Until recently, interpretation of swelling equilibrium experiments rested on the Flory-Rehner equation developed in 1943 for networks which deform affinely. The relationship made possible at least the ranking of series of networks of the same polymer, according to their cross-link densities. However, the theory does not explain the important fact (first observed by Gee in 1965) that there is a maximum in the dependence of λ ln (a 1 c/a 1 u) on λ (where λ is the isotropic deformation v2 −1/3, v2 the volume fraction of polymer in the polymer-solvent system, and a 1 u and a 1 c the solvent activities respectively in uncross-linked and cross-linked polymers). Decisive progress in this field has been achieved by the formulation of a molecular theory of real networks by Flory and Erman in 1979. The present paper reviews the main theoretical advances concerning swelling and swelling equilibrium, and describes useful methods for characterizing both model networks (of controlled structure) and networks in which the cross linking is highly random and essentially uncontrolled.