Dryland salinity poses a major threat to agricultural production in the wheatbelt of Western Australia and much time and effort is expended on understanding the mechanisms which cause it and on developing techniques to halt or reverse its development. Whilst the location of much dryland salinity can be explained by its topographic position, a significant proportion of it cannot. This study investigated the hypothesis that major faults in the Yilgarn Craton represented in aeromagnetic data by intense curvilinear lows explained the location of areas of dryland salinity not explained by topography. Moreover, the causal mechanisms that might underpin a spatial relationship between major faults and dryland salinity were sought. In one fourth order catchment, nearly 85% of the salinity that was not explained topographically was within 2km of the centre line of a major fault, the remaining 15% being in the other 12km of the catchment. Three groups of similar third order catchments in the western wheatbelt of Western Australia were also investigated; in each case the catchment that was underlain by a major fault had dryland salinity an order of magnitude more than the unfaulted catchment(s). This evidence demonstrates a strong spatial association between major faults and the development of dryland salinity. Other evidence suggests that the underlying mechanism is hydraulic conductivity 5.2 to 2.9 times higher inside the fault zone compared to outside it and shows that geomorphology, salt store, regolith thickness, and degree of clearing are not the underlying mechanisms. In one of the groups of catchments, it has been calculated that an amount of recharge, significant in relation to recharge from rainfall, was entering from an adjacent catchment along a major fault. The paper concludes that geological features such as major faults affect the development of dryland salinity in the wheatbelt of Western Australia because of permeability differences in the regolith and therefore computer models of salinity risk need to take these differences into account. Techniques need to be developed to map, quickly and relatively cheaply, the geology-related permeability differences over wide areas of the landscape.