ABSTRACT
Summary Soil-nutrient deficiencies in two cerrados soils have been investigated as a preliminary to small-plot research oriented to estimate fertilizer requirements for establishment of legumes in grass-legume associations. The studies were conducted in the greenhouse, in 2-kg pots, with surface-soil samples taken from two sites representative of the dominant soils in the region. These 321are the dark red latosol (DRL) and the red-yellow latosol (RYL), which qualify as Haplustox and Acrustox, respectively. Two kinds of experiments were carried out. One consisted of a 1/4 replicate of a 28 factorial including potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), zinc (Zn), molybdenum (Mo), copper (Gu), manganese (Mn), and boron (B), in which Cu and Mn were applied together as a single factor. Levels of these nutrients were 40, 100, 10, 15, 1, 0.13, 1, 1, and 0.5 parts/million parts of soil (ppm), respectively. All pots received monocalcium phosphate [Ca (H2PO4)2·2H2O] at a rate of 50 P ppm. A second experiment consisted of four levels of P (25, 50, 100, and 200 ppm) and four levels of Ca (0, 20, 100, and 200 ppm) as CaCO3, with basal application of all other nutrients and rates used in the previous experiment. Test plants used were calopo (Calopogonium mucunoides Desv.), a Centrosema (Benth.) species CIAT 438, stylo (Stylosanthes guianensis Aubl.) CIAT 2243, and gambagrass (Andropogon gayanus var. bisquamulatus [Hochst.] Hack) CIAT 621. Results showed a clear response to P up to 100 P ppm in both soils, when gambagrass or stylo was used, and up to 200 P ppm when the Centrosema sp. or calopo was used. Responses to the highest levels were more evident in the RYL soil. In all species, Ca improved plant responses to low levels of P. Maximum yields were also increased except in the case of stylo. In the second experiment, it was observed that S was deficient in both soils for all test plants, as measured by plant responses to applied S. Potassium responses were also observed in both soils, while Ca response was more common in the RYL soil. Magnesium response was observed only in the RYL soil when gambagrass was used as a test plant. Molybdenum and Zn deficiencies were observed in both soils with some legumes. The results demonstrate the importance of S, Ca, Mg, K, Mo, and Zn deficiencies after P deficiency has been corrected. On the basis of these observations, field experiments have been established to define the required amounts of S and K and to confirm under field conditions the need for Ca, Mg, Mo, and Zn.
