Four species of trees were selected to evaluate the tolerance to heavy crude oil contamination by means of a tolerance index integrating germination, height, biomass and survival as variables. Fresh seeds to Cedrela odorata (tropical cedar), Haematoxylum campechianum (tinto bush), Swietenia macrophylla (mahogany) and Tabebuia rosea (macuilis) were planted in a Vertisol to which heavy crude petroleum was added at four different treatments (C0, 0; C1, 18,940; C2, 44,000; and C3, 57,000 mg kg⁻¹), with the control being uncontaminated soil. The experiment was carried out in a greenhouse during 203 days with a completely random design. The presence of petroleum in soil stimulated and increased germination of S. macrophylla and C. odorata, accelerated the germination of T. rosea and did not affect the germination of H. campechianum. The height and biomass of all species was reduced in the presence of petroleum in the soil. The survival of S. macrophylla and H. campechianum was not affected by petroleum at any concentration studied. On the other hand, C. odorata and T. rosea showed high mortality at all concentrations. The tolerance index showed that S. macrophylla was best at tolerating petroleum in soil and could be employed as a productive alternative for the advantageous use of contaminated sites. The use of tree species could be important because of the great potential of trees for phytoremediation due to their long life, biomass and deep roots that can penetrate and remediate deeper soil layers.

Under greenhouse conditions, we evaluated establishment of four tree species and their capacity to degrade crude oil recently incorporated into the soil; the species were as follows: Cedrela odorata (tropical cedar), Haematoxylum campechianum (tinto bush), Swietenia macrophylla (mahogany), and Tabebuia rosea (macuilis). Three-month-old plants were planted in soil with three treatments of heavy petroleum and a control (C0 0 mg kg⁻¹; C1 18,000 mg kg⁻¹; C2 31,700 mg kg⁻¹; C3 47,100 mg kg⁻¹) with four repetitions per treatment and species; the experiment was carried out for 245 days. Height and biomass of all species significantly diminished as petroleum concentration increased, although plant survival was not affected. The quantity of colony-forming units (CFU) of rhizospheric bacteria varied among tree species and treatments; petroleum stimulated bacterial CFU for S. macrophylla. The number of fungi CFU for S. macrophylla and T. rosea was significantly greater in C0 than in soil with petroleum, but among species and among different concentrations, no significant differences were found. The greatest percentage of total petroleum hydrocarbon (TPH) degradation was found in C1 for soil without plants (45 %). Differences from the remaining treatments (petroleum concentrations in soil and plant species) were not significant (P < 0.05). Among all trees, H. campechianum had the greatest TPH degradation (32.5 % in C2). T. rosea (C1) and H. campechianum (C2) resulted in petroleum degradation at levels ranging from 20.5 to 32.5 %. On the basis of this experiment, the tree species used did not improve TPH degradation. However, all of them showed high rates of survival and vigor. So, as tree species provide goods and services, experiments with inoculation of hydrocarbonclastic microorganisms, addition of fertilizers, and mixture of tree and grasses are recommended.