Biol. Bull. 208: 60-68. (February 2005)
© 2005 Marine Biological Laboratory
Diversification Before the Most Recent Glaciation in Balanus glandula
John P. Wares* and
Clifford W. Cunningham
Department of Biology, Duke University, Durham, North Carolina 27708
* To whom correspondence should be addressed, at Department of Genetics, University of Georgia, Athens, Georgia 30602. E-mail: jpwares{at}uga.edu
 |
Abstract
|
|---|
A deep genetic cline between southern populations of the barnacle Balanus glandula (from about Monterey Bay southward) and northern populations (from northern California through Alaska) has recently been described. If this pattern is due to historical isolation and genetic drift, we expect it to have formed recently and represent a transient, nonequilibrium state. However, this cline appears to have formed well before the last glacial maximum. Our assays of sequence diversity at a region of mitochondrial cytochrome oxidase I, combined with coalescent estimators of the time of separation for these two regions, suggest that a late Pleistocene event more than 100 thousand years ago may be responsible for the initial separation. This suggests that either strong oceanographic mechanisms or natural selection have maintained the cline, because there has clearly been adequate time for this cline or polymorphism to resolve itself by genetic drift and migration. However, reliance on only a single mitochondrial marker for which the substitution rate has been estimated still limits the resolution of our analysis.
Abbreviations: kya, thousand years ago • LGM, last glacial maximum • Ne, effective population size • mtCOI, mitochondrial cytochrome oxidase I
|
Introduction
|
|---|
Patterns of genetic variation in some coastal species indicate the effects of changing paleoclimate on distribution and abundance (e.g., Edmands, 2001; Dawson, 2001; Hickerson and Ross, 2001; Marko, 2004). These effects may lead to significant levels of population structure, where some regions harbor significantly different frequencies of particular alleles or molecular markers (Grosberg and Cunningham, 2001). Predictions range from extreme divergence of regions, indicated by reciprocal monophyly on a gene tree of regional samples (see Avise, 2000), to gradients in overall genetic diversity. Patterns suggesting divergence between populations currently distributed to the north and south of the glacial margin at the last glacial maximum (LGM)—about 18 thousand years ago (kya)—are often explained as representing populations that survived Pleistocene glaciation in "unglaciated refugia" (Holder et al., 1999; Hickerson and Ross, 2001; Wares, 2002). A finding of lower genetic diversity in populations north of the LGM glacial margin is typically interpreted as post-glacial expansion from a southern refugial distribution (see Marko, 2004). In marine organisms, the location and persistence of paleoceanographic phenomena must also be considered, such as the location of extant currents and their persistence through Pleistocene climate change (Herbert et al., 2001; Wares, 2002).
Sotka et al. (2004) characterized the genealogical relationship among individuals of Balanus glandula from coastal British Columbia to southern California, using DNA sequence data from two loci (mitochondrial cytochrome oxidase I [mtCOI] and nuclear elongation factor 1
). The analysis of these data indicates a clinal pattern of allelic class frequencies, suggesting that there are surprisingly low levels of realized migration (gene flow) along the California coast from Point Arena to Monterey Bay. The divergence between the northern and southern lineages does not represent a classical "phylogeographic break" in which the two groups are genealogically distinct (class I phylogeographic status; Avise 2000). Instead, the pattern is consistent with secondary contact between historically isolated lineages, which has proceeded slowly because of selection or physical oceanographic forces. One way to distinguish among the mechanisms that may be maintaining this pattern is to estimate how long ago the cline formed. If the cline is quite ancient, it becomes less likely that neutral processes could have maintained it.
The effective population size (Ne; Hudson, 1990) is a parameter that describes the idealized average number of individuals contributing genetic diversity each generation. When estimated from genealogical data using methods based on coalescent theory, this parameter represents the long-term size of the inbreeding population and thus describes the demography of a population over time scales of 103–105 years (Turner et al., 2002). The time of isolation between populations can also be measured in terms of Ne. Here we apply these methods to mitochondrial data taken from B. glandula populations between southern Alaska and southern California (see Table 1). Although other molecular data sets are available (e.g., Sotka et al., 2004), estimates of the substitution rate µ have only been calculated for this portion of the mtCOI locus in balanoid barnacles (see Materials and Methods). Using these data, we estimate regional Ne and the range of divergence times between the southern and northern allelic classes of B. glandula to determine whether this divergence was associated with the LGM as is commonly predicted, or if it significantly predates the LGM (see Marko, 2004).
View this table:
[in this window]
[in a new window]
|
Table 1 Summary of collection data and estimates of effective population size (Ne) for populations of Balanus glandulus
|
|
Because it is difficult to distinguish between weak selection and physical oceanographic forces in maintaining this genetic cline (Sotka et al., 2004), these demographic parameters may reflect the predictions made by physical oceanographers regarding changes along the continental shelf during the Pleistocene (Gaines and Roughgarden, 1985; Roughgarden et al., 1988; Wing et al., 1995; Lyle et al., 2001) that would create barriers to gene flow, and areas of larval retention. The formation and maintenance of this cline essentially requires such selection or oceanographic structuring, without which it should collapse quickly (Irwin, 2002; Sotka et al., 2004). By narrowing the range of historical demographic scenarios associated with this cline, we can better infer the conditions that are maintaining it.
|
Materials and Methods
|
|---|
Balanus glandula (Darwin, 1854) is distributed in dense aggregates on rocky shores from northern Alaska to just south of the border between the United States and Mexico (Barnes and Barnes, 1956; Newman and Abbott, 1980) and is readily distinguished from other cirriped species by exoskeletal characters of the operculum, parieties, and basis (Newman and Abbott, 1980). Individuals (n = 370) were haphazardly collected in the field (see Table 1 for collection locations) and placed in 95% ethanol prior to DNA isolation. All molecular methods are as in Wares et al. (2001); a 710-bp fragment of the mitochondrial cytochrome oxidase I gene (mtCOI) was amplified using the universal primers of Folmer et al. (1994), with sequence data deposited in GenBank (Benson et al., 2004), as AF234351-234462 and AY795085-795281, and the aligned data set available from EMBL (Kulikova et al., 2004), as ALIGN-000504. This mitochondrial data set is entirely independent of the one presented in Sotka et al. (2004), using different individuals, mostly collected 2–3 years earlier, and a different fragment of the mtCOI gene; these data include almost three times as many individuals as in Wares et al. (2001) and eight newly sampled populations (Juneau, Alaska; Bamfield, British Columbia; Fogarty Creek and Strawberry Hill, Oregon; and Point St. George, Cape Mendocino, Point Arena, and Bodega Bay, California). Phylogenetic analysis of these data was performed using PAUP*4.0b10 (Swofford, 2002). We followed the methods of Wares et al. (2001) to obtain the full set of parsimony trees, and we used Arlequin 2.001 (Schneider et al., 1997) to analyze sequence data for geographic associations.
We estimated effective population size (Ne) within populations and pairwise migration among populations, as well as divergence time, using four related but distinct methods. (1) To treat the continuous coalescent description of population relationships, we used the method of Wilkins and Wakeley (2002) to estimate Ne and neighborhood size (Nb; roughly the per-generation effective gene flow distance) for populations south of the cline (see Fig. 1), within the clinal region, and north of the clinal region. (2) Coalescent estimation of Ne and pairwise migration M was calculated using MIGRATE 1.7.6 (Beerli and Felsenstein, 1999). Replicate analyses were performed using the methodology of Turner et al. (2002). (3) Estimates of diversity in terms of 
(S) and (
) were calculated using Arlequin 2.001 (Schneider et al., 1997). These estimates, combined with the estimated substitution rate at mtCOI for balanomorph barnacles (Wares, 2001; Wares and Cunningham, 2001), were used to calculate frequency-based estimates of allelic age (Slatkin and Rannala, 1997). (4) Recent improvements in Markov chain Monte Carlo (MCMC) coestimation of ancestral parameters allow the coordinated estimation of migration parameters between regions as well as the divergence time between those regions. The program IM (Hey and Nielsen, 2004) estimated these parameters, again grouping the "northern" and "southern" regions, both across Monterey Bay and across Cape Blanco, Oregon. This latter comparison is primarily used as a control estimate of divergence; because analyses of divergence time are rarely verified against populations with known divergence using an independent estimate of µ, we estimate this parameter for Cape Blanco because no phylogeographic transition is present in B. glandula at this location (Sotka et al., 2004; see Results). Thus, only a minimal divergence is expected relative to the divergence across the observed cline.
View larger version (28K):
[in this window]
[in a new window]
|
Figure 1. A. maximum-parsimony phylogram representing phylogeographic diversity of parsimony-informative sites in Balanus glandula (n = 363). For reference, dark circles represent individuals collected from south of Point Conception (34.5°N) and open stars represent individuals collected from the 3 northernmost collection sites. In general, the "northern" clades harbor only a single individual collected from south of Monterey Bay (Kirk Creek). As indicated by the plot in B, the frequency of the "southern" allelic type (see text) decreases rapidly between Monterey Bay (MB; California) and Cape Blanco (CB; Oregon). Here the southern type is defined as those individuals not found in the clades formed by substitutions at sites 273 and 360 from the reported alignment of mitochondrial cytochrome oxidase I. Binomial sampling error of the frequency of the southern type from each population is indicated.
|
|
Because of computational difficulties with searching and resolving large genealogies for ancestral population parameters, two groups of MCMC analyses were performed using IM: the full data set was separated at Monterey Bay for analysis with 106 chains after discarding the first 105 chains. Then, as in Turner et al. (2002), the data partitions were randomly sampled for 20 sequences from each regional group to allow the analysis to stabilize parameter results more efficiently. A program, SEQSAMPLER.RB (written in Ruby and available from JPW), generated 40 pseudoreplicate data sets that were each analyzed twice using distinct starting seeds and the same initial parameters as for the whole data set.
|
Results
|
|---|
Sequences from the southernmost populations of Balanus glandula (south of Monterey Bay) are statistically distinct from those to the north. Although phylogenetic analysis shows this distinction to be robust (Wares et al., 2001), the signal can be summarized by looking at only two nucleotide positions (273 and 360 in the aligned data set of 632 bp, of which 61 positions are parsimony informative and 50 are variable but parsimony uninformative; Fig. 1). The southernmost populations are fixed for an adenine residue at these nucleotide positions; the branches supporting the two clades that are limited to northern populations are themselves supported in 100% of maximum parsimony trees (consensus). This trend is significant when measured by 
st (Wares et al., 2001). No populations from Cape Mendocino northward or Pacific Grove southward are significantly differentiated from each other within regions, but comparisons between regions are highly significant in analysis of molecular variance. Figure 2 shows that this pattern is driven by the secondary contact of genetically distinct populations, as the apparent isolation by distance pattern (Fig. 2a, Mantel test of pairwise Fst matrix against pairwise geographic distance, P < 0.01) breaks down when clades are analyzed separately, even though overall geographic distribution is comparable (Fig. 2b). Equilibrium isolation by distance (sensu Wright, 1943) represents limited dispersal and gene flow, but nonequilibrium dynamics (such as vicariance) can artificially create the same genetic patterns (Peterson and Denno 1998).
View larger version (15K):
[in this window]
[in a new window]
|
Figure 2. Isolation by distance plots for haplotype data in Balanus glandula. (A) All data combined; vertical axis is pairwise Fst between populations and horizontal is pairwise distance in kilometers between populations. A Mantel test (1000 matrix permutations) indicated that the relationship between Fst and distance is significant (P < 0.01). (B) Sequence data from the clades that predominate in northern populations (substitutions 360 and 273 on Figure 1; black) and the clade that predominates in southern populations (white) analyzed separately. Neither relationship is statistically significant.
|
|
Analysis of divergence time between the lineages that dominate either northern or southern populations of B. glandula indicates that the initial formation of the phylogeographic discontinuity is more ancient than the last glacial maximum (LGM) by at least an order of magnitude. Here we assume that the substitution rate for cytochrome oxidase I in B. glandula is similar to that measured for other barnacles (µ 
1.55 x 10–8 in Chthamalus, Wares, 2001; µ 2.18 ± 1.1 x 10–8 in Semibalanus, Wares and Cunningham, 2001). The method of Wilkins and Wakeley (2002), using the mean of these two estimates, indicates that the maximum likelihood estimation of the effective population size (Ne) in populations to the north of the cline is approximately 251,200 and slightly lower in the populations to the south of the cline (199,500) and in the clinal region itself (208,900). In a purely panmictic population, the time to the most recent common ancestor is expected to be 2Ne (Hudson, 1990); thus the combined set of allelic classes in each region indicates that current diversity has arisen in the past 500 thousand years or so. This same method suggests that the neighborhood size is indistinguishable from infinity for the regions to the north and south of the cline, while populations within the cline have a neighborhood size of less than 1000 m, which corresponds to the results of Sotka et al. (2004).
Estimates of = 2Neµ for each population sampled ranged from 1.80 to 4.66 in our MIGRATE analyses. However, for some population samples we were unable to get a stable estimate from this method. Estimates of were somewhat correlated (0.45) with estimates of made using Arlequin, and log plots of this parameter show the same basic trend; because of difficulties with obtaining stable estimates using MIGRATE, however, we report only the results from Arlequin. In Table 1, Ne at each site is estimated using () and the above substitution rates. Frequency-based methods (Slatkin and Rannala, 1997) of estimating the age of alleles indicate that the "northern" mitochondrial haplotype groups are likely to have arisen well before the LGM, as well. The average () within each allelic class and the frequency of each class is used in these estimates: southern group () = 3.192 (119–253 kya); northern groups () = 2.489 and 1.600 (93–197 kya and 59–127 kya, respectively).
The mean time of divergence from our two full-data analyses using IM is t = 4.66 (4.64, 4.68 were the maximum likelihood estimates from each replicate), where t = µt (t is time in years; Hey and Nielsen, 2004). This mean generates a divergence time of 338,000 (±113,000) years using a per-gene substitution rate of 1.4 x 10–5(±6.9 x 10–6; Wares and Cunningham, 2001). Migration across Monterey Bay was limited (estimated as m1 = 0.91 from north to south, and m2 = 0.99 from south to north), consistent with the finding of limited gene flow in Sotka et al. (2004) and the restriction indicated by analysis with the Wilkins and Wakeley (2002) method. The same values estimated from the 80 pseudoreplicate data sets were t = 1.34 ± 1.47, and much lower estimates of m1 and m2 (0.44, 0.39). From the 82 IM analyses, the minimum divergence time between the northern and southern regions across Monterey Bay is t = 0.595 (43.2 ± 14.5 kya). Estimated effective population size in both regions is much higher in the IM analysis (8.7 x 106 in populations north of Monterey Bay, 7.0 x 106 in populations to the south). The pseudoreplicate analyses may be conflated by significantly higher estimates of because any small subsample of the complete, highly diverse data set involves long branch lengths between individuals, perhaps suggesting much more unsampled diversity than exists when the entire data set is available.
Replicate comparisons of populations across Cape Blanco (excluding populations south of the clinal region), despite the lack of phylogeographic signal based on other types of analyses (Wares et al., 2001; Sotka et al., 2004), are not different from the Monterey Bay comparisons. The mean divergence time of pseudoreplicate data sets using IM is in fact higher (t = 1.67 ± 1.64 kya), although the average estimates of migration across Cape Blanco are also much higher (m1 = 3.56 and m2 = 4.16). As with the pseudoreplicate analyses across Monterey Bay, estimates of were highly inflated by the subsampling process by nearly an order of magnitude over estimates based on (). The average across- Cape Blanco estimate using the entire data set (excluding populations south of Monterey Bay) is t = 4.479 (m1 = 3.58 and m2 = 2.72).
|
Discussion
|
|---|
Estimating the age of an evolutionary event by using molecular data is often a circular task. Without adequate fossil or biogeographic evidence, inferences based on genetic diversity must rely on consistency among different analytical methods. Here, calculations of population age and time of separation between populations of Balanus glandula on either side of a significant genetic cline suggest that the separate allelic groups likely formed well before the last glacial maximum (LGM). However, these estimates have broad confidence intervals, indicating that in some cases we cannot reject a more recent event. Few molecular markers are available for which the substitution rate µ is known with confidence, and without better cases for which this parameter has been estimated independently of the biogeographic or paleontological hypothesis being tested, it will be difficult to improve such estimates. In this study, we did not obtain significantly different estimates of divergence when comparing regions on either side of the strong genetic cline than when we compared populations on either side of a separate geographic landmark for which there is no evidence of genetic differentiation. Thus, these methods and estimates must be applied with caution.
When there is complete random mating across the samples, the expected time to the most recent common ancestor is 
2Ne generations (Hudson, 1990), but there is high variance around this prediction (Hudson and Turelli, 2003) and an inference from a single gene may be misleading. Our estimates of Ne = /2µ are on the order of 105 for each sampled location, similar to estimates of Ne (by the method of Wilkins and Wakeley, 2004) when comparing regional groups (north and south) separated by the multilocus cline between Monterey Bay and Pt. Arena. Frequency-based methods (Slatkin and Rannala, 1997) also suggest that these groups diverged during or prior to marine isotope stage 5 (74–131 kya; Lyle et al., 2001). Under panmixia, we expect similar estimates of diversity at multiple spatial scales (Hudson, 1990; Beerli, 2004). However, methods designed to estimate divergence time (t) and gene flow simultaneously (Hey and Nielsen, 2004) suggest much higher Ne in each region (on the order of 106–107) and a much older divergence. Methods that do not simultaneously account for ongoing migration following the divergence of populations frequently estimate lower separation times (Wares et al., 2002). However, estimates of t are similar for populations across Cape Blanco and Monterey Bay in this study, despite strong differences in estimated migration regimes across these two geographic features. These estimates also vary widely depending on the sample size from each region. If within-region coalescence is not significantly different from among-region coalescence, it should indicate high gene flow and the homogenization of population structure.
The tendency to attribute marine phylogeographic patterns to the LGM is probably associated with the overall lack of information on earlier glacial events and the impact each stage had on nearshore geography and oceanographic patterns (Valentine and Jablonski, 1993; Lyle et al., 2001). Nevertheless, it is clear that divergence among populations is often far more ancient than the LGM (Edmands, 2001; Dawson et al., 2001; Hickerson and Ross, 2001; Marko, 2004). The time required for significant divergence between isolated populations is generally on a much longer scale than 20–40 thousand years (Palumbi and Kessing, 1991; Wares, 2002). In B. glandula, the presence of similar levels of diversity [()] throughout the studied range (Table 1) suggests long-term persistence in habitats north of glacial margins (e.g., southern Alaska), indicating that unlike other coastal species (Hickerson and Ross, 2001; Hellberg et al., 2001; Dawson, 2001; Jacobs et al., 2004; Marko, 2004), B. glandula has not expanded northward from a southern refugial population.
It is important to recognize the difference between primary divergence and secondary contact in this and other systems (Endler, 1982). The statistical signal for isolation by distance in B. glandula is driven solely by the divergence of the northern and southern allelic types, because removal of those diagnostic nucleotide sites—or analysis within allelic "type" alone—eliminates any signal of geographic differentiation (Fig. 2). No obvious or statistically significant differences exist between settlement classes at this mitochondrial locus or at several allozyme loci across large spatial scales, different intertidal heights, or among settlement sites (Hedgecock, 1986; Wares, unpubl. data). These patterns are robust across other data types as well—although only a couple of microsatellite loci have been isolated for B. glandula, gene flow also appears to be quite high between populations throughout the northern range of this species, including across Cape Blanco (R. J. Toonen and J. P. Wares, unpubl. data).
The potential for the cline between northern and southern allelic types to be maintained by selection, rather than by the introgression of historically differentiated populations of this barnacle, is not yet known (Sotka et al., 2004), which complicates any demographic interpretations of these data on the basis of neutral expectations. It is now recognized that although mitochondrial sequence data have their merits for phylogeographic analyses (Avise, 2000), they may also reflect non-neutral evolutionary patterns (Ballard and Kreitman, 1995; Schizas et al., 2001). The current position of the cline may not reflect the original geographic disjunction, because ancient clines tend to be "attracted" to contemporary discontinuities in gene flow (Endler, 1977; Hare and Avise, 1996; Irwin, 2002). The nearshore oceanography appears to have changed dramatically during late Pleistocene glacial and interglacial periods (Hendy and Kennett, 2000), and the strength and geographic position of the cline could have been similarly affected.
Clinal genetic variation has been identified in only a few marine metapopulations (Gardner, 1996). Although some hybrid zones are known for intertidal and abyssal species (Hilbish, 1996; Won et al., 2003), few represent interactions among populations of the same species (e.g., Hilbish, 1996)—a set of interactions that may represent either secondary introgression of isolated lineages or primary divergence due to environmental adaptive pressures (Endler, 1982; Nielsen et al., 2003). Sotka et al. (2004) argue conservatively that if the northern and southern populations of B. glandula separated at the LGM, the rate of introgression, or "rebound," from this disruption is extremely slow considering the assumed neutral rate of gene flow for a species with broadly dispersing planktonic larvae. Here we show that this cline has been maintained for far longer, either by abiotic factors that limit the dispersal of Balanus across the coast of central California, or by natural selection that limits the spread of the historically differentiated lineages, or both.
Physical oceanographic patterns, including currents and gyres, probably play the greatest role in driving larval dispersal in B. glandula (Connolly and Roughgarden, 1999; Connolly et al., 2001). If these patterns alone maintain the distinction of northern and southern genetic races of B. glandula, the data presented here suggest that these oceanographic mechanisms would have to be stable since well before the LGM. We still know little about coastal geology and physical oceanographic processes prior to the LGM (Lyle et al., 2001; Herbert et al., 2001). Thus, we must be cautious about the precision of our estimates of Ne (in Table 1, error exceeds the estimates) or the age of the separation between these populations based on estimates of µ obtained from other taxa and without the confirmation of fossil data (Marko, 2002, Marko, 2004; Clarke and Knight, 2003; Graur and Martin, 2004). Here we simply suggest that the divergence of B. glandula populations predates the LGM, and the maintenance of this cline—perhaps by selection, since migration and advection are expected to homogenize these populations—must be considered on this longer time scale.
Since many physical oceanographic structures are believed to be unstable during glacial maxima (Herbert et al., 2001; Graham et al., 2003), the transitions between glacial and interglacial periods are likely to have interrupted or moved any such "selectively neutral" barrier to gene flow. Temperature regimes along the coast also change during climatic cycling, and species are known to extend their range northward during El Niño events (Newman and McConnaughey, 1987; Engle and Richards, 2001; Connolly and Roughgarden, 1999; but see Paine 1986). However, the actual latitudinal distribution and fitness of a species may depend on a number of abiotic variables, including coastal exposure, substrate type, food availability, and other spatial components (Sanford, 2001; Helmuth et al., 2002; Harley and Helmuth, 2003). It is not clear how old B. glandula actually is, with little fossil data available (Pleistocene, Oregon coast; Zullo, 1969, Zullo, 1984), but it has persisted through many climatic changes. Except for species in the B. amphitrite species complex (Puspasari et al., 2001), a detailed phylogeny is not available for Balanus. The peculiar nature of this cline, in which allele frequencies differ significantly but most alleles are still present on both sides of the region, adds to the complexity of this pattern. The long-term maintenance of diversity in both regions is important for our understanding of the interaction between abiotic environment and genetic diversity (e.g., Ma et al., 2000) and the extent to which historical demographics contribute to contemporary character evolution (e.g., Marchinko and Palmer, 2003; Marchinko et al., 2004). Further genetic and ecological analysis of B. glandula will be necessary to fully explain this intraspecific divergence.
|
Acknowledgments
|
|---|
The authors thank R. Toonen, N. Tsutsui, M. Hickerson, E. Sotka, T. "fujitubo" Oakley, S. Palumbi, and R. Grosberg for editorial and analytical assistance. S. Morgan, J. Abrams, C. Thornber, J. Wright, T. Shirley, A. Baldwin, T. Stokell, and D. Goldwater contributed a significant amount of taxonomic and technical effort towards this project. The editorial and technical advice of N. Knowlton and two anonymous reviewers is greatly appreciated. This research was funded by an NSF Biological Oceanography grant (DEB-96808267) to C. W. C.
|
Footnotes
|
|---|
Received 26 May 2004; accepted 29 November 2004.
|
Literature Cited
|
|---|
Avise, J. C. 2000. Phylogeography. Harvard University Press, Cambridge, MA.
Ballard, J. W. O., and M. Kreitman. 1995. Is mitochondrial DNA a strictly neutral marker? Trends Ecol. Evol. 10: 485–488.
Barnes, H., and M. Barnes. 1956. The general biology of Balanus glandula Darwin. Pac. Sci. 10: 415–421.
Beerli, P. 2004. Effect of unsampled populations on the estimation of population sizes and migration rates between sampled populations. Mol. Ecol. 13: 827–836.[Medline]
Beerli, P., and J. Felsenstein. 1999. Maximum-likelihood estimation of migration rates and effective population numbers in two populations using a coalescent approach. Genetics 152: 763–773.[Abstract/Free Full Text]
Benson, D. A., I. Karsch-Mizrachi, D. J. Lipman, J. Ostell, and D. L. Wheeler. 2004. GenBank: update. Nucleic Acids Res. 32: D23–D30.[Abstract/Free Full Text]
Clarke, T., and J. Knight. 2003. Whale genetics study leaves conservationists all at sea. Nature 424: 479.
Connolly, S. R., and J. Roughgarden. 1999. Increased recruitment of northeast Pacific barnacles during the 1997 El Niño. Limnol. Oceanogr. 44: 466–469.
Connolly, S. R., B. A. Menge, and J. Roughgarden. 2001. A latitudinal gradient in recruitment of intertidal invertebrates in the northeast Pacific Ocean. Ecology 82: 1799–1813.[ISI]
Dawson, M. N. 2001. Phylogeography in coastal marine animals: a solution from California? J. Biogeogr. 28: 723–736.
Dawson, M. N., J. L. Staton, and D. K. Jacobs. 2001. Phylogeography of the tidewater goby, Eucyclogobius newberryi (Teleostei, Gobiidae) in coastal California. Evolution 55: 1167–1179.[ISI][Medline]
Edmands, S. 2001. Phylogeography of the marine copepod Tigriopus californicus reveals substantially reduced interpopulation divergence at northern latitudes. Mol. Ecol. 10: 1743–1750.[Medline]
Endler, J. A. 1977. Geographic Variation, Speciation, and Clines. Princeton University Press, Princeton, NJ.
Endler, J. A. 1982. Problems in distinguishing historical from ecological factors in biogeography. Am. Zool. 22: 441–452.
Engle, J. M., and D. V. Richards. 2001. New and unusual marine invertebrates discovered at the California Channel Islands during the 1997–1998 El Niño. Bull. South. Calif. Acad. Sci. 100: 186–198.
Folmer, O., M. Black, W. Hoeh, R. Lutz, and R. Vrijenhoek. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotechnol. 3: 294–299.[Medline]
Gaines, S., and J. Roughgarden. 1985. Larval settlement rate: a leading determinant of structure in an ecological community of the marine intertidal zone. Proc. Natl. Acad. Sci. USA 82: 3707–3711.[Abstract/Free Full Text]
Gardner, J. P. A. 1996. The Mytilus edulis species complex in southwest England: effects of hybridization and introgression upon interlocus associations and morphometric variation. Mar. Biol. 125: 385–399.
Graham, M. H., P. K. Dayton, and J. M. Erlandson. 2003. Ice ages and ecological transitions on temperate coasts. Trends Ecol. Evol. 18: 33–40.
Graur, D., and W. Martin. 2004. Reading the entrails of chickens: molecular timescales of evolution and the illusion of precision. Trends Genet. 20: 80–86.[ISI][Medline]
Grosberg, R. K., and C. W. Cunningham. 2001. Genetic structure in the sea: from populations to communities. Pp. 61–84 in Marine Community Ecology, M. D. Bertness, M. E. Hay, and S. D. Gaines, eds. Sinauer, Sunderland, MA.
Hare, M. A., and J. C. Avise. 1996. Molecular genetic analysis of a stepped multilocus cline in the American oyster (Crassostrea virginica). Evolution 50: 2305–2315.
Harley, C. D. G., and B. S. T. Helmuth. 2003. Wave exposure, thermal stress, and absolute vs. effective shore level. Limnol. Oceanogr. 48: 1498–1508.
Hedgecock, D. 1986. Is gene flow from pelagic larval dispersal important in the adaptation and evolution of marine invertebrates? Bull. Mar. Sci. 39: 550–564.
Hellberg, M. E., D. P. Balch, and K. Roy. 2001. Climate-driven range expansion and morphological evolution in a marine gastropod. Science 292: 1707–1710.[Abstract/Free Full Text]
Helmuth, B., C. D. G. Harley, P. M. Halpin, M. O’Donnell, G. E. Hofmann, and C. A. Blanchette. 2002. Climate change and latitudinal patterns of intertidal thermal stress. Science 298: 1015–1017.[Abstract/Free Full Text]
Hendy, I. L., and J. P. Kennett. 2000. Dansgaard-Oeschger cycles and the California Current System: planktonic foraminiferal response to rapid climate change in Santa Barbara Basin, Ocean Drilling Program hole 893A. Paleoceanography 15: 30–42.
Herbert, T. D., J. D. Schuffert, D. Andreasen, L. Heusser, M. Lyle, A. Mix, A. C. Ravelo, L. D. Stott, and J. C. Herguera. 2001. Collapse of the California Current during glacial maxima linked to climate change on land. Science 293: 71–76.[Abstract/Free Full Text]
Hey, J., and R. Nielsen. 2004. Multilocus methods for estimating population sizes, migration rates, and divergence time, with applications to the divergence of Drosophila pseudoobscura and D. persimilis. Genetics 167: 747–760.[Abstract/Free Full Text]
Hickerson, M. J., and J. R. P. Ross. 2001. Post-glacial population history and genetic structure of the northern clingfish (Gobbiesox maeandricus), revealed from mtDNA analysis. Mar. Biol. 138: 407–419.
Hilbish, T. J. 1996. Population genetics of marine species: the interaction of natural selection and historically differentiated populations. J. Exp. Mar. Biol. Ecol. 200: 67–83.
Holder, K., R. Montgomerie, and V. L. Friesen. 1999. A test of the glacial refugium hypothesis using patterns of mitochondrial and nuclear DNA sequence variation in rock ptarmigan (Lagopus mutus). Evolution 53: 1936–1950.
Hudson, R. R. 1990. Gene genealogies and the coalescent process. Pp. 1–44 in Gene Genealogies and the Coalescent Process, D. Futuyma and J. Antonovics, eds. Oxford University Press, Oxford.
Hudson, R. R., and M. Turelli. 2003. Stochasticity overrules the "three-times rule": genetic drift, genetic draft, and coalescence times for nuclear loci versus mitochondrial DNA. Evolution 57: 182–190.[ISI][Medline]
Irwin, D. E. 2002. Phylogeographic breaks without geographic barriers to gene flow. Evolution 56: 2383–2394.[ISI][Medline]
Jacobs, D. K., T. A. Haney, and K. D. Louie. 2004. Genes, diversity, and geologic process on the Pacific coast. Annu. Rev. Earth Planet. Sci. 32: 601–652.[ISI]
Kulikova, T., et al. 2004. The EMBL Nucleotide Sequence Database. Nucleic Acids Res. 32: D27–D30.[Abstract/Free Full Text]
Lyle, M., L. Heusser, T. Herbert, A. Mix, and J. Barron. 2001. Interglacial theme and variations: 500 k.y. of orbital forcing and associated responses from the terrestrial and marine biosphere, U. S. Pacific Northwest. Geology 29: 1115–1118.[Abstract/Free Full Text]
Ma, X. L., D. L. Cowles, and R. L. Carter. 2000. Effect of pollution on genetic diversity in the bay mussel Mytilus galloprovincialis and the acorn barnacle Balanus glandula. Mar. Environ. Res. 50: 559–563.[ISI][Medline]
Marchinko, K. B., and A. R. Palmer. 2003. Feeding in flow extremes: dependence of cirrus form on wave-exposure in four barnacle species. Zoology 106: 127–141.
Marchinko, K. B., M. T. Nishizaki, and K. C. Burns. 2004. Community-wide character displacement in barnacles: a new perspective for past observations. Ecol. Lett. 7: 114–121.
Marko, P. B. 2002. Fossil calibration of molecular clocks and the divergence times of geminate species pairs separated by the Isthmus of Panama. Mol. Biol. Evol. 19: 2005–2021.[Abstract/Free Full Text]
Marko, P. B. 2004. ‘What’s larvae got to do with it?’ Disparate patterns of post-glacial population structure in two benthic marine gastropods with identical dispersal potential. Mol. Ecol. 13: 597–611.[Medline]
Newman, W. A., and D. P. Abbott. 1980. Cirripedia. Pp. 504–535 in Intertidal Invertebrates of California, R. H. Morris, D. P. Abbott, and E. C. Haderlie, eds. Stanford University Press, Stanford, CA.
Newman, W. A., and R. R. McConnaughey. 1987. A tropical eastern Pacific barnacle, Megabalanus coccopoma (Darwin) in Southern California, following El Niño 1982–83. Pac. Sci. 41: 31–36.
Nielsen, R., and J. Wakeley. 2001. Distinguishing migration from isolation: a Markov Chain Monte Carlo approach. Genetics 158: 885–896.[Abstract/Free Full Text]
Paine, R. T. 1986. Benthic community-water column coupling during the 1982–1983 El Niño: Are community changes at high latitudes attributable to cause or coincidence? Limnol. Oceanogr. 31: 351–360.
Palumbi, S. R., and B. D. Kessing. 1991. Population biology of the trans-Arctic exchange: mtDNA sequence similarity between Pacific and Atlantic sea urchins. Evolution 45: 1790–1805.
Peterson, M. A., and R. F. Denno. 1998. The influence of dispersal and diet breadth on patterns of genetic isolation by distance in phytophagous insects. Am. Nat. 152: 429–446.
Puspasari, I. A., T. Yamaguchi, and S. Kojima. 2001. Phylogeny of the Balanus amphitrite complex occurring in Japan (Cirripedia: Balanidae) inferred from mitochondrial COI gene nucleotide sequences and morphology. Sessile Organisms 18: 7–17.
Roughgarden, J., S. Gaines, and H. Possingham. 1988. Recruitment dynamics in complex life cycles. Science 241: 1460–1466.[Abstract/Free Full Text]
Sanford, E. 2001. The effect of variation in water temperature on the keystone predator Pisaster ochraceus. Am. Zool. 41: 1576.
Schizas, N. V., G. T. Chandler, B. C. Coull, S. L. Klosterhaus, and J. M. Quattro. 2001. Differential survival of three mitochondrial lineages of a marine benthic copepod exposed to a pesticide mixture. Environ. Sci. Technol. 35: 535–538.[Medline]
Schneider, S., J. Kueffer, D. Roessli, and L. Excoffier. 1997. Arlequin. A software for population genetic data analysis; ver. 1.1. Available: http://lgb.unige.ch/arlequin/ [Accessed November 2004.]
Slatkin, M., and B. Rannala. 1997. Estimating the age of alleles by use of intraallelic variability. Am. J. Hum. Genet. 60: 447–458.[ISI][Medline]
Sotka, E. E., J. P. Wares, J. A. Barth, R. K. Grosberg, and S. R. Palumbi. 2004. Strong genetic clines and geographical variation in gene flow in the rocky intertidal barnacle Balanus glandula. Mol. Ecol. 13: 2143–2156.[Medline]
Swofford, D. 2002. Phylogenetic Analysis Using Parsimony (PAUP) v. 4.0b10. Sinauer, Sunderland, MA.
Turner, T. F., J. P. Wares, and J. R. Gold. 2002. Genetic effective size is three orders of magnitude smaller than adult census size in an abundant, estuarine-dependent marine fish (Sciaenops ocellatus). Genetics 162: 1329–1339.[Abstract/Free Full Text]
Valentine, J. W., and D. Jablonski. 1993. Fossil communities: compositional variation at many time scales. Pp. 341–349 in Species Diversity in Ecological Communities: Historical and Geographic Perspectives, R. E. Ricklefs and D. Schluter, eds. University of Chicago Press, Chicago.
Wares, J. P. 2001. Patterns of speciation inferred from mitochondrial DNA in North American Chthamalus (Cirripedia: Balanomorpha: Chthamaloidea). Mol. Phyogenet. Evol. 18: 104–116.
Wares, J. P. 2002. Community genetics in the northwestern Atlantic intertidal. Mol. Ecol. 11: 1131–1144.[Medline]
Wares, J. P., and C. W. Cunningham. 2001. Phylogeography and historical ecology of the North Atlantic intertidal. Evolution 55: 2455–2469.[ISI][Medline]
Wares, J. P., S. D. Gaines, and C. W. Cunningham. 2001. A comparative study of asymmetric migration events across a marine biogeographic boundary. Evolution 55: 295–306.[ISI][Medline]
Wares, J. P., D. S. Goldwater, B. Y. Kong, and C. W. Cunningham. 2002. Refuting a controversial case of a human-mediated marine species introduction. Ecol. Lett. 5: 577–584
Wilkins, J. F., and J. Wakeley. 2002. The coalescent in a continuous, finite, linear population. Genetics 161: 873–888.[Abstract/Free Full Text]
Wing, S. R., L. W. Botsford, J. L. Largier, and L. E. Morgan. 1995. Spatial structure of relaxation events and crab settlement in the northern California upwelling system. Mar. Ecol. Prog. Ser. 128: 199–211.
Won, Y., C. R. Young, R. A. Lutz, and R. C. Vrijenhoek. 2003. Dispersal barriers and isolation among deep-sea mussel populations (Mytilidae: Bathymodiolus) from eastern Pacific hydrothermal vents. Mol. Ecol. 12: 169–184.[Medline]
Wright, S. 1943. Isolation by distance. Genetics 28: 139–156.[Free Full Text]
Zullo, V. A. 1969. A late Pleistocene marine invertebrate fauna from Bandon, Oregon. Proc. Cal. Acad. Sci. 12: 347–361.
Zullo, V. A. 1984. New genera and species of balanoid barnacles from the Oligocene and Miocene of North Carolina. J. Paleontol. 58: 1312–1338.[Abstract]
TOP
Abstract
Introduction
