Body composition interests athletes since athletic performance is influenced by and dependent on the proportion and total amount of fat-free mass (FFM) and fat mass. The use of bioelectrical impedance (BIA) has increased recently since portable instruments make the measurements easy to execute and relatively inexpensive. The purpose of this study was to test the degree of relationship between FFM and fat mass calculated in elite female runners with 12 different BIA formulas reported in the literature and measured by dual-energy x-ray absorptiometry (DXA). The present study shows that body composition by BIA is valid in female runners. Prediction equations used to calculate FFM and fat mass must be appropriate for this population subgroup and validated against other methods, such as DXA and hydrodensitometry. Those formulas that performed well in the controls gave poor results in the female runner and vice versa. The below average fat mass noted in female runners suggests that prediction equations for untrained women with average fat mass are inappropriate. The formula by RJL Systems-2 for women: FFM = 5.091 + 0.6483.height2/resistance + 0.1996.weight gave best predictions of FFM in female runners. Further research is necessary to validate BIA prediction formulas in other athletes.