Objective Magnesium affects blood pressure by modulating vascular tone and reactivity. In obstetric patients, magnesium is administered to prevent eclamptic seizures and as a tocolytic to treat preterm labour. Prior to studying vascular sensitivity in women with pre-eclampsia, we sought to determine the effect of magnesium on venous tone in healthy women of childbearing age. Design Dose-response study.Setting Columbia-Presbyterian Medical Center, New York.Population Ten healthy non-pregnant women (age range 21-47 years).Methods Vascular response to magnesium sulphate (MgSO 4 ) was measured in a dorsal hand vein using the linear variable differential transformer (LVDT) technique. Complete dose -response curves to MgSO 4 (0.0000275 -0.088 mmol/min) were determined after 50% preconstriction of the vein with phenylephrine. Main outcome measures The ED 50 of MgSO 4 (dose resulting in 50% of maximal effect, E max ) was determined. Blood samples from the contralateral upper extremity were obtained to assess total plasma magnesium levels at baseline and at the highest infused dose of magnesium. ED 50 results are expressed as geometric mean (95% confidence interval, CI). E max results and magnesium plasma concentrations are expressed as mean [SD]. Results The ED 50 of MgSO 4 was 0.000307 mmol/min (95% CI 0.138, 0.666) and E max was 102% [20%].Conclusion Magnesium induces dose-dependent venodilation in healthy women in the absence of systemic haemodynamic effects. The dose resulting in vasodilation using the LVDT/hand vein model is two to three orders of magnitude less than the therapeutic doses of magnesium used for tocolysis or seizure prophylaxis. Studies of the effect of systemically administered therapeutic doses of magnesium on vascular reactivity in pre-eclampsia will be of interest.