The physical changes of puberty coincide with an increase in the salience of peer relationships and a growing risk for depression and other forms of psychopathology. Previously, we reported that pubertal tempo, defined as a child's rate of intraindividual change in pubertal status (measured using parent-reported Tanner stages; Marshall & Tanner, 1970), was associated with changes in boys'--but not girls'--depressive symptoms over and above effects explained by pubertal timing (Mendle, Harden, Brooks-Gunn, & Graber, 2010). The present study extends this previous research by examining changes in the quality of peer relationships in the association between individual differences in pubertal development and change in boys' depressive symptoms. Boys (N = 128, M = 9.61 years, SD = 0.70, at Time 1) were recruited from public schools and assessed annually for 4 years. Results from latent growth curve models indicated that earlier pubertal timing and more rapid pubertal tempo were associated with greater decrements in the quality of boys' peer relationships. After accounting for the association between change in peer relationships and depressive symptoms, the direct effects of pubertal timing and tempo on depressive symptoms were no longer significant. These results highlight a multifaceted approach to studying puberty and emphasize how social mechanisms may intersect with biological risk to produce psychological distress. (Contains 2 tables, 1 footnote and 2 figures.)