A field experiment was conducted over four seasons on a clay loam (24% clay in topsoil) in Scotland. Straw incorporation with conventional ploughing, shallow ploughing, rotary and tine cultivation was investigated for continuous winter barley. Residue treatments were either chopped straw plus stubble or stubble only. Incorporation depths ranged from 0–100 to 0–300 mm. Where stubble only was incorporated, mean grain yields were greater with ploughing (7·2 t/ha) than with shallow rotary cultivation (6·3 t/ha). Where straw and stubble were incorporated, mean grain yields were greater with ploughing (7·6 t/ha) than with non-ploughing (mean over six treatments, 7.0 t/ha). Non-ploughing incorporation with deep (200 mm) rotary cultivators gave a greater mean grain yield (7·5 t/ha) than with shallow ( Irrespective of residue treatment, no improvement in yield or soil fertility resulted from ploughing deeper than 100–150 mm. Ploughing distributed residues more uniformly with depth than non-ploughing, which concentrated them near the surface. Residue incorporated near the soil surface decomposed at the same rate as that buried more deeply by conventional ploughing. The most reliable method of minimizing primary tillage for incorporation and for preserving favourable soil conditions was shallow ploughing. Minimizing secondary cultivations and substituting broadcasting for drilling reduced compaction and reduced inputs at minimum risk to the crop. Broadcasting seed also resulted in a small yield advantage over drilling.