Nutritional status diagnosis of cotton and soybean crops through critical level methods is based on sufficiency ranges established many years ago, with little advances since then. However, as nutrient contents in leaves are influenced by sample type and crop yield potential among other factors, it is important to establish regional reference values. The aim of this study was to establish optimum nutrient concentrations in cotton and soybean leaves by adjusting a regression model to nutrient concentration based on DRIS indexes. Database consisted of yield and nutrient content in leaf samples collected in commercial cotton and soybean crops and soybean experimental plots in Mato Grosso do Sul and Mato Grosso States. On each selected place, chemical composition of leaves and crop production (whole cottonseeds or soybean grains) were evaluated at full blooming stage and physiological maturity, respectively. Based on the proposed methodology, sufficiency ranges for indexes leaves of cotton (fifth fully formed leaf) and soybean (third trifoliate leaf, without or with petiole) were determined, whith variation smaller than that presented in other papers reviewed. Specifically for the soybean crop, in addition to confirming the differences in the reference values according to the type of index leaf sampled. Samples without petiole produce significantly higher levels of N, P, B, Fe, Mn and Zn and lower K level, compared with samples with petiole. If disregarded, the manner of sampling can induce false diagnosis of nutritional deficiency or excesss in the crop.
foliar analysis; diagnosis and recommendation integrated system; index leaf; Glycine max; Gossypium hirsutum
