This contribution aims to summarize the present theoretical methods used in treating gas migration in UO2, and to review briefly the extent to which predictions of such theories for the case of Xe in UO2, are compatible with experiment. The theory has two components: first the statistical mechanical treatment of gas trapping and migration; and secondly the calculation by computational methods of those energy terms which are shown by the statistical mechanics to be significant. The applicability of the results to the analysis of the experimental data depends upon certain crucial assumptions concerning the extent of thermodynamic control of the system. These problems will be described after our account of the theory which we present in the next section.