This study explored the association between sodium excretion and blood pressure (BP). A new method was used to minimize the measurement error introduced by the large intrinsic variability of 24-hour sodium excretion. The ratio of intra- to interindividual variation was used to estimate the number of measurements needed to characterize the individual. When seven consecutive 24-hour samples were collected from 73 children, ages 11-14 years, a significant correlation between mean individual sodium excretion and BP was demonstrated. The independent relationship persisted when controlling for height, weight, pulse, age, sex and race (p = 0.045), but was eliminated by simultaneously considering mean creatinine excretion. Although the cross-sectional association described is quantitatively weak, a linear relationship between BP and sodium over the range consumed in this society could be important for prevention.