A long-term field experiment on acutely copper deficient soil commenced in 1967 in south-western Western Australia to measure the residual value of fertiliser copper for grain production of spring wheat (Triticum aestivum L.). Before sowing the first wheat crop, 6 amounts of copper [0, 0.69, 1.38, 2.07, 2.75 and 4.75 kg copper/ha as copper sulfate (25% copper)] were applied to large main plots when the fertiliser was placed (drilled) with the seed during sowing. Thereafter, wheat was grown every fourth year in rotation with subterranean clover (Trifolium subterraneum L.)-based pasture. When each wheat crop was sown in subsequent years, 4 amounts of copper were drilled with the sown seed in subplots within each of the 6 main plots, using new suplots for each wheat crop that were not previously treated with copper since the initial application in 1967. The effectiveness for producing wheat grain from the 6 original amounts of copper was calculated relative to freshly-applied copper by estimating the amount of original copper and freshly-applied copper required to produce the same grain yield. Results for the first 12 years have been published (Gartrell 1980); results for years 16–32 are presented in this paper.

For the first 12 years large wheat grain yield increases were only obtained to freshly-applied copper in the original nil-copper main plots. For the first time, in the 16th year, copper responses also occurred to freshly-applied copper in the original 0.69 kg copper/ha main plots. The recommended amount of copper applied to wheat, for the soil type used, was 1.38 kg copper/ha and this supplied sufficient copper for grain production for 28 years. At this stage, grain yield responses to freshly-applied copper occurred for the first time in these main plots. The copper removal in grain of each wheat crop only accounted for about 2–3% of the original 1.38 kg copper/kg treatment. Otherwise, all freshly-applied copper treatments were on the maximum yield plateau because wheat roots accessed enough copper for grain production from soil treated with the original copper treatments. In years 4–16, the original copper treatments were more effective than freshly-applied copper because cultivating soil to sow wheat mixed the original treatments through soil increasingly improving interception of the original copper treatments in soil by plant roots. However, in years 24–32, freshly-applied copper was more effective as the residual value of the original copper treatments continued to decrease due to continued slow reaction of the original copper treatments with soil.

For wheat at the flag leaf stage, the concentrations of copper in the youngest expanded wheat leaf blades that were related to 90% of the maximum grain yield (prognostic value) was about 1.4 mg/kg.