This study used multilevel regression modelling to longitudinally investigate the influences of age, sex, body size, skinfold thicknesses and maturity on the development of isokinetic knee extension and flexion on eight occasions over a 4-year period. Forty-one subjects (20 boys and 21 girls) were measured and 295 isokinetic leg strength tests and associated measures were successfully completed. Subjects were aged 10.0 +/- 0.3 years at the onset of the study. Stature, body mass, triceps and subscapular skinfold thicknesses, and sexual maturation (according to pubic hair development) were assessed at each test occasion. Isokinetic concentric knee extension and flexion of the dominant leg were determined to elicit maximal peak extension (PET) and flexion torque (PFT). Statistical significance was accepted at p < 0.05. Multilevel regression modelling indicated that stature and mass were significant predictors of both PET and PFT. Age and maturity were non-significant explanatory variables once stature and mass had been accounted for. Skinfold thickness exerted a significant negative effect independent of mass and stature on PFT but not PET. At test occasion 8, cross-sectional areas (CSAs) of the knee extensors (ExCSA) and flexors (FlexCSA) were determined using magnetic resonance imaging on 23 boys and 14 girls and examined as predictors of isokinetic leg strength. There were no significant sex-related differences in PET or PFT. Pearson product moment correlation coefficients indicated a significant relationship between ExCSA and PET and FlexCSA and PFT for both boys and girls. Analysis of covariance (ANCOVA) demonstrated that ExCSA and FlexCSA were significant explanatory variables for PET and PFT, respectively, but became non-significant once stature and mass had been introduced into the analysis. To conclude, there were no significant sex differences in PET or PFT between the ages of 10 and 14 years and the development of PET and PFT could be accounted for by the increase in stature and mass. Age, maturity and thigh muscle CSA were all non-explanatory variables in the production of PET and PFT once body size had been controlled for.