The objective of this paper was to estimate the repeatability coefficient (RC) of the fruit traits such as height (FH), diameter (FD), weight (FW), C vitamin (VitC), total soluble solids (TSS), pulp weight/fruit weight relation (PFR) and the titrable acidity (TA), while the number of measures that should be performed for an efficient selection of the genotypes under evaluation were determined. Analysis were made using the ANOVA, principal components and structural analysis methodology. The estimates of the RC for FH, FD, FW, VitC and TA demonstrated high regularity at the superiority of the individuals from one cycle to another, so the evaluation of two cycles was enough to predict the individuals real value with R² above 90%. To TSS, in average, eight evaluation cycles were necessary to reach a R² above 90%. PFR showed to be highly irregular, so the evaluation of at least 26 cycles turned to be necessary in order to reach a R² above 90%. Such an irregularity was attributed to the method used in analyzing the characteristic, then the search for a more accurate and stable method was necessary.
Malpighia emarginata; chemicophysical properties; selection; quantitative genetics
