Molecular Ecology (2004) 13, 2143–2156 doi: 10.1111/j.1365-294X.2004.02225.x © 2004 Blackwell Publishing Ltd Blackwell Publishing, Ltd. Strong genetic clines and geographical variation in gene flow in the rocky intertidal barnacle Balanus glandula ERIK E. SOTKA,* JOHN P. WARES,† JOHN A. BARTH,‡ RICHARD K. GROSBERG † and STEPHEN. R. PALUMBI * * Stanford University, Department of Biological Sciences, Hopkins Marine Station, Pacific Grove, CA 93950, † Center for Population Biology, Division of Biological Sciences, 1 Shields Drive, University of California, Davis, CA 95616, ‡ College of Oceanic and Atmospheric Sciences, Oregon State University, 104 Ocean Administration Building, Corvallis, OR 97331, USA Abstract A long-standing issue in marine biology is identifying spatial scales at which populations of sessile adults are connected by planktonic offspring. We examined the genetic continuity of the acorn barnacle Balanus glandula, an abundant member of rocky intertidal communities of the northeastern Pacific Ocean, and compared these genetic patterns to the nearshore oceanography described by trajectories of surface drifters. Consistent with its broad dispersal potential, barnacle populations are genetically similar at both mitochondrial (cytochrome oxidase I) and nuclear (elongation factor 1-alpha) loci across broad swaths of the species’ range. In central California, however, there is a striking genetic cline across 475 km of coastline between northern and southern populations. These patterns indicate that gene flow within central California is far more restricted spatially than among other populations. Possible reasons for the steep cline include the slow secondary introgression of historically separated populations, a balance between diversifying selection and dispersal, or some mix of both. Geographic trajectories of oceanic drifters closely parallel geographical patterns of gene flow. Drifters placed to the north (Oregon; ∼44°N) and south (Santa Barbara, California; ∼34° N) of the cline disperse hundreds of kilometres within 40 days, yet over the long-term their trajectories never overlapped. The lack of communication between waters originating in Oregon and southern California probably helps to maintain strong genetic differentiation between these regions. More broadly, the geographical variation in gene flow implies that focusing on species-level averages of gene flow can mask biologically important variance within species which reflects local environmental conditions and historical events. Keywords: California Current, cline, dispersal, marine invertebrate, mtDNA sequences, nuclear sequences, selection Received 22 December 2003; revision received 25 March 2004; accepted 7 April 2004 Introduction Virtually all species of bottom-dwelling marine organisms are distributed in patches that are linked by the dispersal of planktonic larvae (Thorson 1950). As such, larval dispersal plays a fundamental role in the ecology and evolution of marine organisms and their biotic interactions (see re- views in Caley et al . 1996; Botsford et al . 2001; Grosberg & Cunningham 2001; Strathmann et al . 2002). Genetic estimates of the spatial extent of larval dispersal generally support a positive correlation between a species’ ability to disperse — as measured by larval lifespan — and the actual distances they travelled (Bohonak 1999). However, a number of more recent studies on fish and invertebrates with long-lived planktonic larvae indicate that planktonic larvae can be locally retained (e.g. Jones et al . 1999; Swearer et al . 1999; Barber et al . 2000; Taylor & Hellberg 2003), suggesting that the realized dispersal of marine organisms is probably less extensive in distance than estimates based on life-history alone predict. More importantly, there is growing recognition that realized dispersal is probably more complicated than can be estimated from limited sampling from a small portion Correspondence: Erik E. Sotka. Fax: 831-655-6215; E-mail: sotka@stanford.edu