Estimativas de repetibilidade e predições de ganhos genéticos em clones de cajueiro-anão.

Abstract

The cashew clones (Anacardium occidentale L.) of the dwarf type have allowed a great boost to cashews in Northeast Brazil, due to the higher productivity, precocity and uniformity of both the chestnut and peduncle, besides being smaller in relation to the common type, facilitating manual harvesting and fruit retention. Considering the importance of cashew, the main challenge of genetic improvement corresponds to the generation of clones that maximize the production of almonds, as well as the use of the peduncle, to meet the needs of industry and consumers. Because cashew is a perennial plant, one of the main limitations for the availability of new cultivars is the time it takes to develop them. Thus, the repeatability coefficient is an essential parameter for the breeder, since it allows determining how many phenotypic observations must be made in order to select genotypes accurately, optimizing time and resources. Since perennial species require more long-term evaluations and information loss is common, more efficient selection methodologies, such as the REML / BLUP procedure, have been sought in order to increase selective accuracy and obtain greater genetic gains. Thus, the objective of this work was to verify the existence of genetic variability in a dwarf cashew population for some agronomic characters; ii) to infer about the type of interaction between clones of dwarf cashew and vines; iii) to estimate the repeatability coefficients and the selection gains for the most relevant characters of the dwarf cashew; and iv) to identify clones with genetic potential to continue the process of obtaining new cultivars. The experiment was conducted at the Experimental Field of Pacajus, belonging to Embrapa Agroindústria Tropical. The experimental design was a randomized block design with 35 treatments (33 dwarf cashew clones + 2 commercial controls), eight replications and one plant per plot. The spacing of 8 meters between rows and 6 meters between plants was adopted. The analyzed traits were plant height (AP), crown diameter (DC), chestnut productivity (PROD), mean brown weight (PMC), nut percentage (% CF) and disease occurrence, as an anthracnose reaction (ANT), black mold (MP), presence of powdery mildew on inflorescence (OI) and presence of powdery mildew on chestnut (OC). The REML / BLUP was used to predict the genetic values and to estimate the genetic parameters. The clones presented genetic variability for most of the evaluated characters; there was no interaction between clones and crops; it was observed that two measurements are sufficient for chestnut yield and three measurements for average weight of chestnut, plant height and stem diameter; and clones 12, 13, 16, 19, 25, 26 and 31 should be indicated to give continuity to the cultivar cultivation process.