Research Findings: We examined low-income children's temperament (regulatory and reactive) as a predictor of their self-regulation, and teacher-child relationship (closeness and conflict) as a moderator of associations between child temperament and self-regulation. This study involved 291 children (132 girls) (Mage = 53.88 months, SD = 6.44 months) from three EduCare programs. Parents reported on children's temperament and teachers reported on qualities of teacher-child relationships during fall. Direct assessments of self-regulation were conducted during the following spring and summer. Hierarchical regression models using SAS PROCMIXED were employed to account for nesting of children within classrooms. Bivariate analyses revealed that teacher-child closeness was positively associated with children's self-regulation, and teacher-child conflict was inversely associated with children's self-regulation. After controlling for demographic variables, regression analyses showed that higher levels of conflict combined with lower temperamental regulation was related to lower self-regulation. Lower levels of child temperamental regulation was related to higher self-regulation when teacher-child conflict was low. Practice and Policy: Findings suggest that reducing conflictual teacher-child conflict could be beneficial for children's self-regulation, particularly for children with low regulatory temperament. A focus on enhancing teacher self-regulation, for example, through mindfulness practices, is a promising approach to reducing teacher-child conflict.