Diagnose da composição nutricional de tecido foliar e radicular do milho

Abstract

This work and preliminary reference standards for a diagnosis of the nutritional composition of the corn leaf and root tissue were aimed at this work. The experiment developed in 2016, in the experimental area located in the Experimental Farm of Agricultural Sciences of the Federal University of Grande Dourados, in the municipality of Dourados, MS, with a corn crop, in the second harvest of the agricultural year of 2015/2016. Leaf and root samples were collected at the R1 development stage, and the reference reference for the first row and bet of spike and as roots were collected from five plants per plot and washed to remove soil adhered thereto. As subpopulations were separated by crop yield levels and Mahalanobis distance from the multinutrient variables. They were calculated as norms of the Composition of Nutrients Diagnosis (CND - Composition Nutrient Diagnosis), composed of the mean and the coefficient of variation of the multinutrient variables in the reference populations. The method used to separate subpopulations was from Cate-Nelson, dividing it into four subpopulations classified as VN (True Negative), VP (True Positive), FN (False Negative) and FP (False Positive). With a principal components analysis it was possible to identify four main components (CP) that presented the eigenvalues 3,385; 2,4; 1,513 and 1,380 for each plant tissue with accumulated variances of 37.6; 64.3; 81.1 and 96.4%, respectively. The root tissue presented better performance by the methods of Cate- Nelson, diagnostic test and discriminant analysis for selection of populations of high productivity.