Interactions between the weeds Bromussterilis L., Galium aparine L. and Papaver rhoeas L. were investigated over 3 years of continuous winter wheat (Triticum aestivum L.) across a factorial combination of organic and conventional fertilizer, and ploughing and hand-roguing contrasted with minimum tillage and herbicide. The species were sown separately and together at 50 seeds m−2 per species at the start of the experiment in October 1989. In addition, there were weed-free and background-weed controls. Weed densities were monitored at roughly monthly intervals and crop yield recorded for three seasons. B. sterilis populations increased 10-fold under minimum tillage, but declined under ploughing. Densities of P. rhoeas remained largely low. G. aparine increased on the organically fertilized minimum-tillage plots, except where B. sterilis was present; the high densities of B. sterilis reduced the population size of G. aparine. Crop yield was influenced most strongly by the fertilizer treatment in the first season, but later the density of B. sterilis was by far the most important factor; the evidence of interactive effects of the different weed species on crop yield was weak at best. Interactive effects of arable weed species can be observed, but only at very high densities, and so are unlikely to be of widespread economic importance.