Drought tolerant maize (Zea mays L.) hybrids are crucial to sustainability of maize production in West and Central Africa (WCA). Two studies were conducted at three locations in Nigeria for 2 yr to (i) assess performance of 156 early-maturing maize inbreds and three hybrid types and (ii) determine heterosis among the lines and relationship between lines per se and hybrid performance. The inbreds and their hybrids were evaluated separately under drought and well-watered conditions. Genotype, environment and genotype × environment interactions were significant for grain yield under the research conditions. Grain yield of inbreds ranged from 0.06 t ha−1 for TZEI 123 to 1.92 t ha−1 for TZEI 17 under drought. While differences in grain yield among hybrid types (single, three-way and double-cross hybrids) were not significant under drought, significant differences were detected among hybrid types under optimal conditions. GGE biplot analysis identified three inbreds, TZEI 18, TZEI 56, and TZEI 1 and hybrids TZEI 129 × TZEI 16, (TZEI 17 × TZEI 16) × TZEI 157 and (TZEI 16 × TZEI 157) × TZEI 129 as ideal across research conditions. Midparent heterosis (MPH) and high-parent heterosis (HPH) for grain yield were higher in the well-watered conditions than under drought. Positive and significant correlations existed between MPH, HPH and yield under both research conditions. Drought tolerant hybrids with stable and high yield are available for promotion for adoption by farmers in WCA.
The GGE biplot tool has potential for determining combining ability effects, identifying distinct heterotic groups and efficient testers in a line × tester study. However, its use for such analysis has not been adequately explored. The objectives of this study were to (i) assess combining ability of extra-early maturing lines (80–85 days to physiological maturity) and testers for grain yield (ii) classify lines into heterotic groups and (iii) identify most efficient testers using GGE biplot. Sixty-three lines crossed to four testers were evaluated under Strga-infested, drought and nonstress environments for 2 years in Nigeria. Results of GGE biplot analyses of combining ability and heterotic patterns of yield of lines, grouping and identification of testers were close to those of the conventional line × tester method. Testers TZEEI 13, TZEEI 21 and TZEEI 29 were highly efficient in grouping lines under stress environments while testers TZEEI 21 and TZEEI 29 were best under nonstress environments. The GGE biplot identified tester TZEEI 13, TZEEI 21 and TZEEI 29 as most efficient across stress environments and TZEEI 21 and TZEEI 29 across nonstress environments.
Authors:R.O. Akinwale, M.A.B. Fakorede, B. Badu-Apraku and A. Oluwaranti
Genotype main effect plus genotype-by-environment interaction (GGE) biplot produces a graphical display of results that facilitates a better understanding of complex genotype-by-environment interaction in multi-environment trials of breeding and agronomic experiments. However, the full potential and weaknesses of this powerful tool are not fully understood by breeders, agronomists, entomologists and pathologists. The objective of this paper was to review the usefulness of this statistical tool and enumerate some of its weaknesses. Its main application has so far been in the analysis of multi-environment data. It has been used to analyze the performance of crop cultivars under multiple stress environments, from which ideal cultivars, mega-environments, and core testing sites were identified. More recently, GGE biplot has been employed in genetic analysis of diallel data to estimate the combining abilities and identify heterotic groups among inbred parents. Genotype-by-trait biplot has also been utilized in trait profile analysis, and in identification of traits that are reliable for indirect selection of a target primary trait. Two major shortcomings of this tool are (i) failure to identify more than two distinct, contrasting groups in diallel studies and (ii) lack of statistical tests for most of its graphical displays. Other aspects of GGE biplot that need further study and development are (i) estimation of genetic variances, covariances, and heritability, including the analysis of data generated from North Carolina Designs I, II, and III as well as other genetic designs, considering their importance in plant breeding programs; (ii) analysis of Quantitative Trait Loci (QTL) data for proper understanding of the genetic constitution of each individual plant or line; and (iii) analysis of Genotype-by-pathogen or insect strain interaction data. Nevertheless, GGE biplot has helped greatly in the accurate analysis and interpretation of data from breeding and agronomic field evaluation experiments.
Authors:B. Badu-Apraku, M. Fakorede, A. Menkir, A. Kamara and S. Dapaah
Full-sib families derived from Pool 16 DT, a tropical maize (
L.) population, were evaluated in the 1995/96 and 1997/98 dry seasons, using two sites each season in Côte d’Ivoire. In all sites, the crop was irrigated from planting to about 2 weeks before anthesis, irrigation was discontinued thereafter for the rest of the season in one site in 1995 and both sites in 1997. Irrigation was continued till maturity in the second site in 1995. The means and ranges showed that the induced stress environments produced significantly lower grain yield, fewer ears per plant (EPP) and lower grain moisture percentage than the non-stressed site. The coefficients of variation (CVs) associated with the stressed environment were consistently larger than those associated with the non-stressed environment. Similarly, the coefficients of determination (R
) values were generally higher for the non-stressed than the stressed site. Only 7 families were common to the highest-yielding 20 % of the 90 families in each evaluation environment in 1995 and the 7 families were ranked differently in the stressed relative to the non-stressed environments. EPP, plant height (PHT) and ear height (EHT) had consistent positive correlation with grain yield; correlations of days to anthesis and silking date with yield were negative. The induced stress did not properly elicit the true differences among families and may, therefore, need to be modified. We are presently experimenting with a slight modification of the method by irrigating whenever the maize plants start showing signs of temporary wilting such as leaf rolling early in the day.