Previous studies in a population of 721 children aged 2--14 years demonstrated the familial aggregation of blood pressure in children, and a significant regression coefficient (b = 0.25) of follow-up on initial blood pressures over a four-year period. Urinary kallikrein concentration (UKal) also aggregated in families, was lower in black than in white children and was inversely related to blood pressure. Further studies in the same cohort have been conducted. These variables were again measured in 484 children in 129 families seven to eight years after the initial blood pressure and three to four years after the original UKal measurements were made. Familial aggregation again was found for blood pressure and urinary kallikrein. Blood pressure tracking was demonstrated by the finding that blood pressure scores at the third survey were related significantly to those at both previous surveys. Kallikrein concentrations in casual urines at Survey 3 were related to those obtained at Survey 2 (r = 0.499), and were again significantly lower in black than in white children (log = 3.84 +/- 0.8 vs 4.37 +/- 0.7; P less than 0.001). These were significant inverse relations between UKal/creatinine concentration and blood pressure in both white and black children. Thus, familial aggregation of blood pressure, blood pressure rank and concentration of kallikrein in casual urine specimes are relatively stable in children over an eight-year period of observation. This study demonstrates in a free living population of normal children, a stable relation between blood pressure and an enzyme which is involved in the production of potent vasodilator peptides and is related to hypertension in adults.