BIOMETRICS OF MACAÚBA FRUITS

Macaúba (Acrocomia aculeata) is a very promising palm tree, with wide distribution in tropical America, with productive potential above 8 tons/ha of vegetable oil, with the first commercial crops in the state of Minas Gerais. In this context, research has acted strongly for technological advancement for improvements in the production system and selection of superior genotypes, requiring the use of precise and less costly methodologies. Therefore, the objective of this study was to construct prediction models to estimate the weights of the components of macaúba fruits, by means of physical characters that are easy to measure. Five accessions of natural populations of macaúba introduced in the BAG of Embrapa Cerrados were used for the study: Paranoá-DF, Formosa-GO, Carmo do Paranaíba-MG and Lagoa Formosa-MG. 20 ripe fruits from each access were evaluated for the physical characteristics of the bipartite fruit and variables related to these characteristics were obtained. Afterwards, the actual weight of each fruit component (peel, pulp, endocarp and almond) was observed.  Based on the set of variables obtained, stepwise multiple regression analysis was used to construct regression models to obtain the estimated weight of each component. The validation of the models by the correlation between the observed versus estimated data reached coefficients of determination (R²) above 0.8 for pulp and almond weight