Understanding population connectivity is fundamental to ecology, and, for sedentary organisms, connectivity is achieved through larval dispersal. We tested whether coastal topography influences genetic structure in Perna perna mussels by comparing populations inside bays and on the open coast. Higher hydrodynamic stress on the open coast produces higher mortality and thus genetic turnover. Populations on the open coast had fewer private haplotypes and less genetic endemism than those inside bays. Gene flow analysis showed that bays act as source populations, with greater migration rates out of bays than into them. Differences in genetic structure on scales of 10s of kilometres show that coastal configuration strongly affects selection, larval dispersal and haplotype diversity.
KEY WORDS: Dispersal · Coastal topography · ConnectivityResale or republication not permitted without written consent of the publisher Mar Ecol Prog Ser 368: 189-195, 2008 Bay morphology influences oceanography, not only by retaining water and, thus, changing circulation patterns and consequently affecting larval dispersal (Roughan et al. 2005), but also by creating different selective environments from those of the open coast. In bays, populations are often subjected to high thermal stress, different food concentrations and lower salinities compared to open coast sites (Ricketts & Calvin 1968, Castilla et al. 2002, Largier 2004. However, wave action, which can be the major cause of mussel dislodgement in the intertidal habitat (Paine & Levine 1981), is higher on the open coast than in bays (Ricketts & Calvin 1968).In the present study, we test the hypothesis that the influence of coastal topography on larval dispersal and selective regime produces a discernible and predictable effect on the genetic structure of mussel populations. Our null hypothesis is that open coast sites have the same genetic diversity and number of private haplotypes as sites in bays. Mussels disperse by means of planktotrophic larvae that can stay in the water column for a period of weeks to months before settling to the substratum and being recruited into adult populations (Hicks & Tunnell 1995). In addition, wave force was measured at the same sites on 4 different occasions. Mortality rates of Perna perna were measured at 2 bay and 2 open coast sites after a storm to determine the role of hydrodynamic stress in genetic turnover.
MATERIALS AND METHODSSampling, DNA extraction, amplification, and sequencing. Populations of Perna perna (shell length 4 to 5 cm) were sampled at 3 sites inside bays (Plettenberg Bay, Jeffreys Bay, Algoa Bay) and 3 open coast sites (Cape St. Francis, Cape Recife, Kenton-on-Sea; n = 15 each population; Fig. 1) from low intertidal rocky shores of South Africa. These bays are typical of a series of half-heart or log-spiral bays found on this coast. They include a long stretch of sandy beach, with patches of sand-influenced granite rock (Bownes & McQuaid 2006). All populations belong to the same genetic lineage (Zardi et al. 2007)...