Abstract: The objective of this study was to compare the selection of plants bred by different pedigree methods using selection among, among and within and only within families. The haploid induction rate of 14 S0:1 and seven S2:3 families, all crossed with the single-cross hybrid GNZ9501, was evaluated. An experimental area of the Department of Biology of the Federal University of Lavras (UFLA), in Lavras, Minas Gerais, in the growing seasons 2012/2013 and 2014/2015, was used for the experiments. In each growing season, one experiment per was carried out, arranged in a complete randomized design, with one and two replications, respectively. Haploid induction was most effective in the families 2 and 6 in both growing seasons. Selection among and within families resulted in higher genetic gains for haploid induction. The results indicated a high genetic variability for haploid induction rate in plants within families.

ABSTRACT. The breeding program of Urochloa ruziziensis evaluates many genotypes in initial phases. Evaluations through grades might make the selection less costly. The aim of this study was to verify the efficiency of visual selection for green biomass yield in relation to different selection strategies, such as mass selection by phenotypic mean, BLUP (Best Linear Unbiased Prediction) and at random. For this purpose, 2,309 regular genotypes were evaluated in an augmented block design in two cuts. The evaluators gave grades for plant vigor, and later, the plots were measured for green biomass yield. The coincidences of the selected genotypes were estimated by different selection strategies. Then, 254 clones of the genotypes selected in different strategies were evaluated in a clonal test in a triple lattice design in four cuts. The statistical analyses were performed in SAS using the Mixed procedure. The regular genotype level and clone-mean basis heritabilities were 31.16 and 62.91%, respectively, for green mass yield. The expected selection gains were 21.09% (visual), 25.43% (phenotypic mean), and 27.5% (BLUP). Moreover, the realized heritabilities for these strategies were 15.58, 11.87, and 15.86%, respectively, which might be associated with genotype by environment interaction. Therefore, the visual selection could be a useful strategy in initial phases of a U. ruziziensis breeding program because the efficiency was moderate to high in relation to phenotypic mean and BLUP.

ABSTRACT. In most cases, in genetic breeding of Arabica coffee, plot size is defined in an empirical manner. It is often based only on the experience of the breeders and the availability of resources, potentially leading to a reduction in precision. Therefore, the aim of this study was to estimate the size of the experimental plot for evaluation of coffee yield. We evaluated two experiments for validation of cultivars with 12 treatments set up in a randomized complete block design with three replicates and plots composed of 50 plants. Each plant was considered as a basic unit. Estimates of ideal plot size were made by maximum curvature of the coefficient of variation, linear-plateau segmented model and by the resampling methods. We discussed the variations in the parameter estimates for different plot sizes. Divergence was seen among the plot sizes estimated by the different methodologies. Increasing the number of plants per plot led to a higher experimental precision to the point that the increase was no longer significant. The plot size recommended for evaluating coffee production is from seven to 19 plants.

O objetivo deste trabalho foi avaliar a necessidade de bordadura e estimar o tamanho ótimo de parcela em experimentos para avaliação de Urochloa ruziziensis. Foram avaliadas a altura das plantas e a produção de massa verde de oito progênies de meio-irmãos de U. ruziziensis, em dois cortes. Utilizou-se o delineamento de blocos ao acaso, com quatro repetições e parcelas de 16 m². Cada parcela foi subdividida em 32 estratos de 0,5 m², sendo cada estrato considerado como uma unidade básica. Para determinar a necessidade de bordadura, foi realizada análise de variância, tendo-se considerado a posição da unidade básica na parcela. As estimativas do tamanho ótimo da parcela foram realizadas pelo método da máxima curvatura do coeficiente de variação e pelo método da reamostragem. O uso de bordaduras não altera o desempenho médio e a classificação das progênies de meio-irmãos avaliadas. O emprego de parcelas com 3 m² é suficiente para obter boa precisão experimental em experimento com progênies de meio-irmãos de U. ruziziensis.

The objective of this work was to assess the need for a border area and to estimate the optimum plot size in experiments for the evaluation of Urochloa ruziziensis. Plant height and green mass production of eight half-sib progenies of U. ruziziensis were evaluated in two cuttings. A randomized complete block design was used, with four replicates and 16 m² plots. Each plot was subdivided into 32 strata of 0.5 m², with each stratum being considered as a basic unit. To determine the need for a border area, analysis of variance was carried out considering the position of the basic unit in the plot. Estimates of optimum plot size were made by the maximum curvature of the coefficient of variation method and by the resampling method. The use of borders does not affect the mean performance and classification of the half-sib progenies evaluated. The use of 3 m² plots is enough to obtain good experimental accuracy in an experiment with half-sib progenies of U. ruziziensis.