Grow and post (GP) access methods, e.g., B+trees, are the dominant form of index tree access method because of properties not strictly related to search performance. GP methods fit well with the rest of a database system, and indeed profit from their inclusion, e.g, search performance is improved by caching index nodes. Enhancements to GP methods have increased their utility. GP methods solve the multi-attribute point search problem and, more speculatively, the spatial search problem. Their simplicity and flexibility make GP methods applicable in several interesting new areas. This paper examines these topics from the author's personal perspective.