Path analysis of Rosa damascena Mill. performance under different conditions

In this research, effects of different traits on flower yield of 12 Rosa damascena Mill. accessions from different regions of Iran were analyzed under irrigation (control) and non-irrigation with and without mulch conditions. Randomized complete block designs were applied with three replications at Research Station of Homand Absard, Damavand, Iran. Flower weight, flower number, flower dry matter percentage, oil percentage, oil yield (per ha), petal weight, plant height and canopy cover were recorded. According to the correlation analysis, flower yield positively correlated with petal weight under non-irrigation with mulch, irrigated (control) and non-irrigation without mulch conditions. The statistical multivariate methods were also applied included multivariate linear regression and path analyses. Using three separate stepwise-regression analysis made it possible to determine the traits with most important impacts on the flower yield, under irrigation (control) and non-irrigation with and without mulch conditions. Flower number, petal weight, oil yield, and oil percentage under non-irrigation without mulch totally explained 91.7% of the variation among genotypes in the regression model. Petal weight, plant cover, and diameter of the species under irrigation (control) totally explained 98.6% of the variation among genotypes in the regression model. Oil percentage, oil yield, petal weight, plant height and flower number per ha under non-irrigation with mulch totally explained 90.6% of the variation among genotypes in the regression model. The results indicated that petal weight has the most direct impact on flower yield based on the path analysis and oil yield showed a negative direct effect on the flower yield under non-irrigation (without mulch) condition. Oil percentage, oil yield, petal weight, plant height and flower number per plant under non-irrigation with mulch, irrigated (control) and non-irrigation without mulch totally explained 95/5% of the variation among genotypes in the regression model. Petal weight had the highest direct impact on flower yield based on the path analysis under irrigation and non-irrigation conditions. Therefore, it is possible to consider petal weight as a magnificent character to improve flower yield of Rosa damascene.