Animal and cross-sectional epidemiological studies suggest that prenatal lead exposure is related to delayed menarche, but this has not been confirmed in longitudinal studies. We analyzed this association among 200 girls from Mexico City who were followed since the first trimester of gestation. Maternal blood lead levels were analyzed once during each trimester of pregnancy, and daughters were asked about their first menstrual cycle at a visit between the ages of 9.8 and 18.1 years. We estimated hazard ratios (HRs) and 95% confidence intervals (CI) for probability of menarche over the follow-up period using interval-censored Cox models, comparing those with prenatal blood lead level ⩾5 µg/dl to those with prenatal blood lead <5 µg/dl. We also estimated HRs and 95% CI with conventional Cox regression models, which utilized the self-reported age at menarche. In adjusted analyses, we accounted for maternal age, maternal parity, maternal education, and prenatal calcium treatment status. Across trimesters, 36−47% of mothers had blood lead levels ⩾5 µg/dl. Using interval-censored models, we found that during the second trimester only, girls with ⩾5 µg/dl prenatal blood lead had a later age at menarche compared with girls with prenatal blood lead levels <5 µg/dl (confounder-adjusted HR=0.59, 95% CI 0.28–0.90; P=0.05). Associations were in a similar direction, although not statistically significant, in the conventional Cox regression models, potentially indicating measurement error in the self-recalled age at menarche. In summary, higher prenatal lead exposure during the second trimester could be related to later onset of sexual maturation.