Atrazine (6-chloro-N2-ethyl-N4-isopropyl-1,3,5-triazine-2,4-diamine) and metolachlor [2-chloro-N-(2-ethyl-6-methylphenyl)-N-(2-methoxy-1-methylethyl)acetamide] have been found with increasing occurrence in rivers and streams. Their continued use will require changes in agricultural practices. We compared water quality from four crop-tillage treatments: (i) conventional moldboard plow (MB), (ii) MB with ryegrass (Lolium multiflorum Lam.) intercrop (IC), (iii) soil saver (SS), and (iv) SS + IC; and two drainage control treatments, drained (D) and controlled drainage-subirrigation (CDS). Atrazine (1.1 kg a.i. ha-1), metribuzin [4-amino-6-(1,1-dimethylethyl)-3-(methylthio)-1,2,4-triazine-5(4H)-one] (0.5 kg a.i. ha-1), and metolachlor (1.68 kg a.i. ha-1) were applied preemergence in a band over seeded corn (Zea mays L.) rows. Herbicide concentration and losses were monitored from 1992 to spring 1995. Annual herbicide losses ranged from < 0.3 to 2.7% of application. Crop-tillage treatment influenced herbicide loss in 1992 but not in 1993 or 1994, whereas CDS affected partitioning of losses in most years. In 1992, SS + IC reduced herbicide loss in tile drains and surface runoff by 46 to 49% compared with MB. The intercrop reduced surface runoff, which reduced herbicide transport. Controlled drainage-subirrigation increased herbicide loss in surface runoff but decreased loss through tile drainage so that total herbicide loss did not differ between drainage treatments. Desethyl atrazine [6-chloro-N-(1-methylethyl)-1,3,5-triazine-2,4-diamine] comprised 7 to 39% of the total triazine loss.