Mathematical models used previously to obtain transport parameters from experimental investigations of bacterial transport in packed columns by Olson et al. (2005) yielded apparent tortuosity values that were an order of magnitude higher than expected for packed beds. The association of motile bacteria with solid surfaces of porous media, due to their swimming, contributed to the large apparent tortuosity values evaluated from packed column experiments; a modification of the transport equations to account for surface association yielded tortuosity values that were physically more reasonable. Although it was not possible to uniquely determine values of the kinetic parameters for surface association because of limitations with the experimental methods used by Olson et al., it was possible to constrain the range of values. A theoretical derivation of the mean residence time for swimming bacteria at a surface provided qualitative and quantitative support for including surface association terms in the transport model. This analysis shows that surface association can be a significant factor in modeling bacterial migration, and consideration of bacterial swimming properties such as run lengths and turn angles is important for predicting bacterial transport in saturated granular media typical of groundwater environments.