STATISTICAL TOOLS FOR EVALUATION AND SELECTION OF CORN HYBRIDS ACROSS MULTIPLE ENVIRONMENTS AND YEARS
Data of 34 trials established during the last three years (2017-2019), mainly in the Azuero Region in Panama, were used to evaluate some statistical tools for decision-making when releasing new genotypes, with the precision that these are superior to the controls in common use. The number of hybrids evaluated per year varied from 20, 15 and 18 genotypes, respectively. For this study, six hybrids in common were taken over the years, comparing them with the most used control in the country (30F-35). The original experimental design was Alpha Lattice with three repetitions, which varied over years. An analysis of variance was carried out individually and in a combined type REML in a Complete Random Block design for the analysis of the seven selected genotypes. These analyzes showed highly significant differences between the different hybrids evaluated for grain yield variable and other agronomic characteristics. This analysis showed that by reducing the number of cultivars, the variance between genotypes was reduced; while the variance between environments was increased. Within the evaluated hybrids, three exceeded the combined overall average, with P-4039, ADV-9789, and ADV-9779 standing out significantly, with averages exceeding 8,30 ton.ha-1. The Biplot GGE-SReg analysis identified ADV-9779 as the most stable hybrid across localities. The analysis of the reliability of the normalized response indicated that P-4039 exceeds the control in 87% of the locations. The used methodologies were confirmed to be useful and simple to identify superior genotypes for release in the evaluated areas of influence.
| Autores principales: | , |
|---|---|
| Formato: | Digital revista |
| Idioma: | spa |
| Publicado: |
Instituto de Innovación Agropecuaria de Panamá
2021
|
| Acceso en línea: | http://www.revistacienciaagropecuaria.ac.pa/index.php/ciencia-agropecuaria/article/view/417 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|