We hypothesized that 4 weeks of recombinant human erythropoietin (RhEPO) treatment would result in a significant increase in haemoglobin concentration ([Hb]) and arterial blood O2-carrying capacity and that this would (1) increase peak pulmonary oxygen uptake (V̇O2) during ramp incremental exercise, and (2) speed V̇O2 kinetics during 'severe'-, but not 'moderate'-or 'heavy'-intensity, step exercise. Fifteen subjects (mean ± S.D. age 25 ± 4 years) were randomly assigned to either an experimental group which received a weekly subcutaneous injection of RhEPO (150 IU kg-1; n = 8), or a control group (CON) which, received a weekly subcutaneous injection of sterile saline (10 ml; n = 7) as a placebo, for four weeks. The subjects and the principal researchers were both blind with respect to the group assignment. Before and after the intervention period, all subjects completed a ramp test for determination of the gas exchange threshold (GET) and V̇O2,peak, and a number of identical 'step' transitions from 'unloaded' cycling to work rates requiring 80% GET (moderate), 70% of the difference between the GET and V̇O2,peak (heavy), and 105% V̇O2,peak (severe) as determined from the initial ramp test. Pulmonary gas exchange was measured breath-by-breath. There were no significant differences between the RhEPO and CON groups for any of the measurements of interes ([Hb]), V̇O2,peak, V̇O2 kinetics) before the intervention. Four weeks of RhEPO treatment resulted in a 7% increase both in [Hb] (from 15.8 ± 1.0 to 16.9 ± 0.7 g dl-1; P < 0.01) and V̇O2,peak (from 47.5 ± 4.2 to 50.8 ± 10.7 ml kg-1·min-1; P < 0.05), with no significant thange in CON. RhEPO had no significant effect on V̇O2 kinetics for moderate (Phase II time constant, from 28 ± 8 to 28 ± 7 s), heavy (from 37 ± 12 to 35 ± 11 s), or severe (from 33 ± 15 to 35 ± 15 s) step exercise. Our results indicate that enhancing blood O2-carrying capacity and thus the potential for muscle O2 delivery with RhEPO treatment enhanced the peak V̇O2 but did not influence V̇O2 kinetics, suggesting that the latter is principally regulated by intracellular (metabolic) factors, even during exercise where the V̇O2 requirement is greater than the V̇O2,peak, at least in young subjects performing upright cycle exercise.