This experiment was conducted to investigate the genetic variation i different cumin accessions under normal and water deficient conditions and was carried out according to a randomized complete block design with two replicates for each environment at the agriculture experimental station of Isfahan university of technology in 2014-2015. The result of analysis of variance for normal condition showed significant genetic variation for seed yield and it's components and all of studied morphological traits between the accessions. Drought stress had significant effects on morphological traits. The heritability of traits was estimated to be more than 50% and indicated that most of this variation are due to genetic effects. Among all yield components, umbel per plant and seed per umble had the most correlation with seed yield at both moisture conditions. The result of stepwise regression for normal condition showed that umbel per plant defined most of the variation for yield; in this manner at drought stress, seeds per umbel defined most of the variation for yield and can be used as an index for selection. Principle component results for both moisture levels, showed three components justified more than 70% of total variations. The resulting clusters divided accessio into three groups and PCA results largely confirmed cluster analysis. In most cases, the accessions were not represents suitable ability of SRAP markers to determine genetic diversity in cumin. The genetic distance among groups was obtained between 0.04 and 0.14. regarding within groups variation and variation among accessions, results indicated that the most similarity was obtained between 4(Joupar -Kerman) and 3(Marave Tappeh -Golestan), while the lowest one belonged to 4(Joupar -Kerman) and 22(Estahban- Fars). Numerical value of 0.13 obtained for gene flow between groups (Nm) showed that there is gene flow in groups but this current cannot lead to homogenization of the groups. The mean of shanon information index was 0.44 for primer compositions. In the cluster analysis of SRAP data, dendrogram could not Keyword : Cumin, Genetic Variation, SRAP markers, Accessions
