/AnMtgsAbsts2009.52703 The Use of AMMI and GGE Biplot Methods for Evaluation of Tropical Maize Hybrids and Environmental Stratification.

Wednesday, November 4, 2009
Convention Center, Exhibit Hall BC, Second Floor

Renzo Pinho1, ROGÉRIO L. OLIVEIRA2, MÁRCIO BALESTRE1 and DENIS V. FERREIRA3, (1)AGRICULTURE DEPARTMENT, Federal Univ. of Lavras, LAVRAS-MG, Brazil
(2)BIOLOGY DEPARTMENT, Federal Univ. of Lavras, LAVRAS, Brazil
(3)BIOLOGY DEPARTMENT, Federal Univ. of Lavras, LAVRAS-MG, Brazil
Abstract:
The objective of this work was to evaluate the stability, adaptability and environmental stratification analyses, through the AMMI and  GGE biplot methods, using trials data from tropical maize hybrids evaluations, and to compare the efficiency of these methods. For that, data obtained from the evaluation of twenty-three tropical maize hybrids in eleven sites of Brazil and two agricultural harvests, 2005/2006 and 2006/2007, were used.  The data were submitted, first, to variance analyses and following that, to the adaptability, stability and environmental stratification analyses, through the AMMI and GGE biplot methods. Hybrid 15 was the highest yielding. However, hybrids 10 and 16 were those that presented the best combination between adaptability and stability, considering the analysis of the AMMI2 and GGE biplot graphs respectively. The stratification promoted a reduction of up to 28% in the number of locals. The GGE biplot captured a larger portion of the sum of the squares of (G)+(GxA) compared to the AMMI1 model, with a more efficient and of easy interpretation graph, when compared to the AMMI2 graph. By checking the results it is concluded that it is possible to reduce  the number of evaluation sites, and that the GGE biplot analysis was more efficient than the AMMI.