The concentration of plasma immunoreactive atrial natriuretic peptide is positively associated with right atrial and pulmonary capillary wedge pressure, suggesting that blood volume and hence atrial pressure govern its release. Expansion of plasma volume is a central physiological adjustment in normal pregnancy. Conversely, pregnancies complicated by pre-eclampsia are associated with a reduction in plasma volume and central venous pressure. A study was therefore undertaken to test the hypothesis that plasma atrial natriuretic peptide concentrations are low in pre-eclampsia owing to deficient secretion. Concentrations of the peptide were measured by a specific radioimmunoassay. The mean plasma immunoreactive atrial natriuretic peptide concentration in healthy pregnant women (n = 22; third trimester) was higher (56 (1 SD 29) ng/l) than in 25 young, non-pregnant controls (37 (19) ng/l). Concentrations in patients suffering from mild pre-eclampsia (n = 9) were higher (127 (60) ng/l) than in normal pregnant women, and in patients with severe pre-eclampsia (n = 6) concentrations were higher still (392 (225) ng/l). Despite failure of plasma volume expansion and low central venous and pulmonary capillary wedge pressures in pre-eclampsia this condition is associated with greatly increased plasma concentrations of plasma immunoreactive atrial natriuretic peptide, which increase still further with the severity of the disease. These findings are clear evidence that atrial pressure may not be the principal determinant of the release of the natriuretic peptide in pre-eclampsia.