TRENDS IN MAIZE GRAIN YIELDS IN A LONG-TERM FERTILIZER TRIAL

Maize (Zea mays L.) (summer) was rotated with field pea (Pisum arvense Poir.) (winter) on soil of the Hutton form (loamy, mixed, thermic Rho dic Kandiudalf) over a 50-year period. Water (W), nitrogen (N), phosph orus (P), potassium (K) and farm yard manure (M) were each applied at two levels in a 2(5) factorial experiment. The lower levels of N, P, K and M were zero applications. Simple effects for N, P and K and the m ain effect for M were positively correlated with yield level. Over tim e, the simple effect of N exhibited an increased positive trend only w hen pea straw was removed rather than incorporated, while simple effec ts of P and K increased sharply during later years. The zero fertilize r treatment yielded in excess of 2 t maize grain ha(-1) during the las t decade of the trial. Manure provided consistently high yields withou t addition of inorganic fertilizer. This was ascribed mainly to an amp le nutrient supply, as soil organic matter levels were only marginally higher than in the balanced NPK treatment, which received inorganic f ertilizers. The greatest depletion of soil organic matter occurred in imbalanced treatments and the control. Threshold P and K levels in top soil for attaining yields of 4-6 t ha(-1) were ca. 5 mg kg(-1) P and 3 5-40 mg kg(-1) K, respectively, in P- and K-deficient treatments. Fert ilizer N, P and K were utilised more efficiently in fully balanced tha n in imbalanced fertilizer treatments. In treatments receiving abundan t P, about 60% of potential plant-available P could not be accounted f or. The legume crop apparently played a major role in the more efficie nt utilisation of nitrogen and phosphorus by crops in treatments where these elements were not applied.