Experiments were conducted with `BigSur' broccoli (Brassica oleracea L. var. italica) at two sites in Fall 1993 and at two sites in Spring 1994 on a Hayter loam in southwestern Virginia. Our objectives were to determine the effects of tillage main plots (conventional tillage = CT and no tillage = NT) and weed control subplots [no overseeding or preemergent herbicide, oxyfluorofen, red clover (Trifolium pratense L.), `Dutch' white clover (Trifolium repens L.), and hairy vetch (Vicia villosa Roth)] on broccoli yield and weed suppression. In all sites, weed suppression and marketable broccoli yield with NT were equal to or higher than with CT. Overseeded legume living mulches did not affect broccoli yield in any site compared to the control plots and suppressed weeds as well as the oxyfluorofen in three of the four sites. Thus, the NT systems used in these experiments can suppress weeds and produce high broccoli yields. Also, overseeded legume living mulches can be established effectively after transplanting to suppress weeds without reducing broccoli yield. Chemical name used: 2-choro-1-(3-ethoxy-4-nitrophenoxy)-4-(trifluoromethyl) benzene (oxyfluorofen).