[en] The growing interest in medicinal plants from both international industry and local markets requires management of tree bark harvesting from natural forests in order to prevent inappropriate exploitation of target species. This study was designed to determine the bark re-growth response of a selected number of medicinal tree species as a basis for the development of an optimal bark harvesting method. In 2004, bark was harvested from 925 trees belonging to 12 species in 38 sites in a dry forest in Benin, West Africa. Two years later, the response of trees to bark harvesting was examined with respect to re-growth (edge or sheet), development of vegetative growth around the wound, and the sensitivity of the wound to insect attack. Two species, Khaya senegalensis and Lannea kerstingii, showed complete wound recovery by edge growth. At the other extreme, Afzelia africana, Burkea africana and Maranthes polyandra had very poor edge growth. M. polyandra showed good sheet growth, whereas the other 11 species had none or poor sheet growth after total bark harvesting. In contrast, partial bark removal allowed better sheet growth in all 12 species studied. Insect sensitivity was species-specific. Insect attacks were negatively correlated with non-recovered wound area, but there was a marked species effect for the same rate of regeneration. L. kerstingii and K. senegalensis had very good and similar re-growth, but L. kerstingii was very susceptible to insect attack, whereas K. senegalensis appeared to be very resistant. Only a few individuals developed vegetative growth, and each tree usually developed only one or two agony shoots, but there was no significant difference between species. Synthesis and applications. This is the first study to provide data on the ability of trees to close wounds after bark harvesting in West Africa. We report large variability in the response of different species to our bark harvesting technique, and identify just two out of the 12 study species as suitable for sustainable bark harvesting. Based on our results, we developed a decisional step method to help forest managers select the best techniques for managing medicinal tree species as an alternative to bark harvesting, for example, coppice management, harvesting leaves instead of bark, stand establishment, and collaboration with timber companies.