Results for 1998–2004 are reported from four long-term (25–28 years) tillage trials, comparing conventional autumn ploughing with reduced tillage, normally spring harrowing only. Plant residues were retained during the period studied. The weather was somewhat wetter than the 1961–1990 normal. Results with reduced tillage were mostly similar to those seen in earlier trial periods. In Trial 1, mean grain yield was 95% with spring harrowing only versus autumn ploughing, 96% when harrowing in autumn was performed as well and 97% when the soil was ploughed every third year. In Trial 2 positive crop rotation effects were found both with and without ploughing, and reduced tillage gave 5% lower grain yield also in this trial. In Trial 3, tillage system did not affect yields of cereals grown in rotation with potatoes, but reduced tillage gave 12% lower potato yield than ploughing. Little difference in response to N fertilizer was found. In Trial 4, reduced tillage on large-scale (0.7 ha) plots gave 11% lower grain yields than annual ploughing, partly due to shallow sowing depth. No long-term trend in yield responses to tillage was discernible in any trial, and between-year variability was similar with both ploughing and reduced tillage. Percentage yields with reduced tillage relative to annual ploughing correlated positively with rainfall in May and with mean air temperature in August. It is concluded that the reduced tillage systems studied are sustainable in terms of productivity, relative to labour, machinery and energy inputs. Likely benefits of such systems include higher levels of organic matter and aggregate stability in surface soil horizons, but a disadvantage is the need for frequent herbicide use to control perennial weeds.