A phylogeographic analysis of the North American medicinal leech, <i>Macrobdella decora</i> (Say, 1824)

Abstract In spite of their important roles in many ecosystems, data regarding population structure and biogeographic patterns of leeches are scarce. To begin to address this knowledge gap, we herein perform a phylogeographic analysis of the North American medicinal leech, Macrobdella decora (Say, 1824). A total of 224 M. decora specimens were collected from 35 localities across large swaths of USA and Canada and covering most of the known range of the species. Using four loci (mitochondrial cytochrome c oxidase subunit I [COI] and NADH dehydrogenase I [ND1], as well as nuclear 18S rRNA [18S] and 28S rRNA [28S]), we construct phylogenetic trees using several optimality criteria and superimpose geographic patterns onto the trees in order to tease out any potential structure among the populations. Rather surprisingly, given the large geographic range of the species and abundance of potential geographic barriers to gene flow, the analyses showed a conspicuous lack of structure among the different populations of M. decora . However, an AMOVA did show statistically significant differences between the genetic variation within populations and between populations (COI: FST = 0.65412, p &lt; .00001; ND1: FST = 0.69245, p &lt; .00001), which was largely driven by only 6 out of the 35 populations, and indicated a potential barrier for dispersal across the Appalachian Mountains. Finally, a Mantel test showed a weak, but significant, correlation between geographic distance and genetic distance (COI: r = 0.209, p = .027; ND1: r = 0.1289, p = .030); however, this correlation was primarily driven by a single locality. The overall weak structure suggests that M. decora is panmictic throughout its range, and we discuss this in light of previous population level studies in both bloodfeeding and non‐bloodfeeding species, concluding that the lack of structure in M. decora might be due to its high capacity for dispersal via hosts.