Divergência genética de castanha-do-brasil (Bertholletia excelsa Humb. & Bonpl.) considerando variáveis biométricas de sementes.

Abstract

Studies have shown that with the increase in deforestation, exploitation, and excessive collection in areas of native forests where the populations of the chestnut trees of Bertholletia excelsa would meet, they may cause the disappearance of this species in its natural habitat. Therefore, it is of utmost importance to develop research aimed at factors that can influence the quality and germinative potential of seeds, fruit production and the variability of individual production in different matrix trees, in order to define the areas of seed collections with the intention Brazil nuts may be introduced in commercial plantations in the future. The present study aimed to verify the genetic divergence between genotypes of Bertholletia excelsa in three different origins in the Northern Region of Mato Grosso, Amazonian Forest, Brazil, based on the biometric characteristics of the seeds. The fruits and seeds of Bertholletia excelsa were collected in the municipality of Paranaíta, MT. A total of 1,515 seeds were evaluated for length, width, and thickness with the aid of a digital caliper and the width, thickness and total area of seeds were also evaluated with the ImageJ® Software. After obtaining the data, analysis of variance (ANOVA) was performed, and then the averages of the analyzed characteristics were grouped and compared by the Scott-Knott test at 5% probability. Pearson's linear correlation coefficient (r) was calculated using correlation data (Pachymeter x ImageJ). The mean Euclidean distance, cluster analysis using the method (UPGMA), and canonical variables (VC) were generated to obtain the dissimilarity measures. This study obtained results in which it was possible to verify a wide phenotypic variation, mainly for thickness characteristic in the digital caliper. It is also evident that digital image processing is a great tool to verify the existence of genetic divergence in seeds for selection of matrix trees. Thus, the species under study has excellent qualities to be used as a seed collection area, as well as in genetic breeding programs for non-wood forest products.