EFFECT OF REACTIVE PHOSPHATE ROCKS AND WATER-SOLUBLE PHOSPHORUS FERTILIZERS ON EXTRACTABLE PHOSPHORUS CONCENTRATIONS IN SOIL

Soil samples were collected each year from all plots at all sites in t he National Reactive Phosphate Rock Project and analysed for bicarbona te-soluble phosphorus (P) using the Colwell procedure (soil test P). T he relationship between soil test P and the level of P applied was ade quately described by a linear equation. The slope coefficient of this equation provides a measure of the extractability of P from soil treat ed with a particular fertiliser and has been termed the 'extractabilit y' for that particular fertiliser at that site. Extractability values were used to estimate phosphate rock effectiveness; this was done by d ividing the extractability for each phosphate rock by the extractabili ty of the reference water-soluble P fertiliser to provide a relative s oil extractability (RSE). There was a good agreement between the RSE f or a phosphate rock and its substitution value for water-soluble P fer tiliser, which is a measure of fertiliser effectiveness based on pastu re yield. Estimates of fertiliser effectiveness, based on increases in soil test P values, also agreed with estimates based on pasture yield with respect to the ranking of different P fertilisers, and the ranki ng of phosphate rocks in order of their reactivity. Simple and multipl e linear regression analyses were used to identify relationships betwe en a range of soil properties and extractability values across sites, but no relationships were found.