Cutting and removal of oil-impacted marsh plants are still used worldwide as a clean-up and recovery technique. To experimentally test the efficacy of cutting and removing marsh plants under subtropical conditions, we simulated an oil spill (Bunker MF-180) in Spartina alterniflora marshes and compared the responses of plant height, biomass, density of culms and number of flowering plants in high and low energy areas in Paranaguá Bay (S Brazil) for about 9 months. Cutting and removal were inefficient in promoting or accelerating the recovery of the impacted areas. Cut or uncut impacted marshes fully recovered within 6 months, both in low and high energy areas. Plant cutting should be practiced only when there is an effective risk of contamination of groundwater near urban areas, when obvious aesthetical issues are involved in areas of touristic interest or when there are real short-term conservation risks to threatened species.