The wear and friction behavior of ultralow wear polytetrafluoroethylene (PTFE)/alpha-alumina composites first described by Burris and Sawyer in 2006 has been heavily studied, but the mechanisms responsible for the 4 orders of magnitude improvement in wear over unfilled PTFE are still not fully understood. It has been shown that the formation of a polymeric transfer film is crucial to achieving ultralow wear on a metal countersurface. However, the detailed chemical mechanism of transfer film formation and its role in the exceptional wear performance has yet to be described. There has been much debate about the role of chemical interactions between the PTFE, the filler, and the metal countersurface, and some researchers have even concluded that chemical changes are not an important part of the ultralow wear mechanism in these materials. Here, a stripe test allowed detailed spectroscopic studies of PTFE/alpha-alumina transfer films in various stages of development, which led to a proposed mechanism which accounts for the creation of chemically distinct films formed on both surfaces of the wear couple. PTFE chains are broken during sliding and undergo a series of reactions to produce carboxylate chain ends, which have been shown to chelate to both the metal surface and to the surface of the alumina filler particles. These tribochemical reactions form a robust polymer-on-polymer system that protects the steel countersurface and is able to withstand hundreds of thousands of cycles of sliding with almost no wear of the polymer composite after the initial run-in period. The mechanical scission of carboncarbon bonds in the backbone of PTFE under conditions of sliding contact is supported mathematically using the Hamaker model for van der Waals interactions between polymer fibrils and the countersurface. The necessity for ambient moisture and oxygen is explained, and model experiments using small molecules confirm the reactions in the proposed mechanism.