Prediction of plant response to fertilisers by soil tests

Abstract

Abstract Pasture yield responses to applied phosphorus were measured in 19 field experiments on recent and gley soils of Canterbury, and correlated with results from 8 methods of determining available phosphorus in soils. As the apatite content of these soils was high, methods which extracted this form of phosphorus from the soils gave very poor correlations with pasture responses. The best correlation was given by the Olsen method with either a 4– or 16–hr extraction period. This method accounted for 44% of the variation in response. Some of the soil–test methods were compared qualitatively with pasture responses from 94 trials on zonal soils. Overall, there was little difference in predictive ability between the Truog, Bray, and Olsen methods.