In this work, deleterious effects in soils due to the presence of dielectric fluids were investigated. For this purpose, vegetable (Envirotemp® FR3) and mineral (Lubrax AV 66 IN) oils were used for simulating a set of soils contaminated in different oil contents (0.5, 1.0, 2.0, 2.5, 5.0, 7.5 and 10%) in which three 120-days soybean crop periods (SCP) were carried out using the species Glycine max (L.) Merr. Both soil and soybean plant samples were analysed on following the changes on chemical attributes, content of oils and greases (COG) in soils and phytotechnical characteristics of soybean plant. No significant changes on soil chemical attributes were found. For a 0.5% vegetable oil fraction, COG removals of 35, 60 and 90% were observed after the 1st, 2nd, and 3rd SCPs, respectively, whereas removals of 25, 40 and 70% were observed for 0.5% mineral oil fraction after the 1st, 2nd, and 3rd SCPs, respectively. There was an effectively accumulated removal on all tested oil fractions as being proportional to the integrated 120-days SCPs, suggesting a lesser number of crops for a complete abatement of oil fraction in soil. A 100% recovery on the seedlings emergence fractions was also evidenced, revealing that at least a number of 7 and 9 SCPs should be applied continuously in soils contaminated by vegetable and mineral oils, respectively, in order to no longer jeopardize soybean plant growth. Finally, an empirical prediction of the number of SCPs necessary for the complete removal of oil from the soil was proposed.