We prospectively evaluated the mental (MeDQ) and motor (MoDQ) developmental quotients of 395 (67.5% of the eligible) infants of mothers with epilepsy (IME) (mean age: 15 months) enrolled in the Kerala Registry of Epilepsy and Pregnancy between 1998 and 2004. The same developmental pediatricians, blinded to antiepileptic drug (AED) exposure, evaluated the children using the Indian adaptation of the Bayley Scale of Infant Development: Their mean MeDQ was 89.1+/-29.9 and mean MoDQ was 90.7+/-26.9. The MeDQ and MoDQ were impaired (<84) for 150 (37.6%) and 133 (33.5%) IME, respectively. Maternal age, type of epilepsy, seizure frequency, or use of folic acid did not correlate with the mean MeDQ or MoDQ. Maternal education was significantly correlated with the MoDQ, but not with the MeDQ, of the infants. Infants not exposed to AEDs (n=32) had a higher MeDQ (mean: 92.3, 95% CI: 81.4-103.2) and MoDQ (mean 94.7; 95% CI 84.9-104.5) than those exposed to AEDs (MeDQ--mean: 88.6, 95% CI: 85.5-91.6; MoDQ--mean: 90.0, 95% CI: 87.3-92.8). Those exposed to polytherapy had significantly lower developmental quotients than those exposed to monotherapy. Cumulative AED scores during pregnancy had an inverse relationship with developmental quotients. On multiple regression analysis, polytherapy was a stronger predictor of lower developmental quotients than dosage. Compared with carbamazepine monotherapy, valproate monotherapy was associated with significantly lower MeDQ and MoDQ in IME (93.1 and 95 vs 86.9 and 86.1), but the differences between other AEDs were not significant for IME exposed to valproate monotherapy. A limitation of the study is that the influence of maternal intelligence on developmental quotients was not evaluated.