The importance of chronological age (CA) and skeletal age (SA) in explaining variation in somatic dimensions, and the independent contributions of CA, SA, stature (ST) and weight (WT) to variability in physical fitness were investigated in a sample of 6593 girls 6-16 years of age. Body dimensions included lengths, breadths, circumferences, skinfolds, and Heath-Carter somatotype, while fitness tests included measures of health- and performance-related fitness, and cardiovascular and lung functions. Age-specific correlations were calculated between SA and anthropometric dimensions, fitness tests and cardiovascular and lung functions, while age-specific stepwise multiple regressions were used to investigate the relaitive importance of SA, CA, ST and WT in explaining fitness and cardiovascular and lung functions. SA is most highly correlated with lengths and then with breadths, circumferences and skinfolds in this order. SA per se or in interaction with CA is the (only significant predictor of somatic characteristics. Among fitness items, physical working capacity and static strength correlate highest with SA. Bent arm hang, leg lifts and sit-ups correlate negatively with SA but values are low, while all other components correlate at non-significant or low levels. Results of the multiple regression analysis indicate that, with few exceptions, CA, SA, ST and WT and their interactions explain less than 10 % of the variance in most physical fitness items. However, for PWC, arm pull strength, and bent arm hang, the interaction terms explain between 12 % and 67 % of the variance.