Evaluation of methods of chemical analysis for available phosphorus in soils of the "Zona Litoral-Mata" of Pernambuco

Abstract

A greenhouse study was conducted with corn for the purpose of comparing the abilities of five chemical tests to estimate supplies of plant available phosphate in soils of the physiographic region of Pernambuco known as the "Zona Litoral-Mata". Each soil in the study is typical of one of eleven mapping units used in the reconaissance soil survey of the region. The eleven soils represent a wide range of physico-chemical properties. The phosphorus tests examined included the following: 0.05 N HCl + 0.025 N H₂SO₄ (Mehlich method); 0.03 N NH₄F in 0.025 N HCl (Bray method n.° 1); 0.03 NH₄F in 0.1 N HCl (Bray method n.° 2); 0.5 M NaHCO₃ (Olsen method); and 0.5 N HCl (Hawaii method). The five tests are compared on the basis of simple correlation between the amounts of phosphorus extracted from the eleven soils and relative dry matter production as measured in the greenhouse. Relative dry matter production was computed as 100 times the ratio of weight dry matter produced when no P was applied to the weight of dry matter when P was applied. Correlations of the results of five phosphate tests with relative dry matter production were as follows: Mehlich, r = 0.520; Bray n.°  1, r = 0.540; Bray n.° 2, r = 0.398; Olsen, r = 0.170; and Hawaii, r = 0.649. Only the later correlation was significant at the 95% probability level, therefore indicating that there is some advantage to using the Hawaii test to measure available phosphorus in the soils studied. Phosphorus appeared to be strongly deficient in all eleven soils. Relative dry matter production ranged from 22.8 to 71.2. In all instances the response of corn to applied phosphate was significant at the 99% probability level.