STABILITY ANALYSIS OF SOME SOYBEAN GENOTYPES USING A SIMPLIFIED STATISTICAL MODEL | ||||
| Journal of Plant Production | ||||
| Article 4, Volume 6, Issue 12, December 2015, Page 1975-1990 PDF (741.46 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/jpp.2015.52124 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
| A. R. Morsy1; W. M. Fares2; S. B. Ragheb1; M. A. Ibrahim1 | ||||
| 1Food Legume Crops Res. Sec., Field Crops Res. Inst., ARC, Giza. | ||||
| 2Central Lab. for Design & Stat. Analysis Res., ARC, Giza. | ||||
| Abstract | ||||
| The genotype × environment (G×E) interaction is considered a stumbling block to plant breeders, since the presence of significant GxE interaction component can complicate the identification of superior genotypes and reduce the usefulness of selection. Seed yields of 26 soybean genotypes were evaluated in three locations i.e. Sakha, Etay ElBaroud and Mallawy, through four successive summer seasons from 2012 to 2015. The used design was a randomized complete block design with three replications. This research is aimed to estimate the stability parameters of seed yield of 26 soybean genotypes over twelve environmental conditions and to examine the usefulness and validity of a new simple stability method comparing with four widely used methods. The four stability methods follow three main statistical models namely; regression, variance, and non-parametric approaches. Results showed highly significant mean squares for genotypes, environments and G×E interaction indicating that the tested genotypes exhibited different responses to environmental conditions giving the justification for running stability analysis. The terms of predictable (linear) and unpredictable (non - linear) interaction components were highly significant indicating that the tested soybean genotypes were different in their relative stability. The two soybean cultivars Giza 111 and Giza 21 in addition to their high mean yields, they met all the rules of stable genotypes. Therefore, both cultivars could be considered a good breeding material stock in any future breeding program. Also, when the simplified stability method was applied, the unstable eighteen genotypes were differentiated into three classes. These classes included three genotypes (L162, H29 L115, and H2 L12 ) were adapted to the unpredictable low yielding environments, while five others (H15 L273, L163, H3 L4, H4 L24 and DR 101) were adapted to high yielding environments. Whereas, the rest ten genotypes were unstable over the low, medium and high environmental groups. The results proved also that, the proposed stability method of Thillainathan and Fernandez (2002) is very simple and easy to apply, understand and interpret by agronomists and plant breeders than the other popular stability models. Also, it is possible to support the results of this stability method by a scatter plot diagram that enable the researchers to visually, directly and quickly compare the mean yield performance and stability of the tested genotypes. | ||||
|
Statistics Article View: 129 PDF Download: 414 |
||||
