Multivariate Analysis Techniques In Biometric Variability Characterization Of Camu-Camu Fruit On Indigenous Populations.

The camu-camu brings attention to the high vitamin C content (877-6116 mg/100 g pulp). This Amazon indigenous species is in a domestication process, being underexploited in Brazil. Germplasm Banks make possible selection among and within progenies, based on the characteristics of ideotypes for commercial and nutraceutical use of this species. Therefore, this study aimed to identify the fruit biometric characteristics that contributed to the largest intraspecific variability of camu-camu native populations in Roraima state, using the principal component analysis and cluster analysis. Camu-camu accessions were collected between December, 2009 and March, 2011 from 16 indigenous populations. In the prospection sites, each plant was considered as an accession and was also obtained the geographic coordinates with a GPS. During the harvest, the fruits were packed in polystyrene box with ice and at the end of the expedition, taken to the horticulture sector of Embrapa Roraima. We evaluated a total of 247 subsamples, analyzed by computer program- Info Gen, version. 2012. Parameters such as fruit height (mm), fruit diameter (mm), average fruit weight (g), pulp yield (%), epidermal yield (%), seed yield (%), soluble solids content (º Brix), the average seed number per fruit and ascorbic acid (mg100g pulp + peel-1) were evalluated. Multivariate statistical analysis of principal components and cluster helped to determine which variables have entered the higher intraspecific variability in the characteristics evaluated. The analysis suggests that the total mass of fruit pulp yield, soluble solids content and ascorbic acid have the greatest discriminating characters for selection intraspecific camu-camu. The camu-camu populations studied showed significant genetic variability that can be exploited for the conservation and improvement of the species.

 Keywords: Myrciaria dubia, multivariate analysis, prospecting, intraspecific selection.