Influência do tamanho de sementes e de características agronômicas em descritores adicionais de soja

• Ano de defesa: 2013
• Autor(a) principal: Silva, Francisco Charles dos Santos
• Orientador(a): Sediyama, Tuneo
• Banca de defesa: Sediyama, Gilberto Chohaku
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Fitotecnia
• Departamento: Plantas daninhas, Alelopatia, Herbicidas e Resíduos; Fisiologia de culturas; Manejo pós-colheita de
• País: BR
• Palavras-chave em Português: Glycine max (L.); Estabilidade; Características qualitativas
• Palavras-chave em Inglês: Glycine max (L.); Stability; Qualitative characteristics
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::FITOTECNIA
• Link de acesso: http://locus.ufv.br/handle/123456789/4606

The insufficient amount of descriptors revels the importance of this theme to the study related to the soybean breeding. This work was divided in two phases, which the first phase have the objective to verify the influence of seed size and identify new morphological characteristics to discrimination of soybean cultivars. The second phase have the objective to estimate the repeatability coefficient of hypocotyl length, high plant, of length epicotyl, length of the first internode, petiole of the first trifoliade leaf, petiole of the unifoliade leaf, raquis of the first trifoliade leaf, of the angle formed by the petiole insertion of unifoliade leaf, of pod length, of diameter pod, of thickness of pod, of thickness of waist of pod curvature and the hilum length. The first study was conducted in greenhouse of the Department of Plant Science, at Federal University of Viçosa. Two experiments were realized, one in the summer and other in the in winter. The treatments were constituted by 10 soybean cultivars with 3 different sizes of seeds and a sample of the original lot. The experimental design used was a randomized block with four replications in a factorial design. Each treatment was constituted by a pot with two plants, being evaluate 5 and 6 qualitative and quantitative characteristics respectively, and in the end was realized the analyses of discriminaton by the Gower algorithm. The second study was also conducted at Federal University of Viçosa, being realized 5 experiments with 124, 93, 90, 16 and 16 genotypes of soybean for the first, second, third, fourth, and fifth experiment, respectively, which were conducted in a completely randomized design, and from the estimative of main coefficients and the structural analyzes were calculated the number of necessary evaluations. The results of the first study indicate be possible differentiate the cultivars by the characteristics in the two times of sowing, except for the opening of petioles of unifoliade leaf, seeds with lower size originates lower plants in the V3 stage and with epicotyl shorter, and that the use of Gower algorithm showed be efficient in the evaluation of discriminatory capacity of characteristics quantitative and qualitative in a simultaneous way, demonstrating that these are useful like discriminators of soybean. With the second study, was conclude that the hilum length, the length of first internode and the pod curvature requires lower quantity of measurements comparing to the most characteristics for the same confiability level; and with six measurements, was obtained 95% of confiability for the height of plant in V3 stage by the ANOVA, CP(correl), CP(cov) and AE(correl) methods; 90% for the length of the first internode, hilum length, length, diameter and curvature of pod by the ANOVA, CP(correl), CP(cov) and AE(correl), with seven measurements; was obtained 90% of efficiency for length epicotyl for all the used methods; 85% for the length of petiole of the first trifoliade leaf by the methods of CP(cov) ad 90% by the methods of ANOVA, CP(correl) and AE(correl); was obtained yet 90% of confiability by the methods ANOVA, CP(correl), CP(cov) and AE(correl) for thickness of pod and hilum length. With 15 measurements realized, is possible obtain 90% of confiability for the length of petiole of the first trifoliade leaf and angle formed by the petiole insertion of unifoliade leaf for all the methods used; and for the length of raquis of the first trifoliade leaf were needed 37 measurements for the confibilaity of 90% by the ANOVA, CP(correl), CP(cov) and AE(correl) methods.