**4. Discussion**

The phylogeny used in the present study, and reconstructed using two markers of DNA barcodes for plants (rbcLa + matK), has been used in several studies to test various ecological hypotheses [21,34]. Using this DNA barcode phylogeny, our study indicates that there is a phylogenetic signal in the services provided by intentionally introduced alien woody plant species used by humans in South Africa. From a phylogenetic perspective, this is an indication of non-random selection of alien plants. However, taxonomic non-random plant selection was initially suggested almost four decades ago to explain human–plant interactions, particularly for native plants used in traditional medicine [6,41,42]. This was later supported in several other studies, but mostly for native plants used in traditional medicine [8,43–45]. Nonetheless, only a few studies have tested whether this taxonomic signal translates into a phylogenetic signal (e.g., [46,47]). In addition, the question of whether alien species introduced into a new environment follow the general pattern of nonrandom selection is not ye<sup>t</sup> widely explored (but see [48]). The present study contributes to filling such a gap, showing that alien woody species in South Africa are not randomly selected; phylogenetically closely related alien species, intentionally introduced into the region, tend to provide similar services. Can services then be used to predict the invasion status of these alien species?

Contradicting evidence has been provided in several studies that investigated the correlates of invasion success, suggesting the context specificity of invasions. Because species' functional roles (ecology and physiology) in ecosystems are linked to the services that they provide to humans [49], and functional traits drive invasion success [21,28], our expectation is that the services should also correlate with invasion success [29]. In addition, if an alien species provides a higher number of services, the chances are greater for that species not only to be sought after, but also to be introduced in a high number and independently multiple times into new environments. This is predicted in the propagule pressure theory [30–32]. Indeed, the propagule pressure theory has been demonstrated in several studies for different taxa in various geographic regions [30,50,51]. In the present study, we found a significant correlation between the number of services (which may indicate propagule pressure) and invasion status, but, contrary to expectation, it is alien non-invasive species that provide more services than invasive species. This pattern is maintained whether we corrected for phylogeny or not, and supports the finding reported in a recent study that naturalized plants provide more services than plants that are not naturalized [29].

This counter-intuitive finding may be expected if our dataset comprises a large proportion of unintentionally introduced alien species (this is not the case). It is also possible that an alien species can be introduced both intentionally and unintentionally into an area. This possibility may a priori complicate the detection of a strong correlation between services and invasion status. However, in our case, we focused only on alien species recorded as intentionally introduced and for which the services these species provide to humans are relatively well documented. As such, even if some of these intentionally introduced species are also transported through unintentional introduction pathways, this would simply increase the propagule pressure of the species and would support our expectations of strong relationships between the number of services (used as proxy for propagule pressure) and invasion status. Furthermore, even if we assume that some of the alien species in our list (Table S1) are unintentionally introduced, the fact that these species are now recorded as providing some services to humans implies that humans may further cultivate these species (for the services they provide), thus contributing to the spread of the species. In such a scenario, our hypothesis of a strong relationship between services and invasion would still hold, since we are not analyzing species' traits, but the services they provide. This scenario would actually make it meaningless to distinguish between intentional and unintentional introduction, since what matters in our approach is the services that species provide (not their ecological traits). In addition, the counter-intuitive finding reported here could possibly be because alien non-invasive species might not ye<sup>t</sup> have enough residence

time in their new environment to become invasive. It could also be because the variable "number of services" is not a strong proxy for propagule pressure, as an alien species with a high number of services may not necessarily be introduced multiple times in an area; a species S1 with only one known service may be introduced several times and more often than a species S2 with multiple services if those multiple services are less valuable to communities (i.e., lower use value) than the single service of S1. A further possibility is that the lack of positive relationships between the number of services and invasion status could be because of the differences in species' performances (ecological, physiological) in different environments/habitats in the same country; alien species providing a similar number of services for different human communities may perform differently in different environments or ecosystems. Another possibility is that the NEMBA alien categorization itself may be a source of concern, due to human misjudgment or bias, or decisions that are not ecologically informed, since the NEMBA list was generally criticized for not being science-based (it was allegedly influenced by politics). Potential bias in the list may perhaps lead to the unexpected results that we found. However, the fact that our findings mirror what was recently reported at the global scale (see ref. [29]) means that the NEMBA list may not be a profoundly biased representation of alien invasion status in South Africa. These various scenarios that potentially explain our findings call for future studies that link species' use values to their alien invasion status.

When we tested the link between invasion status and each of the service categories, only three services (food, medicine, and fuel) correlated significantly, in a negative direction, with invasion status, suggesting that these services tend to be provided by non-invasive species. All these findings confirm that alien non-invasive species tend to provide more services to humans than alien invasive species, corroborating a recent finding that naturalized species provide more services to humans at the global scale [29]. These findings prompt the following key question: by harvesting alien plants for human use, do humans limit their ability to invade? Although we did not test this hypothesis, we strongly suspect this possibility, given that alien plants providing more services, or even specific services (food, medicine), tend to be non-invasive (naturalized) or are geographically constrained.

Overall, by aiming to link services to invasion, this study pointed to potential roles played by human choices of specific products (e.g., plants for medicine) in driving species invasion. Our tests reveal unexpected evidence that alien non-invasive species provide more or unique services to humans in comparison to alien invasive species, supporting the recent similar finding reported at the global scale (see ref. [29]). Although a number of scenarios are plausible to explain our finding, we sugges<sup>t</sup> that human utilization/harvest of alien species may constrain their ability to spread and become invasive. This requires further investigations. Other studies in other geographies have shown the following similar finding with that reported in the present study: the uses of alien plants by humans determine their outcome along the introduction–naturalization–invasion continuum [52–54]. Our study also provides additional evidence that DNA barcodes, initially thought of as a taxonomic tool (e.g., [55]), can be used beyond taxonomy and for ecological investigations (see reviews in ref. [56]).

**Supplementary Materials:** The following are available online at https://www.mdpi.com/1424-281 8/13/11/553/s1: Table S1. The raw data collected for the present study. The definitions of Services 1–12 are in Table 1.

**Author Contributions:** Conceptualization, K.Y.; methodology, K.Y.; software, K.Y.; validation, K.Y.; formal analysis, K.Y. and A.E.A.; investigation, K.Y. and A.E.A.; resources, K.Y.; data curation, A.E.A.; writing—original draft preparation, K.Y.; writing—review and editing, K.Y.; visualization, K.Y.; supervision, K.Y.; project administration, K.Y.; funding acquisition, K.Y. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by National Research Foundation, gran<sup>t</sup> number Grant No: 112113.

**Institutional Review Board Statement:** Not applicable. **Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available in Supplementary Materials.

**Acknowledgments:** We acknowledge the South Africa's National Research Foundation (NRF) Research Development Grants for Y-Rated Researchers (Grant No: 112113). We thank four anonymous reviewers for their contributions to the improvement of an earlier version.

**Conflicts of Interest:** The authors declare no conflict of interest.
