(1) To determine the prevalence of overweight and obesity in Australian 4-5-year-old children. (2) To investigate associations between socio-economic characteristics and (a) overweight/obesity and (b) waist circumference. Cross-sectional population survey. Wave 1 (2004) of the Longitudinal Study of Australian Children. Nationally representative sample of 4983 4-5-year-old children (2537 boys and 2446 girls; mean age 56.9 months (s.d. 2.64 months; range 51-67 months)). Prevalence of overweight and obesity (International Obesity TaskForce definitions) and waist circumference (cm). Prevalence estimates were obtained as weighted percentages. Uni- and multivariable ordinal logistic regression (using the proportional odds model) were used to assess associations between potential predictors and the risk of higher child body mass index status and a multivariable linear regression model to assess relationships between the same potential predictors and waist circumference. 15.2% of Australian preschoolers are estimated to be overweight and 5.5% obese. In univariate analyses, seven of the 12 variables were associated with higher odds of being in a heavier body mass index category. In a multivariable regression model, speaking a language other than English (particularly for boys), indigenous status and lower disadvantage quintile were the clearest independent predictors of higher body mass index status, with children in the lowest quintile of social disadvantage having 47% higher odds (95% CI 14, 92%) of being in a heavier body mass index category compared to those in the highest quintile. Waist circumference was not related to any socio-economic variable. This nationally representative survey confirms high rates of overweight and obesity in preschoolers throughout Australia. The recent emergence of a substantial socio-economic gradient should bring new urgency to public health measures to combat the obesity epidemic.