Eficiência da seleção em fases iniciais no melhoramento da cana-de-açúcar

• Autor(a) principal: Pedrozo, Cássia ângela
• Data de Publicação: 2006
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/4738
• topic: Cana-de-açúcar; Seleção; Melhoramento genético; Modelos mistos; Correlações; Sugar-cane; Selection; Breeding; Correlations; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::FITOTECNIA::MELHORAMENTO VEGETAL

The objectives of the present study were: a) evaluate the effect of the selection for stem number in phase T1 (preliminary phase of selection); b) estimate coefficients of genotypic correlation within phase T2 (second phase of selection) and between phases T1 and T2; c) estimate genetic parameters and predict genotypic values for clone selection. A total of 5298 genotypes obtained from the crossing between cultivars RB72454 and RB83102, which were carried out in 1998, in Murici AL, constituted selection phase T1. In the year 2000, of this total 130 genotypes were sampled in relation to the number of stems present in the clump, comprising 5 treatments that defined the experiment of phase T2, as follows: Treatment 1 - genotypes with 2, 3 or 4 stems per clump; Treatment 2 - 5, 6 or 7 stems; Treatment 3 - 8, 9 or 10 stems; Treatment 4 - 11, 12 or 13 stems and Treatment 5 - 14 or more stems. The experiments referring to the two selection phases were carried out at the Sugar-cane Improvement Center - CECA, of the Federal University of Viçosa. The Tukey s test (α = 0.05) was used to compare treatment means, using a randomized block design with 5 repetitions. The results indicated that genotypes not presenting at least five stems per clump in phase T1 should be discarded, even if they show other favorable characteristics. Besides recording stem number, Brix, height and stem diameter were also evaluated in phase T1, for the 130 genotypes. In phase T2 the following traits were evaluated: Brix, height, diameter and stem number, mean stem weight, estimated stem production, tons of stems per hectare and tons of Brix per hectare. The genetic parameter estimates and the prediction of genotypic values were obtained by the Selegen - REM/BLUP software (Resende, 2002a; Resende, 2002b). Genotypic values were used to obtain the coefficients of genotypic correlation between the traits evaluated in the phases T1 and T2, correlation coefficients between the traits evaluated in phase T2, rates of genetic progress, minimal selection proportions and percentage of coincidence of genotypes selected in phase T1 that were shown superior in phase T2. The results allowed the conclusion that, among the traits evaluated in phase T1, stem height had more potential for being considered in the selection of genotypes to be included in phase T2, since it gave the highest estimates of genetic parameters (heritability, selection accuracy and CVg/CVe ratio), highest coefficients of genotypic correlation between phases T1 and T2 and highest rates of genetic progress. In phase T2, the estimated stem production gave high genotypic correlation with stem productivity, constituting therefore an alternative to indirect selection for this second trait. Among the clones evaluated in phase T2, RB987620, RB987572 and RB987561 had the highest predicted genotypic values for trait TBH.