In a semiarid Mediterranean site in central Spain, field experiments were conducted on a Calcic Haploxeralf (noncalcic brown soil), which had been managed with three crop rotations and two tillage systems (no-tillage and conventional tillage) since 1987. The crop rotations consisted of barley→vetch, barley→sunflower, and a barley monoculture. The study took place in two growing seasons (1992–1994) to assess the effects of management practices on the weed seedbank. During this period, spring weed control was not carried out in winter crops. In the no-tillage system, there was a significant increase in the number of seeds of different weed species: anacyclus, common purslane, corn poppy, knotted hedge-parsley, mouse-ear cress, spring whitlowgrass, tumble pigweed, venus-comb, and Veronica triphyllos. Conversely, the presence of prostrate knotweed and wild radish was highest in plots under conventional tillage. These results suggest large differences in the weed seedbank as a consequence of different soil conditions among tillage systems, but also the necessity of spring weed control when a no-tillage system is used. With regard to crop rotations, the number of seeds of knotted hedge-parsley, mouse-ear cress, and spring whitlowgrass was greater in the plots under the barley→vetch rotation. Common lambsquarters dominated in the plots under the barley→sunflower rotation, whereas venus-comb was the most frequent weed in the barley monoculture. Larger and more diverse weed populations developed in the barley→vetch rotation rather than in the barley→sunflower rotation or the barley monoculture.