Regression and causality analysis of traits related to fodder performance in different populations Elymus tauri from northwest Iran

This research was conducted in order to investigate simple correlation coefficients, multiple correlation, regression and causality analysis of morphological traits 12 Elymus tauri populations from northwest Iran in Tabriz Botanical Garden in the form of randomized complete block design in 4 replications in the fall of 2012. Simple and multiple correlation coefficients, downward regression, causality analysis of traits related to performance were calculated. The results showed that there was a high correlation between the dry yield with the total number of tillers, the number of fertile tillers, the number of leaves per tiller, the length of the flag leaf, the dry weight of the plant, the wet weight of the plant, and the wet yield. These traits have a direct relationship with performance and were considered as components of performance. Estimation of step-by-step upward correlation revealed the role of traits of the total number of leaves per plant and the number of leaves per leaf in increasing the yield. In addition, the high balanced multiple correlation coefficient of traits with performance (Radj=0.842) indicated that these traits should be taken into consideration in selecting and improving the superior stands in terms of biological performance.