Show simple item record

dc.contributor.advisorDerera, John.
dc.contributor.advisorSibiya, Julia.
dc.creatorChibanda, Mwila.
dc.date.accessioned2020-02-07T07:09:53Z
dc.date.available2020-02-07T07:09:53Z
dc.date.created2017
dc.date.issued2017
dc.identifier.urihttps://researchspace.ukzn.ac.za/handle/10413/16862
dc.descriptionMasters Degrees. University of KwaZulu-Natal, Pietermaritzburg.en_US
dc.description.abstractZambia is among the top 20 leading global producers of soybean (Glycine max L. Merr.) but adequate production is still hampered by low productivity. The yields of soybean in Zambia average below 3.0 t/ha against a yield potential of 5.0 t/ha. This is attributed in part to poor availability of well adapted and improved cultivars. Therefore, selection for high yield potential is the prime objective of the breeding programme in the medium altitude and subtropical environments in Africa. Unfortunately, spatial and seasonal variability is large in this ecological zone. Therefore, the objectives of the study were to assess the nature and magnitude of the genotype x environmental (G x E) interactions for grain yield, to identify stable genotypes; to determine the genetic gains achieved in breeding for grain yield over 12 years, and to determine the secondary traits that directly or indirectly affect yield in soybean cultivars. Thirty genotypes that were drawn from the advanced set of lines in the programme were evaluated across 16 locations in Zambia, Malawi and Zimbabwe. The experiments were laid out in a 6 x 5 alpha lattice design, with three replications at each site. The recommended cultural practices were followed at all sites in all countries. The data were subjected to analysis of variance (ANOVA, correlation and path coefficient analysis, cultivar superiority index, Additive main effects and multiplicative interaction (AMMI) and Genotype, Genotype and Environment (GGE) biplot analyses, in GenStat statistical software. There were significant genotypes main effects, environment main effects and their interaction effects. The G x E of cross over type was observed. The genotypes G2, G10 and G15 were ranked among most stable genotypes by all methods, while G2 was the most desirable genotype across locations, followed by G15. Biplot analysis revealed that E6 was the most discriminative test location while the most representative one was E4. The genetic gain study showed a 21% gain in Zambia and Malawi. No significant gain was registered in Zimbabwe. An across site analysis of all test locations resulted in a disappearance of all genetic gain earlier observed. The cross over GXE interaction negatively affected heritability of grain yield and masked the appearance of any gains. Overall, a 6.5% gain over the population mean, showed that selection was successful in increasing yield. However, there was no significant gain observed relative to the current commercial cultivars, indicating limited breeding progress. The results of PATH correlation analysis showed that yield was positively and significantly correlated with all traits except the number of seeds per pod. However, the correlation was weak with the exception of harvest index. The harvest index, biomass and number of pods per plant had significant influence on yield. Selection for these three traits, Harvest index, biomass and number of pods per plant would be emphasised to improve yield potential in the soybean programme.en_US
dc.language.isoenen_US
dc.subject.otherYield.en_US
dc.subject.otherStability.en_US
dc.subject.otherGenetic gain.en_US
dc.subject.otherSoybean.en_US
dc.subject.otherCoefficient analysis.en_US
dc.titleGrain yield stability, genetic gain and path coefficient analyses in advanced soybean (Glycine max (L.) Merr.) lines.en_US
dc.typeThesisen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record