Seven traits were evaluated: precocity (PRE: number of months from planting to first flowering), number of spaths (NSP: average number of spaths/year; three harvests: 2016/17, 2017/18 and 2018/19), number of fruit bunches (NFB: average number of fruit bunches/year; three crops: 2016/17, 2017/18 and 2018/19), number of fruits (NFR: average number of fruits/year; two crops, 2017/18 and 2018/19), fruit volume (FRV: cm3, sample of five fruits/plant; two harvests: 2017/18 and 2018/19), fruit mass (FRM: g, wet base corrected to 40% of moisture; sample of five fruits/plant; two harvests: 2017/18 and 2018/19), and average fruit productivity (PRD: ton/ha, wet base corrected to 40% moisture; density 400 plants/ha; two harvests: 2017/18 and 2018/19). The fruits were harvested in a staggered manner after the beginning of the natural fall of the fruits for each germplasm bank accession, this being the point of natural maturation. All data were obtained at the plant level (individual). The analysis followed the statistical model number 1: y = Xr + Za + Wp + e, where “y” is the data vector, “r” is the vector of the replication effects (assumed as fixed) added to the general mean, “a” is the vector of individual (random) additive genetic effects, “p” is the vector of plot (random) effects, and “e” is the vector of errors or residues (random). Capital letters represent the incidence matrices for these purposes. Based on the average (BLUP) and variance (REML) components obtained by the model, the predicted additive genetic values “u + a” were estimated. Data were analyzed using the Selegen-REML/BLUP software (Resende 2007).Resende MDV (2007) SELEGEN–REML/BLUP: Sistema estatístico e seleção genética computadorizada via modelos lineares mistos. Embrapa Florestas, Colombo