Growth faltering and malnutrition are highly prevalent in most South Asian countries. Among the serious consequences of malnutrition are increased risks of morbidity and mortality in children as well as deficits in physical stature and lowered cognitive measures. Childhood malnutrition in poor households has been well documented in India, with the highest rates observed in those aged 12-23 months. Countrywide National Family Health Survey II (NFHS II) data show mean underweight prevalence increases from 11.9% among infants under six months of age to 58.4% at 12-23 months of aye. This paper discusses the results of an analysis of a longitudinal community-based study of infant growth feeding and care practices in rural south India, for which data have been collected since 1997 and completed in May 2001. In 1998 an intervention began in which a monthly visit for nutrition counselling was made by a trained worker. Mothers were counselled about the need for exclusive breastfeeding for four to six months, choice of appropriate complementary foods, and feeding frequency. Infants enrolled in the study prior to 1998, did not receive this intervention in the first year of life. Using bivariate analysis and regression modelling, weiyht velocity and feeding practices for intervention and non-intervention infants, controlling for a range of socio-economic and demographic factors for infants aged 6 to 10 months, are compared. Both descriptive and multivariate analyses show a statistically significant positive association between intervention and weight velocity for female infants. The intervention group as a whole had improved feeding practices such as avoidance of feeding bottle and increased variety of foods. The implication of this association, in a culture where discrimination against females exists, is promising for the development of low-cost educational interventions to improve nutritional status without the need to provide supplementary food.