We analyzed data from a population-based case control study to determine whether maternal prenatal smoking or alcohol drinking might increase the risk of craniosynostosis. Between 1986 and 1989, the Colorado Craniosynostosis Registry ascertained 233 children whose diagnoses were confirmed by an independent radiologist. Of the 212 (91%) cases who participated 86 had lambdoid, 69 had sagittal, 25 had coronal, 18 had metopic, and 14 had multiple suture synostosis. Random sampling of birth certificate records yielded 367 eligible controls, of whom 291 (79%) participated. Study staff interviewed case and control mothers between 1989 and 1991 for prenatal smoking and alcohol drinking. We used Mantel-Haenszel and polychotomous logistic regression techniques to estimate the relative odds of craniosynostosis for smoking and drinking in all cases and in relevant case subgroups. Smoking was associated with a relative odds of craniosynostosis of 1.7 (95% confidence interval, 1.2-2.6). For smoking of more than one pack per day, the relative odds were 3.5 (1.5-8.4) for all types of synostosis combined, and 5.6 (2.1-15.3) for coronal synostosis. No strong or significant association was noted for drinking. Maternal prenatal smoking may increase the risk of craniosynostosis in this population.