Comparison of Stability Parameters for Detection of Stable and High Essential Oil Yielding Landraces of Rosa damascena Mill

The essential oil yield stability of damask rose (Rosa damascena Mill.) as an important medicinal and aromatic plant in different environments has not been well documented. In order to determine appropriate stability parameters, six statistics were studied for essential oil stability of 35 Rosa damascena landraces in seven locations (Sanandaj, Arak, Kashan, Dezful, Stahban, Kerman, and Mashhad) and two years (2007-2008) in Iran, using a randomized complete blocks design with three replications. Significant differences (P<0.01) were observed in essential oil ratio among landraces (G), locations (L), and environments (E) and in landrace × location (GL) and landrace × environment (GE) interactions. The positive correlation of environmental variance (S2), coefficient of variation (CV), and regression coefficient of yield over environments (b) with essential oil suggest that only low-yield landraces develop a similar phenotype over a range of environments and show static stability. Although there were not any stable landraces by b (b ≈ 0) and all of the stable ones by S2 produced very low yields, some of the adaptable ones by CV (e.g. KM1) showed high essential oil ratios and stability simultaneously. The stable landraces according to the dynamic stability concept (b ≈ 1, Sd2 or variance due to deviation from regression ≈ 0) produced moderate essential oil. Superiority index (P) determined some of the highest essential oil as adaptable landraces. The stable landraces with the least variance of the years within places (MSY/P) produced the least essential oil. It could be concluded that a genotype can demonstrates both static and dynamic stability with high essential oil content. In addition, CV, dynamic view statistics (b ≈ 1, Sd2≈ 0), and P are proposed as desirable parameters for the evaluation of essential oil stability with different concepts in damask rose genotypes.