Crop productivity in relation to species of previous crops and management of previous pasture

An experiment at Rutherglen in north-eastern Victoria compared 5 grass-removal methods in subterranean clover-based pastures that were grown before cropping sequences of canola-wheat-lupin-wheat or wheat-wheat-lupin-wheat. The cropping sequences were started in 3 successive years to provide replication in time. Grass removal from the pasture was more effective in winter than in spring and led to yield increases by the first and second crops. The largest increase (80%) was by the first canola crop after winter-cleaned pasture. The yield increase by the equivalent wheat crop was 42%. Since annual grasses and canola do not host the same root pathogens, we conclude that the yield responses were not due to root-disease control but probably to increased N supply. Assays of wheat roots confirmed that root disease was negligible throughout the experiment. Wheat growing in the year after canola yielded 11% more than wheat growing after wheat. The most surprising result was a 17% increase in the yield of wheat growing 3 years after canola compared with wheat growing 3 years after wheat, with wheat-lupin sequences in the intervening years for both systems. We suggest that canola and lupin, both of which are non-hosts of arbuscular mycorrhizal fungi, reduced mycorrhizal root colonisation in the fourth-year wheat crop, leading to less drain on assimilates.