Objective: To investigate the relation between maternal fish intake in late pregnancy and the frequency of low birth weight and intrauterine growth retardation (IUGR). Participants: 11 585 pregnant women in south west England. Methods: Information on fish intake was obtained from a food frequency questionnaire sent to the women at 32 weeks' gestation, and used to calculate n-3 fatty acid (n-3FA) intake from fish. IUGR was defined as a birth weight for gestational age and sex below the 10th centile. Confounding variables considered included maternal age, height, weight, education, parity, smoking and drinking in pregnancy, and whether the mother was living with a partner. Only singleton, liveborn infants were included. Main results: Mean daily intakes of fish and n-3FAs were 32.8 g and 0.147 g respectively. In unadjusted analyses there were positive associations between mean birth weight and fish intake or n-3FA intake, but these disappeared on adjustment for potential confounders. The frequency of IUGR decreased with increasing fish intake-the OR (95%CI) of IUGR in those eating no fish was 1.85 (1.44 to 2.38) compared with those in the highest fish intake group. On adjustment this relation was attenuated (adjusted OR 1.37 (1.02 to 1.84)), but the decline in the frequency of IUGR with increasing fish intake remained statistically significant. No relation was observed between mean gestation and fish or n-3FA intake. Conclusions: These results lend some support to the hypothesis that raising fish or n3-FA intake during pregnancy may increase fetal growth rate. However, they provide no evidence that increasing fish consumption is associated with an increase in mean gestation.