Objective: To examine whether an FFQ can be used for assessing exposure to methylmercury (MeHg) by estimating MeHg intake from seafood consumption using the FFQ and confirming the accuracy of the estimated value. Design: Seafood consumption of pregnant women was assessed using the FFQ. Total mercury (T-Hg) concentrations of maternal red blood cells (RBC) and hair were measured as exposure indices of MeHg. Setting: A prospective birth cohort study, the Tohoku Study of Child Development (TSCD), which has been ongoing since 2001. Subjects: The subjects were 609 pregnant Japanese women who were enrolled in the TSCD. Results: MeHg intake was estimated from seafood consumption determined using the FFQ and the MeHg concentrations in each type of seafood. The accuracy of the estimated value was confirmed by comparison with T-Hg in RBC and hair. Estimated MeHg intake was 42?3 mg/week, and 43?0 % of that was from large predatory fish. Compared with the Japanese tolerable weekly intake, in total 12?5 % of the subjects exceeded it. T-Hg concentrations in RBC and hair were significantly correlated with estimated MeHg intake: r 5 0?325 (P , 0?0001) for RBC and r 5 0?305 (P , 0?0001) for hair. Conclusions: Estimated MeHg intake based on the FFQ was significantly associated with T-Hg concentrations in RBC and hair. Although the estimated value involves uncertainties, the FFQ appears to be a useful tool for assessment of exposure to MeHg.