The Astrand-Ryhming est was modified to overcome problems of predicting V̇O2max in children. V̇O2max was measured directly during a maximal treadmill test and estimated by means of a submaximal protocol in a test group of 51 subjects ages 10-18 years. A multiple regression equation was developed with directly measured V̇O2max as the dependent variable and age, body weight, and V̇O2max estimated from the Astrand nomogram as independent variables. The validity and reliability of this equation to predict V̇O2max in children and adolescents was cross-validated. No significant difference was found between the measured V̇O2max and that estimated from the equation. The correlation coefficient between measured and estimated V̇O2max for the cross-validation group was 0.89, with a standard error of estimate of 12%. Test-retest reliability was 0.95. It was concluded that this modification of the Astrand nomogram provides a valid and reliable prediction of V̇O2max in children and adolescents.