Identifying early risk factors for childhood obesity is critical, as weight in infancy and early childhood tracks to later periods. Continued bottle use – primarily from excess milk intake – is emerging as a potential risk factor for early childhood overweight. Over three fourths of US infants drink from bottles beyond the recommended weaning age of 12 months, and two thirds of UK infants use a bottle at 18 months. This paper is divided into three parts. Part 1 reviews the literature on beverage intake, weight and bottle use in young children. Part II describes pilot data on milk bottle use and weight in 12–60-month-olds, collected prior to a randomized controlled (RCT) trial of a bottle-weaning intervention. Median daily milk bottle consumption at 12 months was 5.0 (interquartile range = 3–6). Among 12–36-month-olds, current users were significantly more likely to be >95th% weight-for-height (19% vs. 0%, P < 0.02), and more were >85% weight-for-height (27% vs. 11%, P < 0.11), vs. non-users. In contrast, current bottle use was not associated with either overweight or obesity in 37–60-month-olds. Part III describes the RCT, begun in fall 2008. It is enrolling 464 parent/12-month-old dyads from a nutrition assistance programme for low-income families. Children's bottle use, anthropometrics, dietary intake and nutrient density (via 24 h recall) are assessed quarterly through 24 months of age. For the intervention, site nutritionists employ a project-developed, visually attractive flip chart. An observational study nested within the RCT will describe dietary changes during this period of feeding transitions.