Australian Centre for International Agricultural Research Canberra 2003A mark–capture–release (CMR) study was carried out in Morogoro, Tanzania, from April 1999 to August 2000 to investigate the effect of slashing and burning versus tractor ploughing on the population of rodents in agricultural fields. We found that the spatial distribution of individuals was significantly affected by the land preparation method. The coefficient of dispersion values (based on variance-to-mean ratio calculations) indicated that more animals clustered around the edges in tractor-ploughed fields whereas in the slashed-and-burnt fields, animals were randomly distributed. Before land preparation, animals were randomly distributed everywhere. This suggests that the slashing-and-burning practice does not affect the rodent population distribution in crop fields while tractor ploughing does affect rodents, probably by reducing cover and food availability or even by killing some individuals. Yet, it seems useful as a management tool when it is practised over a large area and if the surrounding fallow lands, which act as donor habitat, are cleared.