Capacidade produtiva e progresso genético de pinhão-manso (Jatropha curcas L.)

Abstract

The jatropha (Jatropha curcas L., Euphorbiaceae), is a perennial cycle plant whose attentions in the last six years has directed its capacity for vegetable oil. Given their biological attributes, agronomic, economic and industrial, the species is one of oleaginous on the agenda of government interest, which through the National Program for Production and Use of Biodiesel has intensified the search for new bioenergy sources. However, this is still oleaginous phase of domestication and quest for technological mastery. Meet the productive potential, the relationship between characters, the genetic variability and infer genetic parameters have directed in recent years, most quantitative genetic studies and breeding in Brazil. Therefore, the aim of the present study was to estimate genetic parameters and predict random variables important to the strategies of selection and breeding; meet the minimum harvest / measurements needed to choose jatropha trees; estimate genotypic and phenotypic correlations; evaluate the progress of genetic grain production and to study the genetic variability. Was evaluated for 24 months at the experimental farm of Embrapa, in Porto Velho - RO, an experiment with three full blocks, four provenances (two of Rondônia, one of Minas Gerais and São Paulo) with 16 trees per plot, spaced 2x3 meters. The traits associated with grain yield and plant architecture. The great genetic evaluation procedure REML/BLUP was used for the prediction of random variables, genetic parameter estimation and inference about repeatability. Inferences about variance components and estimates of genetic parameters for some agronomic traits during the first two years of development of Jatropha curcas, showed genetic variability exists. The numbers of fruits per bunch and maturation index showed little characters are informative and useful in quantitative genetic studies. The stabilization genotypic characters for number of clusters at harvest of half a year, number of branches, height, crown projection in greater spacing and canopy volume was obtained with equal or less than three measurements per year. Already for grain yield and number of clusters at harvest year end, required a greater number of measurements (mi > 4). High correlations (r.) were obtained genotypic and phenotypic characters among themselves for grain yield, its components and those characters associated with plant architecture (r.  0.75). Predicted genetic gains for grain production of the first two years of planting were respectively higher than 20%, considering the selection of the ten best individuals. When considering the two growing seasons for variable grain production, most plants exhibited average performance (adaptability) generally indicated as the centroid method. Considering the information of the individual additive effects, genetic diversity was not necessarily linked to geographical origin, as plants of different provenances formed groups. At the level of individuals, formed several genetically divergent groups.