Premium
Yield Stability of Determinate and Indeterminate Soybeans Adapted to the Northern United States 1
Author(s) -
Beaver J. S.,
Johnson R. R.
Publication year - 1981
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci1981.0011183x002100030024x
Subject(s) - indeterminate , biology , regression , regression analysis , cultivar , repeatability , stability (learning theory) , linear regression , indeterminate growth , productivity , yield (engineering) , gene–environment interaction , habit , genotype , stepwise regression , statistics , agronomy , mathematics , psychology , biochemistry , materials science , macroeconomics , machine learning , computer science , metallurgy , pure mathematics , economics , ideotype , gene , psychotherapist
Stability of performance of determinate and indeterminate soybean [ Glycine max (L). Merr.]genotypes was compared using regression on environmental means. Seed yields from portions of the 1976 to 1978 USDA Uniform Group III Soybean Test and seed yields from eight Illinois environments were analyzed in four separate joint regression analyses. Regression explained a significant portion, but not all, of the genotype × environment interaction. The Group III indeterminate cultivars in this study possessed desirable stability characteristics, having average or greater than average mean seed yields, average seed yield response to environments of varying levels of productivity, and minimum deviations from regression. Mean seed yields of some determinate genotypes were as great as the best performing indeterminate cultivars. The determinate genotypes also showed an average seed yield response to environments of varying levels of productivity. However, deviations from regression for most determinate genotypes were significant, indicating that the determinate growth habit might possess a less predictable performance when grown in the northern soybean growing region. Repeatability estimates of values of the regression coefficients and the deviations from regression across joint regression analyses were significant, suggesting that the selection of soybean genotypes with more stable performances might be possible.