| Data from segregating populations developed from 1990--2002 at College Station (CS), TX and from 1992--2002 at Weslaco (W), TX were used in this evaluation. The early generation test (EGT) at CS utilized yield and quality parameters of the F2 population to identify populations to advance for later generation pedigree selection. A modified EGT was used in W, F2 and F3 populations were evaluated for yield and fiber quality with these data used to determine advanced strain performance potential. Correlation analysis was used to determine the effect of EGT on the number of advanced strains and their performance at a number of locations in central and south Texas.;No associations (P = 0.05) were found between F2 EGT data from CS and the number of advanced strains selected from each population for lint yield, lint percent, fiber elongation, and uniformity index. At W, F 2 yield data were associated (P ≤ 0.08) with the number of selections through the F6 generation, but there were no associations of F 2 fiber properties and the number of strains selected. However, when F3 EGT data from W were used, associations (P ≤ 0.10) were found between fiber length, fiber strength, and uniformity index and the number of advanced strains selected.;Correlations between EGT yield and advanced stain performance trials did not support EGT at CS, however, fiber properties of F2 populations were significantly associated with advanced strain fiber properties. At W, EGT lint yield data did correlate with performance trial data, thereby supporting the use of EGT where lint yield was the primary selection criteria.;Early generation test data can provide the breeder with useful information relative to lint yield, fiber length, fiber strength, and fiber elongation of advanced strains, especially when evaluated at multiple locations and well defined selection goals are observed. The necessity of multiple locations and years probably reflects genotype by environment variation. |
