Path Analysis of Grain Yield and Yield Components and Some Agronomic Traits in Bread Wheat

Development of new bread wheat cultivars needs efficient tools to monitor trait association in a breeding program. This investigation was aimed to characterize grain yield components and some agronomic traits related to bread wheat grain yield. The efficiency of a breeding program depends mainly on the direction of the correlation between different traits and the relative importance of each component involved in contributing to grain yield. Correlation and path analysis were carried out in 56 bread wheat genotypes grown under field conditions of Maragheh, Iran. Observations were recorded on 18 wheat traits and correlation coefficient analysis revealed grain yield was positively correlated with stem diameter, spike length, floret number, spikelet number, grain diameter, grain length and 1000 seed weight traits. According to the variance inflation factor (VIF) and tolerance as multicollinearity statistics, there are inconsistent relationships among the variables and all traits could be considered as first-order variables (Model I) with grain yield as the response variable due to low multicollinearity of all measured traits. In the path coefficient analysis, grain yield represented the dependent variable and the spikelet number and 1000 seed weight traits were the independent ones. Our results indicated that the number of spikelets per spikes and leaf width and 1000 seed weight traits followed by the grain length, grain diameter and grain number per spike were the traits related to higher grain yield. The above mentioned traits along with their indirect causal factors should be considered simultaneously as an effective selection criteria evolving high yielding genotype because of their direct positive contribution to grain yield.