Five groups of children were identified using friendship nominations from the fall and spring of their fifth-grade year: (1) children with a stable best friendship with the same child (same-stable); (2) children with a mutual best friendship at Times 1 and 2, but the best friend was a different child at each time (different-stable); (3) children with a best friendship at Time 1 but not at Time 2 (friendship loss); (4) children who had no best friendship at Time 1 but did have a best friendship at Time 2 (friendship gain); and (5) chronically friendless children. Peer nominations of psychosocial adjustment were gathered at both time points. The friendship gain group became less victimized and the friendship loss group became more victimized by Time 2. The two stable groups of children were rated as prosocial and popular, with low levels of aggression and victimization. Findings suggest that the consistency of having any best friendship across time may be as important to children's adjustment as same-friendship stability. The results of this study also highlight the importance of best friendship "renewal."