Hormonally active environmental exposures may alter timing of puberty. We investigated urinary biomarkers of phenols and age at puberty among girls. The Puberty Studies of the Breast Cancer and Environment Research Program are investigating environmental exposures and onset of puberty in girls. A multi-ethnic cohort study of 1239 girls 6-8 years old was enrolled at three US sites during 2004-2007 with data through 2012 for this analysis. Pubertal stages and other information were collected annually. Urinary metabolites of eight phenols were assayed in baseline samples (benzophenone-3, enterolactone, daidzein, paraben, genistein, 2,5-dichlorophenol, triclosan, bisphenol A, from highest to lowest observed median concentrations). Associations with age at transition to breast (B2) or pubic hair (PH2) stages were estimated using Cox proportional hazards models adjusted for race, ethnicity, body size and caregiver education (aHR, 95% confidence intervals [CI]). Most urinary phenols differed by race/ethnicity and site. Six biomarker concentrations were >100 ug/L for >75% of girls, approaching levels at which effects are seen in animal models. Higher benzophenone-3 concentration was associated with later age at B2 (aHR=0.97 per ln-ug/L; CI=0.94-0.998) but not age at PH2. The aHR was 0.95 per ln-ug/L (CI 0.91-0.98) among non-blacks, possibly as blacks have lower exposures to this sunscreen ingredient. Urinary enterolactone was associated with later age at pubertal development, but effects were entirely explained by body size. Other urinary phenols were not associated with age at onset of puberty. Findings from this prospective study suggest that urinary phenol metabolites measured during childhood are not associated with earlier pubertal onset. The most prevalent phenols may be associated with later pubertal age; however, the association for benzophenone-3 may only exist among non-blacks and the association for enterolactone appears to be completely mediated by body size.