Relationships between echocardiographic dimensions and the Heath-Carter anthropometric somatotype were considered in healthy, non-obese children (8-11 year olds, n = 143), adolescents (12 15 year olds, n = 216) and young adults (16-24 years, n = 190). Cardiac dimensions, measured by M-mode echocardiography at end-diastole, included left ventricular internal diameter (LVIDd), posterior wall thickness (PWTd), and interventricular posterior wall thickness (STd). Left ventricular mass (LVM) and left ventricular end-diastolic volume (LVEDV) were estimated. Partial correlations between cardiac dimensions and each somatotype component were calculated, controlling for age and the other two components. Only 9 out of 45 correlations in males and 7 of 45 correlations in females were significant (p < or = 0.05). LVM, LVEDV, and LVIDd were significantly related to somatotype in males, demonstrating significant positive correlations with mesomorphy (r = 0.25, 0.29 and 0.29, respectively) and ectomorphy (r = 0.22. 0.37, and 0.37, respectively), and LVEDV and LVIDd were related to endomorphy (r = 0.24 and 0.25, respectively) in 8-11 year old boys. In 8-11 year old females, endomorphy was related to STd (r = 0.41) and LVM (r = 0.34), while mesomorphy was related to PWTd (r = -0.34) and ectomorphy was related to PWTd (r = -0.36). In 12-15 year old females, mesomorphy was related to STd (r = 0.26) and in 16-24 year old females, endomorphy was related to LVIDd (r = 0.29) and LVEDV (r = 0.32). Overall, the correlations between somatotype and cardiac dimensions were low, ranging from -0.36 to +0.41, with no clear pattern in either sex. Additionally, a backward stepwise regression analysis indicated that body size was more important in predicting echocardiographic dimensions than somatotype. Thus, physique, as estimated with the Heath-Carter anthropometric somatotype, is not related to echocardiographic dimensions in children, youths and young adults.