Understanding why alien plant species are incorporated into the medicinal flora in several local communities is central to invasion biology and ethnobiology. Theories suggest that alien plants are incorporated in local pharmacopoeias because they are more versatile or contribute unique secondary chemistry which make them less therapeutically redundant, or simply because they are locally more abundant than native species. However, a lack of a comprehensive test of these hypotheses limits our understanding of the dynamics of plants knowledge, use and potential implications for invasion. Here, we tested the predictions of several of these hypotheses using a unique dataset on the woody medicinal flora of southern Africa. We found that the size of a plant family predicts the number of medicinal plants in that family, a support for the non-random hypothesis of medicinal plant selection. However, we found no support for the diversification hypothesis: i) both alien and native plants were used in the treatment of similar diseases; ii) significantly more native species than alien contribute to disease treatments particularly of parasitic infections and obstetric-gynecological diseases, and iii) alien and native species share similar therapeutic redundancy. However, we found support for the versatility hypothesis, i.e., alien plants were more versatile than natives. These findings imply that, although alien plant species are not therapeutically unique, they do provide more uses than native plants (versatility), thus suggesting that they may not have been introduced primarily for therapeutic reasons. We call for similar studies to be carried out on alien herbaceous plants for a broader understanding of the integration of alien plants into the pharmacopoeias of the receiving communities.