This review examined recent evidence on associations between objectively measured habitual physical activity and adiposity. A search for observational studies was carried out using several electronic databases from June 2004-June 2008. Of 1 255 potentially eligible papers, 47 papers were included, which described 48 studies. Most studies (41/48; 85%) were cross-sectional and 31/48 (65%) used proxies for adiposity, such as body mass index (BMI) or BMI z-score as the outcome measure. Few studies (10%; 5/48) focused on pre-school children. There was consistent evidence of negative associations between objectively measured physical activity and adiposity: significant negative associations were observed in 38/48 (79%) of studies overall. The present review supports the hypothesis that higher levels of habitual physical activity are protective against child and adolescent obesity. However, prospective longitudinal studies are warranted; there is a need for more research on younger children, and for more 'dose-response' evidence.