Objective: To determine possible relations between early adiposity rebound and adolescent development. Methods: Prospective children cohort from 2 kindergartens selected through clustering convenience sampling method in Anhui Province was established since Sep. 2010. Participants were classified as Cohort 1 (2010), Cohort 2 (2011) and Cohort 3 (2012) according to the recruiting year. Till Sep. 2015, a toal of 802 girls were included in this study, and received follow-up till primary school. During kindergarten period, physical examination was carried out every 3 months, 8 times in total. In primary school, physical examination was carried out annually; till Sep.2015, the cohort 1, 2, 3 took physical examination for 12, 11, 10 times, respectively. Information on household economic status and child physical activity was acquired through parents questionnaire survey, and breast development were assessed through visual inspection and palpation. Adiposity rebound was determined according to Rolland-Cachera's method for each girl. Differences between early adiposity rebound and normal adiposity rebound groups were compared by using t test and χ(2) test. Multivariate regression analysis was applied to explore the association between early adiposity rebound and breast development. Results: The average age of participants was (8.90±0.87) years old and the BMI was (17.48±2.70) kg/m(2). The average age at adiposity rebound was (6.16±0.90) years old and the BMI was (15.33±1.82) kg/m(2). Premature breast development was found significantly higher in girls in early adiposity rebound group (27.8%, 54/802) than it in normal adiposity rebound group (13.7%) (P<0.001). After current adiposity, age, household economic status, childhood physical activity adjusted, the OR of premature beast development in early adiposity rebound group was 2.41(95%CI: 1.41-4.12). Conclusion: Early adiposity rebound increases the risk of premature puberty in girls.