Effects of soil composition on the relationships between soil test values for phosphorus fertilizer requirements

Abstract

The correlations between the P test values obtained by a series of standard procedures were found to be improved by allowing for the effects of soil composition, particularly with respect to reactive iron, aluminium, magnesium, and calcium. The high correlations thus obtained suggest that most of the P test procedures will be of about equal value for estimating P fertilizer requirements if allowance is made for the effects of variations in soil composition. They also suggest that P test calibration equations should include variables representing soil Fe, Al, Mg, or Ca, as well as the actual P test.