Data from the 1988 National Maternal and Infant Health Survey were used to examine the effect of smoking on the risk of gestational hypertension (GH). GH was defined as the occurrence of two consecutive diastolic blood pressure readings of at least 90 mmHg after the 20th week of gestation in the absence of proteinuria in subjects normotensive prior to pregnancy. One-hundred ten subjects with gestational hypertension were compared with 4371 subjects free of all hypertensive disorders. Smoking during pregnancy was associated with a reduction in risk of GH overall (Odds Ratio (95% CI) = 0.18 (0.07–0.43)) and within each race and parity subgroup. When the protective effect of smoking was examined by race, black subjects appeared to benefit less from smoking compared with whites but the weaker effect in blacks was attributable to less smoking by blacks. Parous subjects experienced a larger reduction in the risk of GH associated with smoking than nulliparous subjects and this persisted at each dose level of smoking. Neither the protective effect of smoking nor the difference by parity appeared to be due to confounding by race, age, weight gain, alcohol use or marijuana use (adjusted OR (95% CI): nulliparous 0.28 (0.09–0.94) vs. parous 0.10 (0.03–0.30)).