Background: A few prospective studies suggest an association between maternal smoking during pregnancy and lower risk of type 1 diabetes. However, the role of unmeasured confounding and misclassification remains unclear. Methods: We comprehensively evaluated whether maternal smoking in pregnancy predicts lower risk of childhood-onset type 1 diabetes in two Scandinavian pregnancy cohorts (185,076 children; 689 cases) and a Norwegian register-based cohort (434,627 children; 692 cases). We measured cord blood cotinine as an objective marker of nicotine exposure during late pregnancy in 154 cases and 476 controls. We also examined paternal smoking during pregnancy, in addition to environmental tobacco smoke exposure the first 6 months of life, to clarify the role of characteristics of smokers in general. Results: In the pregnancy cohorts, maternal smoking beyond gestational week 12 was inversely associated with type 1 diabetes, pooled adjusted hazard ratio (aHR) 0.66 (95% CI = 0.51, 0.85). Similarly, in the Norwegian register-based cohort, children of mothers who still smoked at the end of pregnancy had lower risk of type 1 diabetes, aHR 0.65 (95% CI = 0.47, 0.89). Cord blood cotinine ≥30 nmol/L was also associated with reduced risk of type 1 diabetes, adjusted odds ratio 0.42 (95% CI = 0.17, 1.0). We observed no associations of paternal smoking during pregnancy, or environmental tobacco smoke exposure, with childhood-onset type 1 diabetes. Conclusion: Maternal sustained smoking during pregnancy is associated with lower risk of type 1 diabetes in children. This sheds new light on the potential intrauterine environmental origins of the disease.