Aim: Alien plants with certain characteristics may have been introduced earlier and more frequently than others. Such introduction bias may cause spurious associations between plant characteristics and naturalization (the establishment of self-sustaining populations in the wild). We aimed to disentangle direct and indirect (i.e. mediated by introduction history) effects of species characteristics on the naturalization success of alien plants introduced for horticulture. Location: Germany (non-native range); rest of the world (native range). Methods: We compiled a dataset of 435 alien plant species introduced in cultivation in Germany, including data on their year of introduction, the number of botanical gardens where they are planted, native range, biological traits and naturalization success. We used path analysis to estimate the direct effects of geographical origin and biological traits on naturalization, and their indirect effects mediated by year and/or frequency of introduction. Results: We found significant direct positive effects of native range size and winter hardiness on naturalization. Alien species native to other parts of Europe and species with a large native range were brought to the country earlier than other species. In addition, woody species, winter-hardy species and tall species were planted more frequently than others. Because the number of botanical gardens where a species is planted increased naturalization success directly, and residence time did so indirectly through a significant positive association with the number of botanical gardens, most of the species characteristics had indirect effects on naturalization. Main conclusions: Our approach allowed us to show that apparent effects of species characteristics on naturalization success can be at least partly indirect, due to introduction biases. This indicates that failure to recognize such introduction biases could impair our ability to explain the success of alien plant species.

DatasetThe file contains the dataset. A brief description of the variables is available in the ReadMe file. The data are explained in more detail in the Methods section of the paper (see Maurel et al., 2016, Global Ecology and Biogeography).Maurel_GEB_2016_dataset.csv