Agricultural land in China is being degraded, with soil erosion becoming an increasing problem. In Yunnan Province, south-west China, there is a long history of soil erosion due to soil type, climate, anthropogenic influence and because 95% of the Province is mountainous. Population pressure and lack of flat land necessitate cultivation of steep slopes. The Yunnan Government prohibits cultivation of slopes >25°, however policy enforcement would result in food shortages in the Province, due to a lack of suitable land <25°. Therefore, the most appropriate way to curb soil erosion in Yunnan is to devise affordable agronomic means of reducing soil loss, which do not decrease crop productivity on sloping land currently under cultivation. At present, very little research has addressed these issues. A research project, building on existing work from 1993-1996, was initiated in 1998. The aim was to test the hypothesis that contour cultivation and contour cultivation plus straw mulch decrease runoff and soil erosion rates on sloping land in Yunnan Province under maize cultivation and to assess the impact of these conservation measures on maize productivity and soil nutrient status. Thirty runoff plots, located on three different slope angles (I 3°, II 10° and III 27°), in three groups of 10, were used to examine three cropping treatments in a replicated plot design in 1998 and 1999. Treatments were downslope cultivation (control), contour cultivation and contour cultivation plus straw mulch. In addition, there was an unreplicated bare plot in each group. Runoff and soil loss were measured on a storm-by-storm basis. Soil nutrient status was measured at the beginning and end of each cropping season. Crop growth parameters and soil physical properties were measured throughout the cropping seasons (21/05–7/10 in 1998 and 22/05–2/10 in 1999). In 1998, seasonal rainfall was 1024 mm, ~28% greater than the 30-year mean. Soil loss was significantly reduced by contour cultivation on Slopes I and II. On Slope I, downslope cultivation produced 3.07 t ha-1 soil loss and contour cultivation reduced this by 81.4 %. On Slope II, downslope cultivation produced 19.11 t ha-1 and contour cultivation reduced this by 58.0%. The addition of straw mulch gave a further, nonsignificant, reduction on both slopes. On Slope III, downslope cultivation and contour cultivation produced 6.92 and 6.29 t ha-1 of soil loss, respectively, with contour cultivation plus straw mulch having 99.4% less erosion than downslope cultivation. In the much drier 1999 season, no treatment significantly reduced soil loss on Slope I. Contour cultivation significantly reduced soil loss on Slopes II and III. On Slope II, downslope and contour cultivation produced 11.52 t ha-1 of soil loss and contour cultivation reduced this by 85.8%. On Slope III, downslope and contour cultivation produced 8.62 and 0.23 t ha-1, respectively; a reduction of 97.3% by contour cultivation. The addition of straw mulch did not further decrease soil loss. Treatment effects on soil nutrient status varied between the two years. At the end of the 1998 season, there was significantly higher soil available N under contour cultivation plus straw mulch on all three slopes (Slopes I and II P <0.001, Slope III P <0.05), an effect that was not found in 1999. At the end of the 1999 season, soil available K was significantly (P <0.001) higher under contour cultivation plus straw mulch on Slope III. In both years, contour cultivation plus straw mulch significantly reduced soil temperature. However, this did not result in yield reductions in comparison with the control. There was an increase in soil moisture content under contour cultivation plus straw mulch during dry periods, which was particularly noticeable in 1999. In 1998, there were no significant treatment effects on grain or shoot yield. In 1999, on Slope II, contour cultivation plus straw mulch significantly increased grain yield by 50.3% compared with the downslope treatment (P <0.05). In 1999, contour cultivation plus straw mulch also significantly increased leaf plus stem yield on Slopes I and II by 12.4 and 36.8%, respectively. It is concluded that on ≤10° slopes, contour cultivation alone is a suitable soil conservation measure. However, use of straw mulch would benefit soil moisture and nutrient status and could, therefore, increase crop yield. On ≥27° slopes, it is recommended that contour cultivation plus straw mulch be used as a soil conservation measure to ensure maximum soil conservation, even in extreme rainfall conditions.