OBJECTIVE: To study whether differences in delivery outcome correlated with pollution levels in the municipalities of a county in southern Sweden (Malmöhus). METHODS: The effects of air and ground pollution on delivery outcome were studied in 38,718 women who lived, and were delivered, in Malmöhus county in southern Sweden during the period 1985-90. Delivery outcome was compared between different municipalities arranged according to the degree of pollution. RESULTS: Neither air nor ground pollution affected the odds ratio for short gestation or low birthweight. No significant effect was seen on perinatal deaths or non-trivial malformations, but in the municipality with the highest arsenic concentration, 14 infants died perinatally versus the expected 7.4, and the odds ratio for stillbirths reached formal statistical significance. CONCLUSIONS: Environmental pollution in this county does not seem to be a major determinant of delivery outcome.