Animal production from naturalized and sown pastures at three stocking rates under coconuts in the Solomon Islands

Abstract

SUMMARY: A grazing study, comparing a naturalized and a sown pasture at three stocking rates, was conducted under a uniform stand of 65-year-old coconuts on a fertile soil in the Russell Islands. In a plot trial on this site there was no significant response to application of up to 400 kg N/ha/year over 2 years to Brachiaria decumbens. Average daily transmission of photosynthetically active radiation through the coconut canopy was 60% of full sunlight. The sown pasture consisted initially of para (Brachiaria mutica), signal (B. decumbens) and koronivia (B. humidicola) with the legumes Centrosema pubescens, Pueraria phaseoloides and Stylosanthes guianensis. The main species in the naturalized pasture were Axonopus compressus, Mimosa pudica, C. pubescens and Colopogonium mucunoides. Pastures were set stocked in two replicates over 3 years at 1·5, 2·5 and 3·5 animals/ha. There was no significant difference in live-weight gain between pastures in any year. Live-weight gain per head declined linearly with increasing stocking rate. Highest live-weight gain was 437 kg/ha/year in the 1st year at 3·5 animals/ha. The planted grasses declined from 60% at the start of grazing to 6% 8 months later. A. compressus increased from 2 to 24% over 3 years in the sown pasture, and from 12 to 34% of yield in the natural pasture. M. pudica increased from 27 to 44% in the sown, and remained approximately constant at 37% in the natural pasture. G. pubescens increased at 2·5 animals/ha, but was replaced by P. phaseoloides at 1·5 animals/ha and by A. compressus and M. pudica at 3·5 animals/ha. There were no significant effects of pasture treatments or stocking rate on copra yields. Where there is a cover of naturalized grasses and legumes under coconuts, cultivation and planting of exotic species cannot be recommended. Major improvement will come from thorough weed control and maintaining stocking rates between 1·5 and 2·5 animals/ha.