Soils and environmental pollution by total petroleum hydrocarbons (TPHs) is one of the most important environmental challenges in Iran especially around oil refineries like the Tehran oil refinery complexes. Presence of petroleum contaminants in soils may be toxic to human and organisms and act as a source of ground and surface waters contamination. Also, there are serious concerns regarding the presence of TPH in soil, especially because of their potential for bioaccumulation in organism’s food chain. Therefore, remediation of these compounds from environment is necessary. There are many techniques for remediation of petroleum-contaminated soils but, some of them are high-cost and have side effects on environment, therefore nowadays bioremediation technologies such as phytoremediation (removal of soil contaminants via plant growth) are investigated. In this study, phytostimulation was used as an environmentally and economically acceptable option to remediate petroleum-contaminated soils. At first, germination and subsequent growth of seven plants were tested in three soils with different petroleum contamination levels. Contamination treatments consisted of C 0 (uncontaminated soil), C 1 and C 2 treatments (1:1 and 1:3 w/w, uncontaminated: contaminated soil, respectively). Thereafter, four plant genotypes with the highest germination and growth were selected for phytoremediation trails. The results showed that the presence of TPH in the soil had no effect on seed germination of agropyron, white clover, sunflower and safflower. In contrast, there were 32 and 56% decreases in germination of puccinellia in the C 1 and C 2 treatments, respectively. Canola and white clover seedlings were sensitive to these compounds and failed to produce dry matter yield at the end of trial period. Therefore, four plant genotypes including tall fescue, agropyron, safflower and sunflower were selected for the phytoremediation trails. In the phytoremediation study, shoot and root dry matter yield of all studied plants significantly reduced in the contamination treatments, particularly C 2 treatment, as compared to the control (C 0 treatment). Also, microbial activity in the rhizosphere of agropyron was greater than tall fescue in the C 2 treatment. However, there were no significant differences in the microbial activity in the rhizosphere of sunflower and safflower with the control. Decreases of the TPH concentration in the presence of the plant genotypes were variant. There were 71 and 69% decreases in TPH concentration in C 1 treatment, and 45 and 42 decreases in TPH concentration in C 2 treatment in the presence of agropyron and tall fes
