The prevalence of iron deficiency and its impact on performance have not been adequately investigated in non-endurance athletes. This study evaluated these factors in 100 female intercollegiate athletes in various sports, and in 66 non-athletes. All subjects had determinations of hemoglobin, ferritin, and transferrin saturation. Athletes reported on diet, menses, and symptoms. Athletes and coaches described mood and performance. Iron-deficient athletes were treated with iron; others received placebo. The same data were collected at season's end. Initially 31/100 (31%) athletes had iron deficiency (ferritin less than 12 ng.ml-1; transferrin saturation less than 16%, or both) compared to 30/66 (45.5%) controls (not statistically significant). Compared to normal athletes, iron-deficient athletes did not have more symptoms of iron deficiency or differences in mood state, but they considered their performance to be worse (P less than 0.05). Their total iron intakes were similar, as were menstrual blood losses. At re-evaluation, 7/45 (15.6%) initially normal athletes were iron-deficient; 14/22 (63.6%) initially iron-deficient athletes were normal. Athletes receiving an iron supplement and their coaches did not report a greater improvement in performance or mood than athletes receiving a placebo. Female college athletes frequently had iron deficiency that could be successfully treated during the season, while some untreated normal athletes became iron-deficient. Iron deficiency and its treatment had no significant impact on symptoms or mood, but affected subjective assessment of performance.