Abstract

The estimate of genetic progress indicates the effectiveness of selection and the need to use new selection methods and strategies. Hence, the objective of the present study was to accuracy estimate the genetic contribution of agronomic traits in the irrigated rice breeding program in the state of Minas Gerais, Brazil between 1998 and 2012. For this goal, the following traits were evaluated: Grain yield (Kg.ha^-1), 100-grain weight (g), plant height (cm), days to flowering (days), tillering (score) and lodging (score). These traits were evaluated for 108 inbred lines in advanced yield trials (ATs), conducted in randomised blocks with three to four replicates in four regions of Minas Gerais between 1998 and 2012. Not all sites were included in all crop years, according to the selection accuracy. The restricted maximum likelihood/best linear unbiased predictor Restricted Maximum Likelihood/Best Linear Unbiased Predictor (REML/BLUP) technique was used to obtain reliable genetic value estimates. The genotypes were evaluated for sites within each year and between years by analysis of deviance. Further, genetic and environmental progress were estimated. There was a significant effect for genotype and for the interactions genotype x site and genotype x year. The following results were obtained: 195.91 kg.ha^-1 for yield, 0.10 g for 100-grain weight, 1.50 cm for plant height, 3.17 days to flowering, 0.01 points for tillering, and -0.18 points for lodging. Although these results are satisfactory, new strategies are suggested to increase the genetic progress values of the agronomic traits of interest in the coming years.

Key words: Genetic contribution , value for cultivation and use, selection accuracy, Restricted Maximum Likelihood/Best Linear Unbiased Predictor (REML/BLUP), Oryza sativa L.