Capacidade de combinação para seleção de genótipos de milho

• Ano de defesa: 2015
• Autor(a) principal: Giorgenon, Carlos Henrique Braz [UNESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Milho; Plantas - Melhoramento genetico; Genetica vegetal; Seleção de plantas - Melhoramento genetico; Plant-breeding
• Link de acesso: http://hdl.handle.net/11449/131928; http://www.athena.biblioteca.unesp.br/exlibris/bd/cathedra/09-12-2015/000855250.pdf

The heterosis discovery enabled the development of highly productive corn hybrids. However, some questions about the efficiency of testers for selection of superior lines are found. This work aims to study the combining ability of testers from different genetic structures with a population of inbred lines. The trial was conducted in 2012/2013 in the experimental area of Fazenda de Ensino, Pesquisa e Extensão (FEPE) of Faculdade de Ciências Agrárias e Veterinárias da UNESP, Jaboticabal - SP. The experiment consisted of six different testers that were crossed in topcross scheme with a population of 39 maize genotypes. These topcrosses were evaluated in experimental design of randomized blocks with two replications for each treatment. The agronomic trait evaluated was grain yield obtained by the average grain weight of the portions of each topcross. Data were subjected to variance analysis. With the averages for the different testers, the combining abilities of the 39 genotypes were estimated. Then the data were subjected to correlation analysis to check the level of similarity between the testers. Data were also subjected to analysis by REML/BLUP method and compared with data from the combining abilities. It was concluded that the combination ability information and REML/BLUP can be used for selecting maize genotypes. The association between the information of the combining ability and BLUP varied depending on the tester used, but, in general, showed high values. The combining ability and REML/BLUP can be used alternatively to compare a tester, but, they will be complementary when the objective is to compare more than one tester. Significant correlations were found between the testers, but the estimates are low magnitude, thus, this is not sufficient to exclude the combining ability, even with related testers. The testers classified differently the genotypes and could be necessary to use more than one tester for greater efficiency ...