stepwise regression analyses revealed that weight was the best predictor of spine BMD in both groups. Estradiol and FE2 were significant predictors of the BMD of the proximal femur in the eumenorrheic group, but did not predict any BMD site in the amenorrheic group. In conclusion, amenorrhea in athletic women affects trabecular and cortical bone in both axial and appendicular skeleton. However, some of the discrepancy can be explained by a lower body weight. Physical weight-bearing activity does not seem to completely compensate for the side effects of reduced estrogen levels even in weight-bearing bones in the lower extremity and spine.