SELEÇÃO ENTRE E DENTRO EM GERAÇÃO SEGREGANTE F5 DE PIMENTEIRAS ORNAMENTAIS (Capsicum annuum L.)

Abstract

The peppers originate in the Americas, belonging to the genus Capsicum. They have a wide diffusion in the vegetable market and can be used in various forms. There is a great insertion of the peppers in the ornamental plant market because they possess characteristics such as low size, leaves, fruits and colorful flowers. Capsicum annuum L. Is the most cultivated pepper species. There are few accessions destined to ornamentation even having accessions with ornamental potential, where the best player must access all this variability through breeding programs and hierarchical methods for access to variability. Thus, this study aimed to select between and within a segregating generation F5 of ornamental peppers (Capsicum annuum L.) based on their genetic diversity for selection of families and genotypes, continuing the breeding program and Development of new cultivars. The experiment was developed in a greenhouse of the Plant Biotechnology Laboratory in the Agrarian Sciences Center of the Universidade Federal da Paraíba (CCA/UFPB). Seven segregating populations of a F5 generation of chillers and 4 additional witnesses were characterized, the UFPB 77.2 and UFPB134 accesses and 2 commercial varieties, Etna and yramid. In the first chapter, 20 Plant, flower and fruit descriptors suggested by the International plant Genetic Resources Institute (IPGRI) were used, and the experimental design was completely randomized (DIC), with additional witnesses. The data were subjected to analysis of variance (ANOVA), by the F test and subsequent averages test by the Scott-Knott criteria, (P ≤ 0.05). All analyses were performed using GENES statistical software. In the second chapter, analyses were performed based on 13 quantitative characters and 14 qualitative traits of plant and fruit. The data were subjected to analysis of variance (ANOVA), by the F test, calculation of heritability estimates, ratio of genetic and environmental coefficients of variation, canonical variables and graphic dispersion of the genotypes. The data were also submitted to multivariate analysis of variance (MANOVA), the relative importance of the characters was made based on the Singh method, from the generalized mahalanobis distance, the qualitative data subjected to Gower distance and the Tocher method through one of the three types of distance matrices, isolated and joint analysis of qualitative and quantitative data. The non-metric multidimensional scaling (nMDS) was used for graphic representation of the distance matrices and the adjustment level calculated by the Kruskal Stress. The analyses were performed using the softwares GENES and R version 3.0.3. There is genetic variability among the populations evaluated for most of the studied traits, and it is possible to select families 17.15, 56.26, 55.50, 56.8, 30.16, 30.22 and 55.45 to continue the program, a study of diversity and selection Within these families. For evaluation within each family, non-metric scaling was efficient only for the analysis of quantitative and mixed data, qualitative data showed very high stress levels, not being expressive in determining the variability Within this generation, the importance of the characters showed that for some are not being performed selection, therefore the great variability found in the genotypes in an advanced generation. The analyses were efficient in the expression of the variability within, allowing the selection within the families.