Repetibilidade e estabilidade de características agronômicas de Heliconia spp

• Ano de defesa: 2018
• Autor(a) principal: ALBUQUERQUE FILHO, João Carlos Cezar de
• Orientador(a): LOGES, Vivian
• Banca de defesa: MUSSER, Rosimar dos Santos, LIRA JUNIOR, Mario de Andrade, MENDONÇA JÚNIOR, Antonio Francisco de, SILVA, Gheysa Coelho
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal Rural de Pernambuco
• Programa de Pós-Graduação: Programa de Pós-Graduação em Melhoramento Genético de Plantas
• Departamento: Departamento de Agronomia
• País: Brasil
• Palavras-chave em Português: Helicônia; Melhoramento genético; Planta perene; Fitotecnia
• Área do conhecimento CNPq: FITOTECNIA::MELHORAMENTO VEGETAL
• Link de acesso: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/8382

There is a great demand of manpower for the number of characteristics studied and for the time spent in them in studies of characterization of perennial plants. The selection of the most important characteristics and the number of measurements to be sure of the correct evaluation of the individuals is essential. When successive evaluations are performed on the same individual, the performance of the genotype can be checked by the coefficient of repeatability, being this one dependent of the studied characteristic, of the genetic properties that governs it in the population and of the environmental conditions which are submitted, being useful in the aid of the quantification of phenotypic observations to be made in the same individual so that one has greater accuracy in the selection, smaller bias between the measure obtained and the real value of the individual, as well as in the identification of the environmental influence in the studied trait and reliable selection with a decrease in the cost of labor. In studies of perennial plants, when there is interaction genotype-environment, it is observed that the productive order of the accessions will not be repeated over time, making it necessary the use of methodologies to evaluate the adaptability and stability, the definition of environments in this sense is very broad, being able to be considered different localities, planting times, years, different forms of cultivation, etc. Studies related to stability can be carried out at the beginning of an breeding program, when the purpose of this study is to identify species and/or varieties with potential for this purpose, as in the final stages of a program, when the purpose is to select pre-selected species and/or varieties for a continuation of the program. Thus, a grouping was carried out regarding the size of 18 Heliconia accessions through 11 characteristics analyzed for three consecutive years, the determination of characteristics that contributed the most to the diversity among the groups and the verification of the proportion of the genetic and environmental expression by the coefficient of repeatability by different methods in each group. A study of adaptability and temporal stability was also made by different methodologies, knowing that these have different results, a comparison was made between similarities and divergences. It was found that the use of principal components methodology based on the covariance matrix was more efficient in the data analysis due to the higher coefficients of determination and repeatability. The characteristics length and diameter of the stem, as well as the length and width of the inflorescences, have little environmental influence. While for the study of adaptability and stability, it was concluded that for beginning of improvement process aiming at this characteristic, production of floral stems the accessions H. psittacorum and its hybrids along with H. × nickeriensis, of small and medium size, and the accesses H. rostrata and H. caribaea 'Brazilian Bomber', of large size, are the most indicated. The model with Heterogeneous Auto Regressive matrix was the most suitable for Heliconia studies and it is possible to choose between the use of the Annicchiarico method and Lin and Binns since they produce similar results for use concomitant with the Eberhart and Russell method.