In two experiments maximal aerobic power ([(V)\dot]_{O2 max} )(\dot V_{O_{2 max} } ) calculated from maximal mechanical power (W _max) was evaluated in 39 children aged 9–11 years. A maximal multi-stage cycle ergometer exercise test was used with an increase in work load every 3 min. In the first experiment oxygen consumption was measured in 18 children during each of the prescribed work loads and a correction factor was calculated to estimate [(V)\dot]_{O2 max} \dot V_{O_{2 max} } using the equation [(V)\dot]_{O2 max} = 12 ·W_max + 5 ·weight\dot V_{O_{2 max} } = 12 \cdot W_{\max } + 5 \cdot weight . An appropriate increase in work rate based on height was determined for boys (0.16 W · cm^–1) and girls (0.15 W · cm^–1) respectively. In the second experiment 21 children performed a maximal cycle ergometer exercise test twice. In addition to the procedure in the first experiment a similar exercise test was performed, but without measurement of oxygen uptake. Calculated [(V)\dot]_{O2 max} \dot V_{O_{2 max} } correlated significantly (p<0.01) with those values measured in both boys (r=0.90) and girls (r=0.95) respectively, and the standard error of estimation for [(V)\dot]_{O2 max} \dot V_{O_{2 max} } (calculated) on [(V)\dot]_{O2 max} \dot V_{O_{2 max} } (measured) wass less than 3.2%. Two expressions of relative work load (% [(V)\dot]_{O2 max} \dot V_{O_{2 max} } and %W _max) were established and found to be closely correlated. The relative work load in % [(V)\dot]_{O2 max} \dot V_{O_{2 max} } could be predicted from the relative work load in % W _max with an average standard error of 3.8%. The data demonstrate that calculated [(V)\dot]_{O2 max} \dot V_{O_{2 max} } based on a maximal multi-stage exercise test provides an accurate and valid estimate of [(V)\dot]_{O2 max} \dot V_{O_{2 max} }