OBJECTIVES: Pre-eclampsia (PE) is associated with an increased incidence of cardiovascular disease in later life. Daughters of PE mothers have an increased risk of developing the disease; recent epidemiological data suggest a (grand)paternal contribution. We have directly studied the parents of 673 women with stringently defined PE in relation to their daughters' disease. METHODS: (Grand)parental medical history, current medication and blood pressure (using an Omron 705 automated monitor) were recorded, with obstetric history for the grandmother, including directly verified pregnancy hypertension. RESULTS: The age of the 649 participating grandmothers was 55.5 ± 7.5 years (mean ± SD) and that of the 542 participating grandfathers was 58.0 ± 7.3 years. Essential hypertension (EHT) requiring therapy was present in 23.4% of the grandmothers and 22.8% of the grandfathers. Patients had moderate to severe PE; a quarter were delivered before 34 weeks' gestation. A third of the babies had birthweights below the third centile; the perinatal mortality rate was 2.1%. Grandpaternal absolute systolic pressures and EHT status were highly significant determinants of maternal systolic pressure during gestation (F = 11.8, P < 0.001; F = 8.91, P = 0.003, respectively); maternal body mass index (BMI) had less effect. A similar, less marked, pattern was seen for diastolic pressure (F = 6.01, P = 0.014; F = 11.50, P < 0.0001). Grandmaternal EHT did not influence her daughter's systolic or diastolic pressure (P > 0.2 for both). CONCLUSIONS: A paternal, but not maternal, history of EHT is associated with increased risks of non-pregnant hypertension in the children, the risk being greater in daughters than sons. Pregnancy may unveil or exacerbate this effect, possibly reflecting underlying endothelial vulnerability.