Relações lineares entre caracteres fenológicos, morfológicos e produtivos em milho

• Ano de defesa: 2013
• Autor(a) principal: Casarotto, Gabriele
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; BR; Agronomia; UFSM; Programa de Pós-Graduação em Agronomia
• Palavras-chave em Português: Correlação linear; Multicolinearidade; Análise de trilha; Zea mays L.; Linear correlation; Multicollinearity; Path analysis; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA
• Link de acesso: http://repositorio.ufsm.br/handle/1/5084

This study aimed to verify the existence of linear relationships among phenological, morphological and productive characters of maize cultivars (Zea mays L.) of early and veryearly cycle and transgenic class and also to identify which characters have high correlation and direct effects on grain productivity. Six experiments were performed with early and veryearly and transgenic maize cultivars in the growing seasons 2009-2010 and 2010-2011, in the experimental area of the Department of Plant Science, of Federal University of Santa Maria. In the 2009-2010 harvest were evaluated 36 early cultivars, 22 veryearly and 18 transgenic and 2010-2011 harvest, 23 early, 9 veryearly and 27 transgenic. The experimental design was a randomized block design with three replications. The experimental unit consisted of two rows of five meters in length, spaced at 0,80 m. The seeding rate was adjusted to 62,500 plants ha-1. In each experimental unit it were tagged, randomized, three plants, and it were evaluated 15 characters of each one. The average of these three plants was the value of repetition. It were evaluated phenological (total number of leaves per plant (NFO), phyllochron estimated with the number of expanded leaves(FNFE), phyllochron estimated with the total number of leaves (FNTF) in ° C day leaf-1, the number of days of seeding until male flowering (FM) and number of days of seeding until female flowering (FF)), morphological (plant height (PH) and ear insertion height (AE), in cm) and productive (ear weight (PE), in g, number of kernel rows per ear (NFI), ear length (CE), in cm, ear diameter (DE), in mm, cob weight (PS), in g, cob diameter (DS), in mm, hundred kernel weight (MCG), in g, and grain productivity (PRO) in g ear-1) characters. Analysis of variance (ANOVA) was performed and the means of the cultivars were compared by Scott-Knott test at 5% probability. The linear correlation coefficients of Pearson among 15 evaluated characters were estimated for each experiment. For the path analysis, the PRO was considered the main character and the other characters were considered explanatory ones. It was accomplished multicollinearity diagnosis in the correlation matrix among the explanatory characters and the characters causing high degree of multicollinearity were eliminated. The direct and indirect effects on the PRO were estimated using path analysis and the verification of characters that influence PRO and their contribution in predicting the PRO were estimated by stepwise regression analysis. There are linear relationships among the phenological, morphological and productive characters maize plants. The characters PE and DE showed linear correlation coefficients of Pearson very strong (r≥0,97) and moderate to strong (0,55≤r≤0,78), respectively, with the PRO. In general, the character DE has high correlation and positive direct effects (0,6686 ≤ direct effect ≤ 1,1818) on the PRO. Allied to DE, the CE has a high positive contribution in predicting the PRO. Therefore, they can be used for indirect selection in maize breeding programs.