Testing the role of meander cutoff in promoting gene flow across a riverine barrier in ground skinks (Scincella lateralis).

Despite considerable attention, the long-term impact of rivers on species diversification remains uncertain. Meander loop cutoff (MLC) is one river phenomenon that may compromise a river's diversifying effects by passively transferring organisms from one side of the river to the other. However, the ability of MLC to promote gene flow across rivers has not been demonstrated empirically. Here, we test several predictions of MLC-mediated gene flow in populations of North American ground skinks (Scincella lateralis) separated by a well-established riverine barrier, the Mississippi River: 1) individuals collected from within meander cutoffs should be more closely related to individuals across the river than on the same side, 2) individuals within meander cutoffs should contain more immigrants than individuals away from meander cutoffs, 3) immigration rates estimated across the river should be highest in the direction of the cutoff event, and 4) the distribution of alleles native to one side of the river should be better predicted by the historical rather than current path of the river. To test these predictions we sampled 13 microsatellite loci and mitochondrial DNA from ground skinks collected near three ancient meander loops. These predictions were generally supported by genetic data, although support was stronger for mtDNA than for microsatellite data. Partial support for genetic divergence of samples within ancient meander loops also provides evidence for the MLC hypothesis. Although a role for MLC-mediated gene flow was supported here for ground skinks, the transient nature of river channels and morphologies may limit the long-term importance of MLC in stemming population divergence across major rivers.