| Summary: | This study was planned with the purpose of evaluating the drought tolerance of advanced breeding lines of chickpea in natural field conditions. Two methods were employed to impose field conditions; the first: simulating drought stress by growing chickpea genotypes at five rainfed areas, with Faisalabad as the non-stressed control environment; and the second: planting chickpea genotypes in spring to simulate a drought stress environment, with winter-sowing serving as the non-stressed environment. Additive main effects and multiplicative interaction (AMMI) and generalized linear models (GLM) models were both found to be equally effective in extracting main effects in the rainfed experiment. Results demonstrated that environment influenced seed yield, number of primary and secondary branches, number of pods, and number of seeds most predominantly; however, genotype was the main source of variation in 100 seed weight and plant height. The GGE biplot showed that Faisalabad, Kallur Kot, and Bhakkar were contributing the most in the GEI, respectively, while Bahawalpur, Bhawana, and Karor were relatively stable environments, respectively. Faisalabad was the most, and Bhakkar the least productive in terms of seed yield. The best genotypes to grow in non-stressed environments were CH39/08, CH40/09, and CH15/11, whereas CH28/07 and CH39/08 were found suitable for both conditions. CH55/09 displayed the best performance in stress conditions only. The AMMI stability and drought-tolerance indices enabled us to select genotypes with differential performance in both conditions. It is therefore concluded that the spring-sown experiment revealed a high-grade drought stress imposition on plants, and that the genotypes selected by both methods shared quite similar rankings, and also that manually computed drought-tolerance indices are also comparable for usage for better genotypic selections. This study could provide sufficient evidence for using the aforementioned as drought-tolerance evaluation methods, especially for countries and research organizations who have limited resources and funding for conducting multilocation trials, and performing sophisticated analyses on expensive software.
|
|---|
