This study was carried out to determine the yield performances of 37 Damask rose (Rosa damascena Mill.) genotypes across eight environments of Iran over 2 years (2007-2008). The experimental layout was a randomized complete block design with three replications. Data were collected and analyzed using the AMMI and Eberhart and Russel stability analysis methods. Significant differences (p <= 0.01) were observed for genotypes, Environment, and genotypexenvironment interaction (GE). The first five principal component axes were significant (p<0.01) and cumulatively contributed to 92% of the total GE interaction. A biplot generated using genotypic and environmental scores of the first two AMMI components also showed that a genotype that falls near the center of the biplot (small IPCA1 and IPCA2 values) may have wide adaptation. The genotypes of 4, 33, 36, 38 and 39 as general stable and adaptable had so high flower yield. The genotypes that occur close to particular environments on the IPCA2 vs. IPCA1 biplot show specific adaptation to those environments. The best genotypes with higher specific adaptability were recognized for 8 environment as follows: Markazi (37, 2 and 31), Kurdistan (16, 3, 36 and 39), Fars (3, 33, 36 and 39), Kerman (2, 40, 9 and 16), Khozestan (7, 22, 17 and 39), Hamadan (16, 32, 3 and 11), Khorasan (16, 15,28 and 6), Esfahan (2, 10, 8 and 17). According to Eberhart and Russel method 6 the genotypes 4, 33, 36 and 39 were most stable coupled with higher flower yield over eight environments.