Acessibilidade / Reportar erro

Multi-environment tests in the selection of rice genotypes using sites regression model

The analysis of genotype x environment interaction in plant breeding have been enlarged with new methodologies in the last decade, improving its efficiency on the selection of genotypes under different environmental conditions. The objective of this research was to analyze the yield and stability of twelve genotypes of rice, in eight environments, during the years 2005 and 2006 in Colombia. Completely randomized block designs with four replications were used. The phenotypic stability parameters and grouping of environments were estimated by the genotype-environment interaction study according to SREG (Sites Regression) method and its biplot graphic (GGE). The statistical analysis indicated significant differences (with 5% probability error) among genotypes and among environments. Also, it pointed out the significance (with 5% probability error) of the genotype-environment interaction, indicating different responses of genotypes confronted with different environments. In SREG method, the two first principal components of interactions explained 75.29% of the interaction. The genotypes 400094, 350361 and the variety Fedearroz were found as the highest yields. According to the biplot GGE graphic the environments La Libertad and Escobal were the ones with small variations during the years of study.

phenotypic adaptability; biplot GGE graphic; linear-bilinear model; SREG


Universidade Federal de Santa Maria Universidade Federal de Santa Maria, Centro de Ciências Rurais , 97105-900 Santa Maria RS Brazil , Tel.: +55 55 3220-8698 , Fax: +55 55 3220-8695 - Santa Maria - RS - Brazil
E-mail: cienciarural@mail.ufsm.br