In any single season, wheat yields were unaffected by rotation or N level. However, 2-year biomass production was significantly greater (32%, on average) in the continuously cropped plots than in the wheat-fallow rotation. On average, <10% of the N in the wheat crop came from fertiliser in the season of application, and <1·2 kg N/ha of the residual fertiliser was recovered by a subsequent wheat crop. Chickpea fixed 16-48 kg N/ha, depending on the season, but a negative soil N budget was still likely because the amount of N removed in the grain was usually greater than the amount of atmospheric N₂ fixed. Uptake of soil N was similar in the cereal phase of all 3 rotations (38 kg N/ha, on average), but over the whole rotation at least 33% more soil N was removed from continuously cropped plots than from the wheat-fallow rotation, suggesting that the latter is a more sustainable system.

A laboratory study showed that although wheat and chickpea residues enhanced the gross rate of N mineralisation by c. 50%, net rates of N mineralisation were usually negative. Given the high C/N ratio of the residue, immobilisation, rather than loss processes, is the likely cause of the decline in the mineral N content of the soil. Consequently, decomposition of crop residues in the field may in the short term reduce rather than increase the availability of N for crop growth.