Premium
Stability Analysis of Yield in Maize, Wheat and Sorghum and its Implications in Breeding Programs *
Author(s) -
Yue Guilan,
Perng S. K.,
Walter T. L.,
Wassom C. E.,
Liang G. H.
Publication year - 1990
Publication title -
plant breeding
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.583
H-Index - 71
eISSN - 1439-0523
pISSN - 0179-9541
DOI - 10.1111/j.1439-0523.1990.tb00405.x
Subject(s) - sorghum , hybrid , gene–environment interaction , biology , linear regression , stability (learning theory) , agronomy , adaptability , plant breeding , regression analysis , poaceae , yield (engineering) , statistics , zea mays , mathematics , genotype , ecology , genetics , machine learning , computer science , metallurgy , gene , materials science
Abstract Six stability statistics: (b i , s 2 di , , , and ) were estimated for maize, wheat and sorghum in different environments by using three statistical models. The significant linear portion of genotype × environment interaction for maize indicates different hybrids responded differently to environments, whereas the non‐significant genotype × environment interaction (linear) were found for wheat and sorghum suggest that all genotypes responded similarly as the environments change. However, the highly significant pooled deviations (deviation from regression) for all three crops make yield predictions from the model less reliable. When regression coefficients (b i ) were non‐significant, s 2 di , became an important statistic in estimating stability. It appears that the regression coefficient, b i , was best used to estimate genotypic adaptability, whereas s 2 di , for stability. Maize and sorghum had negative correlations between the mean yield and stability statistics, s 2 di , , and , suggesting that high yield and stability are not mutually exclusive in the range of environments used in this study; however, such correlations did not occur in wheat. Thus, maize and sorghum hybrids with high yield potential and high stability could be identified and selected. Correlations between mean yield and b i , or , were positive and significant for maize and sorghum but were non‐significant for wheat, indicating that such relationships may be species specific. Under a given set of testing environments, the stability ranking associated with each maize hybrid is correlated to and depends on other hybrids included in the analysis.