Canonical correlation and direct and indirect effect on the selection of sweet potato accessions collected in Vale do Ribeira

• Main Author: Lopes, Beatriz Garcia
• Publication Date: 2022
• Other Authors: Otoboni, Maria Eduarda Facioli, Vargas, Pablo Forlan, Pavan, Bruno Ettore, Oliveira, Darllan Junior Luiz Santos Ferreira de
• Format: Article
• Language: por
• Source: Research, Society and Development
• Download full: https://rsdjournal.org/index.php/rsd/article/view/34022
• topic: Ipomoea batatas; Melhoramento genético; Eficiência de utilização.; Ipomoea batatas; Mejora genética; Eficiencia de utilización.; Ipomoea batatas; Genetic improvement; Utilization efficiency.

Sweet potato is an important vegetable in Brazil and in the world, due to its productive capacity and nutritional quality of its roots. Despite this importance, there are few breeding programs aimed at the development of new cultivars. Therefore, it is extremely important to know information about the nature and magnitude of phenotypic variations observed in a plant population. Therefore, the objective was to evaluate the canonical correlations between productive characters, as well as the direct and indirect effects of sweet potato accessions collected in traditional communities in Vale do Ribeira. The design was in randomized blocks, with 95 accessions (treatment) and two commercial cultivars, Brazlândia Branca and Brazlândia Roxa, thus totaling 97 accessions with three replications. The following were evaluated: total and commercial productivity of roots, percentage of commercial products, the content of dry mass of roots, the total productivity of dry mass of roots, productivity of fresh vines, percentage of dry matter in the vines, productivity of dry mass of the vines. Each characteristic was submitted to analysis of variance, trail analysis, and canonical correlations were made between the group of variables. After analyzing the results, it is concluded that the character's Total production of vines and Dry mass of vines did not show importance in the study because their direct effects on the main variable are low. The canonical correlation showed that there was no association between the variables evaluated.