Some of the theories that have been developed to explain the origin of bound rubber are critically reviewed and discussed with respect to published data. Theories for carbon black filled compounds and for silica-silicone rubber mixtures are considered; the phenomena involved are likely to be very different, with clear chemical aspects for the latter systems. A common feature emerges, however, from these theories : the area of the polymer-filler interaction site, which is generally considered as a fitting parameter in most approaches. This article concentrates on this aspect and suggests that, with respect to recent findings about the very surface of carbon black particles, an explanation for bound rubber can be offered that considers strong topological constraints exerted by the filler surface on rubber segments. Calculations of interaction site area made with experimental data give values close to a fraction of the half-lateral surface of the structural unit representative of the rubber considered. It follows that the bound rubber variation during storage can now be understood by considering a slow replacement of short rubber chains initially adsorbed on filler particles by larger ones, as demonstrated by calculated data.