Inverse associations between size at birth and blood pressure (BP) in later life are commonly statistically significant only after adjustment for current size, consistent with change in size as the determinant. Few studies have been prospective or have included a range of potential confounders. Using regression models, including maternal and demographic variables, we examined associations between size at birth and BP in Australian children followed from week 16 of gestation to the age of 8 years. BP measurements were available from 1417 children born after 37 weeks gestation without congenital abnormalities. In models adjusted only for sex, the birthweight (BW), birth length, ponderal index, head circumference, chest circumference, abdominal girth, mid-arm circumference, triceps skinfold, placental weight, or BW/placental weight ratio did not significantly predict SBP in 8-year-olds. With adjustment for current size, associations were inverse but not statistically significant (regression coefficients: BW, -1.11; 95% confidence limits [CL], -2.22, 0.01; birth length, -0.25; 95% CL, -0.52, 0.24) and remained nonsignificant after adjustment for confounders. Current weight, height, or body mass index significantly predicted SBP and DBP (P<0.001) with differences of 8/4 mm Hg between upper and lower quartiles; effects were similar in infants with lower and higher BW. These findings are consistent with postnatal change in size as the major determinant of BP in 8-year-olds and are important in the context of the worldwide "epidemic" of obesity in childhood as a likely precursor of increasing rates of hypertension in adults.