Twenty-nine adult iron-deficient anemis subjects (13 men and 16 women) with hemoglobin levels of 4.0 to 12.0 g/100 ml blood were divided into either an iron treatment or placebo group. Hematological, cardiorespiratory and performance data were collected before, during, and after treatment and compared with data from a control group of subjects (4 men and 6 women) from the same socioeconomic population. Hemoglobin levels for the iron treatment group improved from 7.7 to 12.4 g for the women and from 7.1 to 14.0 g for the men. Values for the control group were 13.9 g and 14.3 g for the women and men, respectively. The placebo group showed virtually no change over the 80-day period (8.1-8.4 g for women and 7.7-7.4 g for men). Peak exercise heart rates (5 min, 40-cm step test) were significantly reduced after treatment from 155 to 113 for the iron treatment men and 152 to 123 for the women compared with the placebo group which showed no changes. Values for the control group were 119 and 142 for the men and women, respectively. In response to the exercise test, no difference in oxygen consumption was found between the iron treatment and placebo group although 15% more O2 was delivered per pulse in the iron treatment group. Blood lactates were significantly highein the placebo than iron treatment group both at rest, 1.18 versus 0.64 mmole/liter, and 1 min after exercise, 5.30 versus 2.68 mmoles/liter. No changes in handgrip or shoulder adductor strength were observed following treatment. These results clearly support the concept that performance requiring high oxygen delivery is significantly affected by hemoglobin levels.