Previous research, mainly from developed populations, has shown that a variety of early life factors (prenatal and postnatal), and socio-demographic factors that influence early life growth and development, are important determinants of childhood cognitive function. The main objective of the current study was to examine maternal nutritional indices, body size at birth, postnatal growth and infant feeding practices as predictors of childhood cognitive function independently of socio-demographic factors, in an Indian population. The Mysore Parthenon Cohort comprises 674 babies born in the Holdsworth Memorial Hospital, Mysore, India, in 1997-1998. Maternal anthropometry and serum vitamin D, vitamin B12, folate and homocysteine concentrations were measured at 30±2 weeks of gestation. Detailed newborn anthropometry was performed. Data on breast-feeding was collected during the first, second and third year follow-up visits. The children had repeat anthropometry every 6-12 months until adolescence. Their cognitive function was assessed at age 9.5 years (n=542) and 13.5 years (n=545) using three core tests from the Kaufman Assessment Battery for Children and additional tests measuring learning, long-term retrieval/storage, short-term memory, reasoning, verbal fluency, visuo-spatial ability, and attention and concentration. Data on the parents' socio-economic status and education, maternal intelligence and home environment were recorded. For each SD increase in birthweight, length, head (HC) and mid-upper-arm (MUAC) circumferences, and sum of skinfolds (SS) at birth there was a 0.10-0.14 SD increase in cognitive scores, independent of socio-demographic confounders (P<0.05). The strongest associations were with learning and visuo-spatial ability. Effects of similar size were found for gain in length, HC and MUAC during infancy and early childhood, and gain in SS during late childhood (p<0.05). HC at birth and head growth during early childhood were the strongest predictors, independent of other body measurements. Offspring cognitive scores increased by 0.12-0.16 SD per each SD increase in maternal serum folate concentrations during pregnancy (p<0.05). Maternal vitamin B12 concentrations were inversely related to some cognitive domains in the child (p<0.05), apparent mainly in Hindus. Breast-feeding duration was unrelated to children's cognitive function. There were positive and independent associations of maternal education and intelligence with children's cognitive function (p<0.001). The variance in cognitive function explained by maternal folate status (1-3%), newborn HC and childhood head growth (4-9%) and current socio-demographic factors (6-17%) appeared to be additive (total variance explained 16- 25%). In conclusion, maternal folate status in pregnancy, HC at birth, head growth during infancy and early childhood, and current socio-demographic factors independently predict childhood cognitive function in this cohort. The findings suggest a need for investment in integrated interventions, targeting both the mother and the child, to improve prenatal and postnatal nutrition and maternal education, in order to promote optimal cognitive development among Indian children.