Responses to fertiliser N, P, K, and S by maize on three soils of Rarotonga, Cook Islands

In a recent Soil Survey of the Cook Islands the fertility of some major soil types was assessed from chemical analysis and glasshouse pot trials using the grass green panic (Panicum maximum var. trichoglume). The glasshouse study has been extended to field trials with maize on three major agricultural soils in Rarotonga: Matavera clay loam, Pouara clay loam, and Tikioki clay loam. A 'subtractive' design was used to assess the response to fertiliser N, P, K, and S, and the maize was harvested at the tasselling stage and again at grain maturity. Relationships between maize yields from the subtractive treatments, nutrient levels in soils before planting, and nutrient levels in leaves at silking were examined. Nitrogen was shown to be the nutrient most limiting to the growth of maize; yield responses to N were obtained on all three soils. Responses to P were obtained on the Pouara and Tikioki soils, but not on the Matavera soil, reflecting its high level of available phosphorus. The response patterns observed in the field experiments were consistent with those observed in a glasshouse study using the same soil types. Added N significantly increased leaf levels of N, P, K, S, Mn, and Zn and levels of crude protein in grain. The trials also showed that grain yields of up to 11 000 kg/ha may be obtained in a growing period of 4 1/2 months, which compares favourably with yields for this maize variety grown in New Zealand in a 6 month growing period.