The upward migration of radionuclides in the (238)U decay series in soils and their uptake by plants is of interest in various contexts, including the geological disposal of radioactive waste and the remediation of former sites of uranium mining and milling. In order to investigate the likely patterns of behaviour of (238)U-series radionuclides being transported upward through the soil column, a detailed soil-plant model originally developed for studying the behaviour of (79)Se in soil-plant systems has been adapted to make it applicable to the (238)U series. By undertaking a reference case simulation and a series of sensitivity studies, it has been found that a wide variety of behaviour can be exhibited by radionuclides in the (238)U decay chain in soils, even when the source term is limited to being a constant flux of either (238)U or (226)Ra. Hydrological conditions are a primary factor, both in respect of the overall advective flow deeper in the soil, which controls the rate of upward migration, and in the influence of seasonally changing flow directions closer to the soil surface, which can result in the accumulation of radionuclides at specific depths irrespective of changes in sorption between the oxic and anoxic regions of the soil. However, such changes in sorption can also be significant in controlling the degree of accumulation that occurs. This importance of seasonally varying factors in controlling radionuclide transport in soils even in very long-term simulations is a strong argument against the use of annually averaged parameters in long-term assessment models. With a water table that was simulated to fluctuate seasonally from a substantial depth in soil to the surface soil layer, the timing of such variations in relation to the period of plant growth was found to have a major impact on the degree of uptake of radionuclides by plant roots. In long-term safety assessment studies it has sometimes been the practice to model the transport of (226)Ra in soil, but to assume that both (210)Pb and (210)Po can be treated as being present in secular equilibrium with the (226)Ra. This simplification is not always appropriate. Where geochemical conditions are such that the (226)Ra migrates upward in the soil column faster than (210)Pb and (210)Po, disequilibrium is not a significant issue, as the (226)Ra supports (210)Pb and (210)Po at concentrations somewhat below those estimated on the basis of assumed secular equilibrium. However, for low, but realistic, values of the distribution coefficients for (210)Pb and (210)Po and high, but realistic, distribution coefficients for (226)Ra, the (210)Pb and (210)Po can reach the surface soil in high concentrations that are not locally supported by (226)Ra. This means that models based on the assumption of secular equilibrium should not be employed without a careful consideration of the hydrological and hydrochemical situation of interest.