We studied phenotypic and clinical outcome data in an observational, multicenter cohort study of 1001 Western European white women and their singleton babies, with stringently defined moderate-to-severe preeclampsia. Ninety women admitted to being current smokers; 71 had stopped smoking before entry to the study. Across the categories of never-smoker, stopped, and current smoker there were significant increases in the proportion of women delivering before 34 weeks' gestation (P=0.011), delivering a baby below the third birth weight centile (P<0.001), or delivering a baby with any adverse outcome (P=0.011). By comparison with never-smokers, smoking during pregnancy was associated with a doubling of risk of being delivered before 34 weeks' (odds ratio: 1.98; 95% CI: 1.24 to 3.16; P=0.004), of delivering babies below the third centile of corrected birth weight (odds ratio: 2.20; 95% CI 1.41 to 3.44; P<0.0001), or for their babies to have any adverse outcome (odds ratio: 1.87; 95% CI: 1.19 to 2.95; P<0.006). Worryingly, the risk of developing eclampsia was increased 5-fold (odds ratio: 4.88; 95% CI: 1.44 to 16.61; P=0.005). The proportion of smokers in these preeclamptic women was lower than in our pregnant population generally. However, preeclampsia still carries significant perinatal morbidity, and cigarette smoking in preeclamptic pregnancies exacerbates this. Stopping smoking decreases the risks. Smoking in young women should be a particular target for advice by general practitioners before pregnancy, with active encouragement after conception to enroll in such trials as the current Smoking, Nicotine and Pregnancy Trial to support cessation.