Iconic, threatened, but largely unknown: Biogeography of the Macaronesian dragon trees (Dracaena spp.) as inferred from plastid DNA markers

• Published in: Taxon 2020-04, Vol.69 (2), p.217-233
• Main Authors: Durán, Iván, Marrero, Águedo, Msanda, Fouad, Harrouni, Cherif, Gruenstaeudl, Michael, Patiño, Jairo, Caujapé-Castells, Juli, García-Verdugo, Carlos
• Format: Article
• Language: English
• Subjects: biogeographic disjunction; Dracaena draco; haplotype diversity; human‐mediated dispersal; island biogeography; Rand Flora; SYSTEMATICS AND PHYLOGENY
• ISSN: 0040-0262; 1996-8175
• DOI: 10.1002/tax.12215

The genus Dracaena in Macaronesia comprises two threatened species of arborescent monocots that are often associated with one of the most intriguing biogeographic disjunctions: the Rand Flora pattern. Molecular information is, however, largely missing for the Macaronesian Dracaena taxa ("MDT", hereafter), and the biogeographic or population genetic patterns of this lineage have not yet been thoroughly assessed. To fill this gap, we generated plastid DNA sequence data of 14 Dracaena populations representing the entire natural distribution of MDT (including mainland Morocco and all recognized subspecies), 9 additional populations of subspontaneous origin, and a set of related species of the genus. We performed phylogenetic, biogeographic, and population genetic analyses at different spatial scales and conducted a comparative review on plant haplotype diversity in Macaronesian plants. The results of our phylogenetic analyses indicated the monophyly of the MDT and an origin separate from a clade of geographically distant species that so far were postulated as their closest living relatives (D. cinnabari, D. ombet, D. schizantha, D. serrulata). The results of our phylogeographic analyses indicated that diversification within D. draco occurred throughout the Pleistocene and that wild peripheral populations (Madeira, mainland Morocco) may have a recent origin from Canarian source populations. Recent dispersals, coupled with remarkably low levels of haplotype diversity, probably account for the weak phylogeographic signal observed across wild populations. However, our results suggested that human-assisted expansion of Dracaena inflates the extant phylogeographic signal by non-random translocation of a specific subset of haplotypes. Our study demonstrates that many of the previous biogeographic scenarios on MDT are not supported by molecular data. Instead, our results highlight (i) the impact that human activity may have on the phylogeographic pattern of island plants, and (ii) the need of a deeper taxonomic sampling in future investigations on MDT and close relatives.