Statistical Analysis of Grain Yield in Iranian Cultivars of Barley (Hordeum vulgare)

Abstract

To determine the most important traits influencing grain yield of barley, 10 cultivars were evaluated in three field experiments in randomized block design with three replicates during 2013–2014 growing season. In each experiment, 10 yield component traits were evaluated. Following a combined analysis of variance, multivariate statistical analyses such as simple correlation coefficients, factor analysis, stepwise analysis, path analysis and cluster analysis were administered on combined means. The correlation coefficient analysis showed significant positive correlation of grain yield with grain filling period, biological yield, harvest index, number of seed/spike, number of seed/m2, 1000-seed weight and number of spikes/m2. Factor analysis divided the 11 measured variables into five factors. Factor 1 was affected by grain yield and grain filling period. Factor 2 was affected by number of spikes/m2 and number of seed. Factor 3 was associated with days to hardening and days to maturity. Factor 4 was influenced by number of seed/spike, number of spikes/m2 and 1000-seed weight. Fifth factor was strongly affected by biological yield and plant height. In stepwise regression analysis for grain yield as independent variable, two traits include biological yield and harvest index entered to regression model in four steps with R-square = 0.98. Regression coefficient of both entered variables was positive and significant at 1% probability level. The path coefficient analysis based on grain yield, as a dependent variable, implicated that biological yield had the highest positive direct effect on grain yield. Highest positive indirect effect on grain yield was related to harvest index through biological yield. Cluster analysis using Ward’s method divide 10 investigated cultivars into three clusters. Cluster 1 specified to cultivars with the highest plant height, number of seed/spike and 1000-seed weight. Cluster 2 was specified to lines with higher amount of grain yield, biological yield and number of seeds/m2. Two control cultivars include Sina and Nosrat were located to cluster 3. Finally, based on GGE biplot analysis, Sina, which has higher yield and more relative stability, can be suggested as superior than the other cultivars.