PRECISION OF SOME DROUGHT TOLERANCE INDICES IN EVALUATING THE BI-PARENTAL PROGENIES FOR DROUGHT TOLERANCE IN EGYPTIAN COTTON (Gossipum barbadense L.)

Two field experiments were conducted to study the response to drought stress of sixty four bi-parental (It helps in choice of breeding procedure for genetic improvement of polygenic characters) progenies in two crosses (G93x Menofy) and (G96 x C.B58) belong to Egyptian cotton (Gossypium barbadense L.). Randomized complete blocks design (RCBD) was used in well watering and stress condition arrangement with three replications during two successive cotton seasons 2016 and 2017. . Keeping the importance of production of cotton crop in view, yield and its component traits were studied to satisfy the study objectives.

Results for bi-parental analysis of variance revealed significant differences between genotypes in yield and its component traits for each of the two crosses under well watering and stress conditions. Also the results showed that differences between males and Female/males were significant. Results of correlation analysis between seed cotton yield in both conditions and calculated drought indices for each of the two crosses revealed that stress tolerance index (STI), mean productivity (MP), geometric mean productivity (GMP), yield index (YI), harmonic mean (HM) and drought resistance index (DI) were the best indices for identifying high yielding genotypes in both conditions (drought tolerant genotypes). Therefore, these indices could be used successfully as selection criteria for the screening of genotypes for performance under different water regimes. Screening drought tolerant genotypes using mean rank, standard deviation of ranks and rank sum (RS) distinguished the most drought tolerant genotypes G13 and G17 in cross (G93x Menofy) and G7 in cross (G96 x C.B58). Cluster analysis showed that the genotypes, based on indices tended to group into three groups: tolerant, semi-tolerant and sensitive genotypes. Principal component analysis (PCA), indicated that first and second components justified 99.759% and 99.892% variations for cross I and cross II, respectively among drought tolerance indices.