Molecular Ecology (2005) 14, 2789–2802
doi: 10.1111/j.1365-294X.2005.02635.x
Phylogeographical disjunction in abundant high-dispersal littoral gastropods
Blackwell Publishing, Ltd.
J . M . W A T E R S ,* T . M . K I N G ,† P . M . O ’ L O U G H L I N ‡ and H . G . S P E N C E R † *Department of Zoology, University of Otago, PO Box 56, Dunedin, New Zealand, †Allan Wilson Centre for Molecular Ecology and Evolution, Department of Zoology, University of Otago, PO Box 56, Dunedin, New Zealand, ‡Marine Biology Section, Museum Victoria, GPO Box 666E, Melbourne, Victoria 3001, Australia
Abstract Phylogeographical disjunctions in high-dispersal marine taxa are variously ascribed to palaeogeographical conditions or contemporary ecological factors. Associated biogeographical studies, however, seldom incorporate the sampling design required to confidently discriminate among such competing hypotheses. In the current study, over 7800 gastropod specimens were examined for operculum colour, and 129 specimens genetically, to test ecological and historical biogeographical hypotheses relating to biogeographical disjunction in the Southern Hemisphere, and to southern Australia in particular. Mitochondrial DNA sequence analysis of the high-dispersal intertidal gastropod Nerita atramentosa in southern Australia (88 specimens; 18 localities) revealed an east–west phylogeographical split involving two highly divergent clades (26.0 ± 1.9%) exhibiting minimal geographical overlap in the southeast. The eastern clade of Nerita atramentosa is also widespread in northern New Zealand (43 specimens, 10 localities), but no significant genetic differentiation is explained by the Tasman Sea, a 2000-km-wide oceanic barrier. Spatial genetic structure was not detected within either clade, consistent with the species’ dispersive planktotrophic phase lasting for 5 –6 months. Digital analysis of operculum colouration revealed substantial differences between eastern (tan) and western (black) specimens. Genetic analysis and visual inspection of 88 Australian specimens revealed a completely nonrandom association between mtDNA data and operculum colouration. Independent examination of a further 7822 specimens from 14 sites in southern Australia revealed both colour morphs at all localities, but reinforced the phylogeographical data by indicating a marked turnover in colour morph abundance associated with a palaeogeographical barrier: Wilsons Promontory. This sharp biogeographical disjunction is in marked contrast to the species’ high dispersal abilities. The genetic similarity of Nerita morio (Easter Island) and the eastern Australian + New Zealand lineage (1.1 ± 0.3%) provides further evidence of long-distance dispersal in southern Nerita. Phylogenetic relationships of nine species (four genera) of Neritidae, an almost exclusively tropical gastropod family, are consistent with the hypothesis that southern temperate black nerites comprise a monophyletic radiation. Keywords: dispersal, mtDNA, Nerita atramentosa, Nerita melanotragus, Nerita morio, operculum colour, palaeogeography, systematics Received 26 January 2005; revision received 6 April 2005; accepted 28 April 2005
Introduction The physical and ecological forces that promote genetic differentiation among contiguous coastal populations are Correspondence: Jonathan Waters, Fax: +64 3 4797584; E-mail:
[email protected] © 2005 Blackwell Publishing Ltd
little understood (Dawson 2001). The intertidal biota of southern Australia, for instance, exhibits strong biogeographical structure along a continuous coastline (Fig. 1; Bennett & Pope 1953, 1960), but there is little consensus regarding the origin of this biodiversity. Ecological biogeographers ascribe this structure to extant factors such as habitat barriers or temperature gradients (e.g. O’Hara &
2790 J . M . W A T E R S E T A L .
Fig. 1 Evolutionary hypothesis for Australia’s warm-temperate marine biota, showing isolation of Peronian (eastern; shaded) and Flindersian (western; dark) provinces. (A) Bass Strait was a dry land bridge during Pleistocene glacial maxima when sea levels were around 150 m lower than present (dotted line). Colder temperatures restricted southern distributions of warm-temperate taxa. (B) Postglacial population expansion southward produced a zone of overlap between divergent eastern and western lineages.
Poore 2000), whereas historical biogeographers typically invoke palaeogeographical barriers existing during glacial maxima (Fig. 1; Burridge 2000). There is a similar lack of consensus on the biogeographical forces that generate intercontinental distributions. Vicariance theorists (Rosen 1978) view animal distributions as passive phenomena, with population divergence driven by plate tectonics, whereas ‘dispersalists’ ( McDowall 1978; Pole 1994) invoke transoceanic migrations. It should be noted that these differing biogeographical explanations are not always mutually exclusive; it is conceivable that both historical and ecological factors have interacted to shape extant distributions.
DNA sequence analysis has the power to reveal intraspecific genealogies (Avise et al. 1987; Avise 2000) significant for our understanding of marine palaeogeography and ecology (e.g. Benzie & Williams 1997; Chenoweth et al. 1998; Benzie 1999a, b; Dawson 2001). For instance, recent genetic analyses of southern Australian sea stars (Waters & Roy 2003; Waters et al. 2004), revealed phylogeographical patterns consistent with suggested biogeographical provinces, indicating that southeastern Victoria is an important point of biogeographical disjunction for Australia’s temperate marine taxa (Waters et al. 2004): that is, lineages common to the central and western regions of Victoria’s coast were not detected in the east. Although these data are consistent with a vicariance model, there is additional (counter-intuitive) evidence of long-distance dispersal (e.g. widespread haplotypes) in some of the taxa studied. For instance, one clade of Coscinasterias is widespread in eastern Australia and New Zealand (NZ), but absent from Western Australia (Waters & Roy 2003). Phylogeographical analyses of NZ’s intertidal biota have revealed fine-scale spatial genetic structure (Apte & Gardner 2002; Star et al. 2003: high-dispersal mollusc; Sponer & Roy 2002: brooding ophiuroid; Waters & Roy 2004a: high-dispersal asteroid) although not as marked as that detected within Australia (above). In each case, shared haplotypes found either side of a north–south NZ phylogeographical break suggest that associated oceanographic barriers are recent and/or intermittent. The superfamily Neritoidea is a diverse assemblage of largely marine gastropods, widely distributed along the world’s tropical and subtropical coasts. Relatively few species are found on temperate coastlines: just one of 20 Australian taxa has been documented from the temperate zone, Nerita atramentosa Reeve, 1855, abundant between southern Queensland (east) and Western Australia (west). This black nerite is the dominant herbivore of many southern Australian intertidal communities, and is also common in northern NZ, where it has long been known as Nerita melanotragus E.A. Smith, 1884 (Powell 1979). Temperature may determine the southern limit of N. atramentosa as it is relatively rare or absent from the cold temperate waters of southern NZ (Morton & Miller 1968; Morley 2004; > 40°S; Fig. 2) and southern Tasmania (Richmond 1990; > 43°S; Fig. 3). Morphologically similar nerites are apparently widespread in the Southern Hemisphere: a stout black species, Nerita morio (G.B. Sowerby I 1833), for example, is described from a narrow band of the eastern Pacific between Easter Island and the Austral Islands (including Pitcairn, Gambia, Rapa Islands) (Rehder 1980). Another black nerite ( Nerita lirellata Rehder, 1980), possibly a close relative of N. morio, is also described from Easter Island. The life history of N. atramentosa has been well characterized: eggs are deposited between November and January, hatch after 2 weeks, and larvae settle between May and © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
M A R I N E P H Y L O G E O G R A P H Y 2791 Fig. 2 Sampling localities for temperate Australian Nerita. Site abbreviations are from Table 1. Grey and dark shading represent the distribution of eastern and western mtDNA clades, respectively. Extensive sandy areas (Coorong, 90 Mile Beach; arrowed) are indicated by dotted lines, thin dashed lines represent state boundaries, and the thick dashed line represents the approximate southern limit of Nerita.
July (Underwood 1975). The planktotrophic veliger phase stays at sea for approximately 5 – 6 months. Sexual maturity is reached in the second summer after hatching, and the lifespan extends for up to 5 years. Adults may be found at all intertidal levels, but most commonly are found aggregating between the high- and midtide levels (Dakin & Bennett 1987; Morley 2004) with population densities up to 50– 140/m2 (Underwood 1975). In the current study we analyse mtDNA and operculum colour variation to discriminate between historic and contemporary biogeographical hypotheses for the widespread black nerite N. atramentosa. Ecological hypotheses might predict that genetic diversity in southern Australia is correlated with latitudinal variation in temperature, or habitat barriers such as 90 Mile Beach (hypothesis 1), whereas historic vicariant hypotheses predict that genetic diversity is correlated with the Bassian Isthmus, a palaeogeographical barrier (hypothesis 2; Fig. 1). Such hypotheses are testable through distributional analysis of clades, or through analysis of molecular variance (amova) of a priori sample groupings. On an intercontinental level, vicariant hypotheses predict reciprocal monophyly associated with tectonicmediated divergence of Australian and NZ populations some 80 million years ago (Ma) (hypothesis 3), whereas dispersalist hypotheses predict continental paraphyly due to recent gene flow (hypothesis 4). These conflicting hypotheses are also readily testable under the assumption of a molecular clock (Sanmartin & Ronquist 2004). Specifically, ecological biogeographical and dispersalist explanations ( hypotheses 1, 4) predict shallow divergences © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
Fig. 3 Sampling localities for New Zealand Nerita. Site abbreviations are from Table 1. The dashed line represents the approximate southern limit of Nerita.
2792 J . M . W A T E R S E T A L . associated with recent dispersal or the postglacial establishment of contemporary conditions, whereas the historical explanations (hypotheses 2, 4) predict older divergence events.
Materials and methods Molecular analyses One hundred and twenty-nine specimens of Nerita atramentosa were obtained for genetic analysis ( Figs 2 and 3; Table 1) incorporating 18 localities from southern Australia (86 specimens; collected in August–September 2003) and 10 localities from northern NZ (43 specimens; collected in August–September 2004). A number of additional Neritidae
Table 1 Geographic details of 129 Nerita samples included in DNA sequence analyses Country/region/location New Zealand Three Kings Islands North Island Cable Bay Matauri Bay Paihia Mangawhai Heads Mathesons Bay Kawakawa Bay Maketu Omaio Little Huia Australia New South Wales Mona Vale Shell Harbour Dolphin Point Guerilla Bay Tathra Merimbula Eden Victoria Mallacoota Cape Conran Port Welshpool Walkerville Skenes Creek Port Fairy Portland South Australia Robe Port Elliot Normanville Pine Point
n
Code
2
NZ-TK
5 4 5 6 5 5 2 5 4
NZ-CB NZ-MB NZ-PA NZ-MH NZ-MA NZ-KB NZ-MK NZ-OM NZ-LH
5 5 5 5 5 5 5
AU-MV AU-SH AU-DP AU-GB AU-TA AU-ME AU-ED
5 5 3 5 5 3 5
AU-MA AU-CC AU-PW AU-WK AU-SC AU-PF AU-PL
5 5 5 5
AU-RB AU-PE AU-NM AU-PP
taxa were sampled for phylogenetic comparison: the black nerite Nerita morio from Easter Island, and two specimens each of several Indo-Pacific taxa, collected from tropical northern Australia: Nerita chamaeleon Linnaeus, 1758; Nerita balteata Reeve, 1855; Nerita polita Linnaeus, 1758; Nerita undata Linnaeus, 1758; Neritina violacea Gmelin, 1791. DNA was extracted from foot tissue using a 5% Chelex solution (Walsh et al. 1991) containing 20 µg proteinase K. A 1107-bp portion of the mtDNA cytochrome oxidase I gene (COI) was amplified for most specimens using primers LCO1490 (Folmer et al. 1994) and H7005-mod1 (modified from H7005 of Folmer et al. 1994; M. Kennedy, unpublished). For the few specimens that failed to amplify with the above primers, a nested 658-bp fragment was amplified using LCO1490 and HCO2198 (Folmer et al. 1994). PCR (25 µL) conditions comprised 40 cycles (94 °C for 30 s; 45 °C for 60 s; 72 °C for 60 s) and contained 1 unit of RedHot Taq (ABGene) and 1.5 mm MgCl2. Sequencing reactions were conducted using primer LCO1490. Published COI sequences from two additional neritid genera were included as outgroups: Clithon spinosus (GenBank Accession no. AF236070; Myers et al. 2000), Theodoxus fluviatilis (AF120633; Giribet & Wheeler 2002). Distinct COI haplotypes were identified using phylogenetic software (paup*4.0b10; Swofford 1998). Pairwise sequence divergences among haplotypes were calculated using a best-fit model of sequence evolution (HKY + I + Γ) selected using modeltest 3.06 (Posada & Crandall 1998) and paup*. Bayesian phylogenetic analysis of COI data was performed using mrbayes 3.0B4 (Huelsenbeck & Ronquist 2001; Altekar et al. 2004) under the HKY + I + Γ model (‘nst = 2’; rates = ‘invgamma’) using default priors. The Markov chain Monte Carlo search was run with four chains for 106 generations, with trees sampled every 100 generations (the first 5000 trees were discarded as ‘burn-in’). Population differentiation was assessed by amova, incorporating distance information among haplotypes (Tamura–Nei + Γ; arlequin’s best available approximation of HKY + I + Γ) performed by arlequin version 2.000 (Schneider et al. 2000). Significance of resultant phi statistics was assessed by 100 000 permutations of individuals among populations, and populations among a priori geographical groupings. F-statistics (incorporating TrN + Γ distances) were calculated using arlequin, and associated probability values were calculated using 100 000 permutations. The relationship between genetic divergence and geographical distance was assessed using a Mantel (1967) test, implemented in genalex version 5.3 (Peakall & Smouse 2001) based on linearized FST values (incorporating TrN + Γ distances) vs. geographical distance among sample sites. Coastal distance calculations did not follow deep indentations such as Port Philip Bay (Victoria), as these seem unlikely to disrupt coastal dispersal of pelagic larvae. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
M A R I N E P H Y L O G E O G R A P H Y 2793
Operculum colour analyses Preliminary investigations of operculum colour indicated substantial variation (‘black’ vs. ‘tan’) among Nerita populations. This variable colouration was digitally quantified for a range of Museum Victoria Nerita collections from New Zealand and Australia (56 specimens; six localities). Photographs taken under constant light and magnification were imported into Adobe photoshop version 7 (300 dpi) and digitally analysed. Specifically, samples of 5 × 5 pixels were measured from five sites across each operculum, and colour was quantified according to hue (0 –360°colour wheel), brightness (0 –100%) and saturation (0 –100%). To ensure repeatability, we sampled similar portions across each operculum, but avoided regions that were shadowed or highly reflective. The five measurements for each operculum were averaged. Mean values for hue and colour saturation per specimen were graphed following the method of Deutsch (1997). Each specimen included in genetic analyses was examined by eye and classified as having either a tan or black operculum. An additional 7822 individuals from 14 localities (September 2004 –January 2005) were similarly classified according to operculum colour. At eight of these sites where Nerita were abundant, tight aggregations of 25 specimens were selected randomly from the upper littoral zone down to the lower littoral zone, to facilitate examination of the spatial distribution of operculum colour morphs.
Results Phylogenetic analysis The 129 Nerita specimens obtained from Australia and NZ yielded 116 haplotypes (GenBank Accession nos DQ060936– 051). Seven of these haplotypes were recorded from multiple localities, and two of these were detected in both Australia and NZ (Fig. 4). Bayesian phylogenetic analysis of COI strongly supported the monophyly of black nerites from the Southern Hemisphere (posterior probability 1.00: Nerita atramentosa + Nerita morio), and the combined monophyly of the genus Nerita (posterior probability 1.00; Fig. 4). A striking east– west phylogeographical disjunction was detected within Australia, with two strongly supported clades (posterior probability 1.00; Fig. 4) separated by a mean ML divergence of 26.0 ± 1.9% (minimum 22.8%). Divergences among haplotypes within each of these clades were comparatively small (range 0.1–2.2%; mean 1.0 ± 0.3%). The two Easter Island haplotypes of N. morio were strongly monophyletic (posterior probability 1.00; Fig. 4), genetically similar to each other (divergence 0.3%) and to eastern Australian and NZ haplotypes (mean divergence 1.1 ± 0.3%). © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
Distributional analysis of these clades revealed an apparently narrow point of disjunction in southern Victoria (AUS-PW, AUS-WK, Fig. 2). The western clade was detected only in southwestern Victoria and South Australia, whereas the eastern clade was widespread in New South Wales, eastern Victoria, and also in NZ and Easter Island. Posterior probability analysis indicated only limited phylogenetic structure within either the eastern or the western clade. Aside from the monophyly of the two Easter Island samples, there was little evidence of phylogeographical structure: for instance, most supported nodes within the eastern lineage included mixtures of Australian and New Zealand haplotypes. Similarly, we detected no evidence of phylogenetic disjunction across potential ecological barriers such as the extensive sandy regions of southern Australia (90 Mile Beach, Coorong; Fig. 2; hypothesis 1).
Spatial structure of genetic diversity Analysis of molecular variance of the eastern clade indicated that the majority of genetic variance occurred within sample sites (99.16%). No significant component of genetic diversity was attributable to continental differences (hypothesis 3): hierarchical grouping of east Australian (10 localities; 47 specimens) and NZ (10 localities; 43 specimens) accounted for only 0.56% of the observed genetic variance, a figure not greater than that which could be attributed to chance partitioning of localities among groups (P = 0.17). Similarly, variation among localities within these continental groupings was nonsignificant (0.27%; P = 0.43). Overall, differences among localities accounted for only 0.84% of genetic variance (P = 0.36). Similarly, FST analysis yielded significant values for only 5 of 190 pairwise comparisons of eastern populations, and a Mantel test did not reveal a significant relationship between genetic divergence and geographical distance among eastern Australian populations (Fig. 5; P = 0.37). We used amova of the western form to test for regional differentiation mediated by the sandy Coorong region (Fig. 2; hypothesis 1) — a potential ecological barrier in southern Australia. No significant population genetic differentiation (−2.6% of variance; P = 0.77) was detected between localities east of the Coorong (AU-WK, AU-SC, AU-PF, AU-PL, AU-RB; 22 samples) vs. samples west of the Coorong (AU-PE, AU-NM, AU-PP; 15 samples). We also detected no significant variation among samples within either group (−3.5% of variance; P = 0.62). Similar results were obtained when western N. atramentosa samples were partitioned into South Australia vs. Victoria groups (not shown). FST analysis yielded significant values for only one of 28 pairwise comparisons of western populations. A Mantel test did not reveal a significant relationship between genetic divergence and geographical distance among western populations (Fig. 5; P = 0.11).
2794 J . M . W A T E R S E T A L .
Fig. 4 Bayesian phylogenetic analysis of Nerita COI haplotypes. Outgroup genera Theodoxus, Clithon and Neritina are excluded from the tree for diagrammatic purposes. Numbers at nodes are posterior probability values > 0.50. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
M A R I N E P H Y L O G E O G R A P H Y 2795
Fig. 5 Regressions between genetic distance (linearized FST) and geographical distance for (A) the eastern lineage, and (B) the western lineage of temperate Australian Nerita.
Analysis of operculum colour Digital photographic analysis of hue vs. saturation in 56 museum specimens of Nerita revealed substantial operculum colour variation, with two major groupings of individuals (Fig. 6). These clusters represent tan (25 specimens) vs. black (31 specimens) opercula that are readily distinguishable by eye (see Fig. 7). Two of the museum collections analysed (Mallacoota, Cape Conran) contained both colour morphs (Figs 6 and 7). In addition, substantial variation in hue and saturation was detected among individuals within the black operculum cluster, with minimal overlap, for instance, between specimens from southern Victoria (Wilsons Promontory) and Western Australia (Point Peron) (Fig. 6). Relatively little variation was detected within the tan cluster, with New Zealand specimens closely resembling eastern Australian specimens (Fig. 6). Visual inspection of specimens included in genetic analyses revealed a fully nonrandom association between operculum colour and mtDNA: all specimens with eastern mtDNA (eastern Australia, NZ, Easter Island) had tancoloured opercula, whereas all with western mtDNA had black opercula. This character remained consistent with mtDNA in each of the sample sites at which both mtDNA lineages were detected (AU-WK; AU-PW). Although genetic analyses detected only minimal geographical overlap between mtDNA clades (88 Australian specimens; Fig. 2), the more intensive sampling for operculum colour (7822 specimens; Fig. 8) indicated that black and tan forms are present at all 14 field sites. Nevertheless, operculum colour dimorphism provided corroborative evidence of a major biogeographical disjunction around © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
Fig. 6 Graph of operculum colour variation among museum specimens of Nerita from southern Australia and New Zealand. Hue is represented on a circular scale (degrees) and saturation is represented on a linear scale (%) increasing from the centre.
Wilsons Promontory in Victoria. We identified a dramatic shift in their relative abundance (Fig. 8): all six sites west of Wilsons Promontory were dominated by black-operculum individuals (frequency = 0.987 ± 0.004), whereas all eight
2796 J . M . W A T E R S E T A L .
Fig. 7 Photos showing operculum colour variation among representative museum specimens of Nerita from southern Australia and New Zealand (magnification = ×2 life size). (A) Point Peron, Western Australia (3 black operculum specimens); (B) Marino Rocks, South Australia (3 black); (C) Wilsons Promontory, southern Victoria (3 black); (D) Cape Maria van Diemen, New Zealand (3 tan); (E) Mallacoota, eastern Victoria (2 tan, 1 black); (F) Cape Conran, eastern Victoria (2 tan, 1 black).
sites to the east were dominated by tan-operculum specimens (0.924 ± 0.116). The relative abundance of operculum colour morphs was remarkably consistent within each of these regions, with the exception of small samples from Waterloo Bay (immediately east of Wilsons Promontory) and Pirates Bay (southern Tasmania) which both yielded relatively even abundances of colour morphs. It should also be noted that the West Point of Cape Conran had a 10fold higher frequency of the rare black morph (0.028) than three other sites from far eastern Victoria and New South Wales (mean 0.003 ± 0.001; Fig. 8). Analysis of additional Nerita material confirmed that the eastern colour morph (tan operculum) dominates Australia’s eastern seaboard from subtropical northern New South Wales to the cool waters of eastern Tasmania. Museum Victoria records from eastern Tasmania (64 specimens from 12 localities) indicate a 15:1 ratio of eastern vs. western colour morphs (data not shown). In contrast, the western form dominates Tasmania’s northern coast (12 localities; 39 black-operculum vs. 5 tan-operculum museum specimens). Nerita individuals inhabiting the relatively dry upper-
littoral zone were typically small and found under rocks, whereas specimens in the lower zone were larger and found exposed on the surface of rocks. There was little evidence of microhabitat segregation among colour morphs in sympatry: ‘rare’ forms were typically found all over the intertidal zone, and not aggregated with respect to operculum colour. At Cape Paterson, for instance, 40 cluster samples of 25 individuals (n = 1000) yielded 16 clusters with no tan-operculum specimens, eight clusters containing one tan-operculum specimen, and one cluster containing two tan-operculum specimens (Table 2). These data are consistent with the Poisson distribution predicted for randomly distributed colour morphs.
Discussion Trans-Tasman gene flow The presence of shared haplotypes (and lack of genetic differentiation) between New Zealand and east Australian Nerita atramentosa is obviously inconsistent with a © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
M A R I N E P H Y L O G E O G R A P H Y 2797
Fig. 8 The relative abundance of Nerita operculum colour morphs across southern Australia. Black and tan colour morphs are indicated by black and grey bars, respectively, and the frequency of the black form is given for all sites.
Table 2 Co-occurrence of Nerita operculum colour morphs at eight sampling localities in southern Australia (see Fig. 8), with the frequency at which various ratios of common/uncommon colour morphs were observed. Tight groupings of 25 specimens were sampled from a number of random points throughout the intertidal zone of each site Common/uncommon colour morph ratio Locality (counts × 25)
Rare freq.
25/0
24/1
23/2
22/3
21/4
20/5
Flinders (30) Shoreham (10) Cape Paterson (40) Walkerville (40) Cape Conran (w) (30) Cape Conran (E) (20) Mallacoota (40) Eden (30)
0.016 0.012 0.010 0.014 0.028 0.004 0.003 0.003
18 7 31 29 20 18 37 28
12 3 8 8 4 2 3 2
0 0 1 3 3 0 0 0
0 0 0 0 2 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 1 0 0 0
Gondwanan (vicariant) explanation for their wide distribution (hypothesis 3). Clearly, the Tasman Sea is not a significant phylogeographical barrier for eastern N. atramentosa (hypothesis 4), likely reflecting their extended (5– 6 month) planktotrophic larval phase (Underwood 1975). This lack of trans-Tasman genetic differentiation is © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
in keeping with data for other southern marine species with long-lived larvae. For instance, the southern rock lobster, Jasus edwardsii, lacks substantial genetic divergence across the Tasman Sea (Brasher et al. 1992; Ovenden et al. 1992; Booth & Ovenden 2000). Jasus has a plankotrophic larval phase that extends for up to 2 years, and oceanographic
2798 J . M . W A T E R S E T A L . eastward movement of planktonic larvae (Chiswell et al. 2003), but little evidence of westward movement.
Disjunction in southern Australia
Fig. 9 Projected transit times for planktotrophic larvae traveling between eastern Australia and NZ (after Chiswell et al. 2003). Simulated larval trajectories are based on mean current flows, and results are similar for contrasting wind regimes. Estimated median transit time to NZ is approximately 800 days for larvae originating from New South Wales (A), and approximately 1000 days for larvae from Bass Strait (B).
modelling suggests that this is ample time for trans-Tasman dispersal (Chiswell et al. 2003; Fig. 9). Similarly, a lack of genetic divergence across the Tasman Sea is seen in cheilodactylid fishes (Grewe et al. 1994; Burridge & Smolenski 2003), which have a pelagic larval phase lasting up to one year. On the basis of Chiswell et al. ‘s (2003) study, the 5 - to 6-month planktonic phase of N. atramentosa would provide a low probability of successful transit to NZ (perhaps around 6% for larvae originating in New South Wales; 0.2% for larvae originating in Bass Strait; Fig. 9). Nevertheless, the high abundance and fecundity of the species implies that very low percentages of successful dispersal will still confer a considerable level of gene flow. Traditionally, most dispersalist hypotheses for the southern temperate region have assumed that west–east dispersal predominates (McDowall 1978; Fleming 1979; Pole 1994; Waters et al. 2000). It has recently been suggested, however, that east–west dispersal from NZ (e.g. Winkworth et al. 2002) may also represent a significant evolutionary force for the region. While our genetic data for the eastern lineage are unable to discriminate between eastward and westward gene flow, the presence of substantial diversity in Australian N. atramentosa (two deep lineages) vs. limited diversity in NZ (one lineage) is consistent with eastward gene flow. This phylogenetic approach to biogeographical inference has been used elsewhere to support eastward oceanic dispersal in a widespread southern temperate sea star (Waters & Roy 2004b). In the case of trans-Tasman dispersal, satellite tracking of drifting buoys provides strong support for the passive
Given the lack of genetic differentiation across a 2000-km oceanic barrier (the Tasman Sea separating Australia and NZ), it may seem remarkable that there is a major biogeographical disjunction within Australian N. atramentosa. We have already established that Nerita has high dispersal ability: neither eastern nor western mtDNA clade shows evidence of population differentiation, even across substantial ecological barriers such as the Coorong. Likewise, the Easter Island taxon Nerita morio, some 5000 km to the east of NZ, comprises haplotypes that are genetically very similar to Australian and NZ haploptypes of N. atramentosa, providing further evidence of recent long-distance dispersal. Similarly, the young geological age of Easter Island (around 3 Ma) implies recent colonization. The sharp biogeographical disjunction detected in southern Australian N. atramentosa is unrelated to the Coorong and 90 Mile Beach (extensive regions devoid of rocky habitat) and is not correlated with latitudinal variation in temperature (hypothesis 1). Indeed, based on museum specimens, the eastern lineage dominates the east coast from the warm waters of northern New South Wales to the cool waters of eastern Tasmania (Museum Victoria material). The western lineage is similarly dominant from the warm temperate waters of Western Australia (98 specimens, six sites: Museum Victoria material) to the cool waters of northern Tasmania. The transition in relative abundance of eastern and western lineages of N. atramentosa is correlated with a palaeogeographical barrier in the Bass Strait region (hypothesis 2), as predicted by our evolutionary model based on Australia’s glacial history (Fig. 1; Waters et al. 2004). The precise point of disjunction — Wilsons Promontory — represents the northern extension of a granite ridge that was not breached until sea levels exceeded −50 m around 10 ka (Davies 1974). This isthmus was exposed on numerous occasions for extensive periods of the last 3 Myr. We therefore infer that N. atramentosa retains the phylogeographical signature of palaeogeographical history, despite the high dispersal ability of this species. We also suggest that Australia’s distinct eastern (East Australian Current, EAC) and western (Leeuwin Current, LC; see O’Hara & Poore 2000) currents have helped to maintain this biogeographical disjunction. While the south-flowing EAC links populations on Australia’s eastern coast, satellite data provide little evidence of westward movement of this water mass into Bass Strait (Chiswell et al. 2003). The LC, by contrast, flows eastwards from Western Australia to Bass Strait. We presume that the EAC and LC, respectively, facilitated postglacial population expansion of the isolated eastern (southward © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
M A R I N E P H Y L O G E O G R A P H Y 2799 expansion) and western (southeastward expansion) lineages of Nerita (Fig. 1). Growing evidence suggests that Wilsons Promontory has considerable biogeographical significance for Australia’s southern temperate region. The sharp phylogeographical disjunction we observed across this location is consistent with data for other species pairs of coastal taxa: the two subspecies of Patelloida latistrigata, for example, apparently have a similar east–west disjunction (Dartnall 1974). Additionally, the sea star Patiriella medius is replaced by Patiriella oriens to the east of the promontory (O’Loughlin et al. 2003). More broadly, O’Hara & Poore (2000) noted that the eastern coast between Wilsons Promontory and New South Wales houses echinoderm and decapod assemblages distinct from those of western Victoria and South Australia. The clear biogeographical disjunction reported here in southeast Australia adds to a growing list of important marine phylogeographical barriers: Atlantic vs. Gulf coasts of Florida (see Avise 2000); Indian vs. Pacific Ocean coasts of northern Australia (see Benzie 1999a, b); Indian vs. Atlantic Ocean coasts of South Africa (Lessios et al. 2003). Future multispecies phylogeographical research should help to test the biological generality of this newly documented point of southern temperate disjunction.
Operculum colour: ecological variation? We detected substantial and consistent differences in operculum colouration between eastern and western specimens of Nerita (Figs 6 and 7). A similar east–west (tan vs. black) colour morph disjunction was detected among specimens sampled for genetic analyses (Fig. 2) and in specimens examined in the wild (Fig. 8). The relatively minor divergence in colour detected among museum collections of the western black form (Fig. 6) might reflect population differentiation within this lineage, but could alternatively be attributable to variable collection dates and preservation histories. The black and tan operculum colour morphs of N. atramentosa occupy similar habitats in eastern and southern Australia. There is no evidence that the associated biogeographical disjunction is related to ecological variation (see above). Similarly, we detected no evidence of microhabitat segregation among the colour morphs in sites where they co-occur. Our ability to detect this, however, may have been limited by the scarcity of one or other colour morph at most field sites (Fig. 8; Table 2). Nevertheless, our results confirm that the two forms are ecologically similar and regularly cluster together. This contrasts with results from rocky shores of tropical Australia, where four sympatric Nerita taxa [Nerita chamaeleon (generalist), Nerita costata (vertical surfaces), Nerita polita (boulder bases), Nerita undata (crevices)] occur in monospecific aggregations that show substantial interspecific differences in microhabitat © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
usage (Neil 2001). Such ecological segregation is typically thought to reflect competition. To date, only the eastern lineage of N. atramentosa (New South Wales) has been the subject of detailed ecological studies (Underwood 1974, 1975). As both forms lack significant spatial genetic structure, however, the western lineage may have a comparable life history. Alternatively, the wider biogeographical distribution of the eastern lineage (incorporating much of the South Pacific) could be taken as evidence that it has the higher dispersal ability of the two lineages. Obviously, future ecological studies are required to assess the larval duration of the western lineage and shed light on its dispersal ability. Little is known about the genetic basis of shell colour polymorphism in gastropods, and essentially nothing about genes likely to control operculum colouration. Genetic studies of the landsnail Cepaea nemoralis indicate that distinct shell colour morphs are controlled by an allelic series at a single nuclear locus, with dark shades dominant to lighter forms (Cain et al. 1960). Variable banding patterns in the same species are controlled by a second, tightly linked locus. It has been argued that selective forces associated with predation and environmental heterogeneity may help to maintain shell-colour polymorphism in Cepaea (Cook 1998). It has also been suggested that variable salinity tolerance may influence the relative abundances of colour morphs of the intertidal periwinkle (Littorina saxatilis) (Sokolova & Berger 2000). In the current study, however, we found no evidence that the relative abundances of tan vs. black opercula were related to ecological variation in Nerita; rather, we suggest the forms represent fixed differences between two distinct evolutionary lineages. Given the completely nonrandom association between mtDNA and operculum colour (the latter presumably reflecting variation at one or more nuclear loci), including two sites of sympatry, it is possible that the eastern and western lineages of Nerita are reproductively isolated. Moreover, it seems possible that the low population density of rare colour morphs (frequencies typically ≤ 0.02; Fig. 8) could give rise to an Allee effect that reduces their reproductive success, which would also help to maintain the biogeographical disjunction. Future studies incorporating mtDNA, nuclear loci, and colour dimorphism will test for hybridization among these divergent lineages. We will argue elsewhere that these genetically and phenotypically divergent forms of N. atramentosa represent taxonomically distinct species.
Evolution of southern temperate nerites Neritomorphs are an ancient and diverse gastropod assemblage first appearing in Silurian–Devonian strata (c. 400 Ma; Fryda & Blodgett 2001; Colgan et al. 2003). Fossils directly attributable to Neritidae date back to the Triassic
2800 J . M . W A T E R S E T A L . (Knight et al. 1960). Nerita atramentosa, as it is currently understood, is represented in the Australian fossil record from the early Pliocene of Victoria (Beesley et al. 1998) and the late Pleistocene of South Australia (Ludbrook 1984). The single NZ record (under the name N. melanotragus E.A. Smith, 1884) is from the early Pliocene (3.6 –5.0 Ma; Laws 1936; Beu & Maxwell 1990). Under a standard molecular calibration for gastropods (2.4% per Myr; Hellberg & Vacquier 1999), the Kimura 2parameter (K2P; Kimura 1980) divergence between eastern and western lineages of N. atramentosa (minimum 12.8%; mean 13.7%; note that divergences are considerably higher under the parameter-rich HKY + I + Γ model) suggests a phylogenetic split approximately 5 – 6 Ma. This estimate predates the late Pliocene (around 3 Ma) initiation of glaciation, an event hypothesized to have facilitated allopatric divergence of these lineages. We therefore suggest that a faster divergence rate may be more appropriate for N. atramentosa. Alternatively, if the Pliocene estimate is correct, this suggests the cold-water barrier of southern Australia in the mid-Pliocene was sufficiently strong to promote divergence prior to the initiation of glaciation around 3 Ma. Our data for Nerita are consistent with a single southern radiation within this largely tropical genus. The K2P divergence (minimum 13.2%; mean 15.1%) between temperate (N. atramentosa, N. morio) and tropical (N. undata) sister groups suggests the associated divergence occurred in the early Pliocene (roughly 5 – 6 Ma), only shortly before the divergence of eastern and western temperate lineages (see above). The early Pliocene appearance of fossil N. atramentosa is also consistent with this molecular time frame. It has often been suggested that Australia’s ongoing tectonic movement northward has helped it ‘acquire’ a large number of tropical Indo-Pacific marine taxa (O’Hara & Poore 2000). Similarly, Fleming (1979) suggested that Australasia is an important entry point for tropical taxa into the southern temperate region. Improved sampling of tropical Nerita species will shed additional light on these hypotheses.
Acknowledgements J. Eason, M. Finney, A. Smith, and M. Stewart assisted with the collection of temperate Nerita specimens. M. do ceo Fernandes Soares, L. Devlin, N. Choochinprakarn, S. Choochinprakarn, K. Prasitmonthon, Y. Techatanadirek, and A. Waters assisted with field observations in southern Australia. B. Marshall (Te Papa Museum of New Zealand) provided samples from Three Kings Islands, and R. Willan, M. Chaddock, and G. Dally (Northern Territory Museum and Art Gallery) provided tropical taxa from northern Australia. K. Miller and C. Rowley kindly assisted with photograph production. The manuscript was improved by comments from C. Burridge, R. Harrison, E. Taylor, and two anonymous reviewers. A University of Otago Research Grant and Marsden Contract UOO914 (Royal Society of New Zealand) provided financial support for the research.
References Altekar G, Dwarkadas S, Huelsenbeck JP, Ronquist F (2004) Parallel metropolis coupled Markov chain Monte Carlo for Bayesian phylogenetic inference. Bioinformatics, 20, 407–415. Apte S, Gardner JPA (2002) Population genetic subdivision in the New Zealand greenshell mussel (Perna canaliculus) inferred from single-strand conformation polymorphism analysis of mitochondrial DNA. Molecular Ecology, 11, 1617–1628. Avise JC (2000) Phylogeography: The History and Formation of Species. Harvard University Press, Cambridge, Massachusetts. Avise JC, Arnold J, Ball RM et al. (1987) Intraspecific phylogeography: the mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systematics, 18, 489–522. Beesley PL, Ross GJB, Wells A (1998) Mollusca: The Southern Synthesis. Fauna of Australia, vol. 5. CSIRO Publishing, Melbourne. Bennett I, Pope EC (1953) Intertidal zonation of the exposed rocky shores of Victoria, together with a rearrangement of the biogeographical provinces of temperate Australian shores. Australian Journal of Marine and Freshwater Research, 4, 105–159. Bennett I, Pope EC (1960) Intertidal zonation of the exposed rocky shores of Tasmania and its relationship with the rest of Australia. Australian Journal of Marine and Freshwater Research, 11, 182–221. Benzie JAH (1999a) Genetic structure of coral reef organisms: ghosts of dispersal past. American Zoologist, 39, 131–145. Benzie JAH (1999b) Major genetic differences between crown-ofthorns starfish (Acanthaster planci) populations in the Indian and Pacific Oceans. Evolution, 53, 1782–1795. Benzie JAH, Williams ST (1997) Genetic structure of giant clam (Tridacna maxima) populations in the west Pacific is not consistent with dispersal by present-day ocean currents. Evolution, 51, 768–783. Beu AG, Maxwell PA (1990) Cenozoic Mollusca of New Zealand. New Zealand Geological Survey Paleontological Bulletin 58. Lower Hutt, New Zealand. Booth JD, Ovenden JR (2000) Distribution of Jasus spp. (Decapoda: Palinuridae) phyllosomas in southern waters: implications for larval recruitment. Marine Ecology Progress Series, 200, 241– 255. Brasher DJ, Ovenden JR, White RWG (1992) Mitochondrial DNA variation and phylogenetic relationships of Jasus spp. ( Decapoda, Palinuridae). Journal of Zoology, London, 227, 1–16. Burridge CP (2000) Biogeographic history of geminate cirrhitoids (Perciformes: Cirrhitoidea) with east–west allopatric distributions across southern Australia, based on molecular data. Global Ecology and Biogeography Letters, 9, 517–525. Burridge CP, Smolenski AJ (2003) Lack of genetic divergence found with microsatellite DNA markers in the tarakihi Nemadactylus macropterus. New Zealand Journal of Marine and Freshwater Research, 37, 223–230. Cain AJ, King JM, Sheppard PM (1960) New data on the genetics of polymorphism in the snail Cepaea nemoralis. Genetics, 45, 393 – 411. Chenoweth SF, Hughes JM, Keenan CP, Lavery S (1998) Where oceans meet: a teleost shows secondary integration at an Indian– Pacific interface. Proceedings of the Royal Society of London. Series B, Biological Sciences, 265, 415–420. Chiswell SM, Wilkin J, Booth JD, Stanton B (2003) Trans-Tasman sea-larval transport: is Australia a source for New Zealand rock lobsters? Marine Ecology Progress Series, 247, 173–182. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
M A R I N E P H Y L O G E O G R A P H Y 2801 Colgan DJ, Ponder WF, Beacham E, Macaranas JM (2003) Gastropod phylogeny based on six segments from four genes representing coding or non-coding and mitochondrial or nuclear DNA. Molluscan Research, 23, 123 –148. Cook LM (1998) A two-stage model for Cepaea polymorphism. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 353, 1577–1593. Dakin WJ, Bennett I (1987) Australian Seashores. Angus and Robertson, North Ryde, Australia. Dartnall AJ (1974) Littoral biogeography. In: Biogeography and Ecology in Tasmania (ed. Williams WD), pp. 171–194. Dr. W. Junk, The Hague, The Netherlands. Davies JL (1974) Geomorphology and Quaternary environments. In: Biogeography and Ecology in Tasmania (ed. Williams WD), pp. 17–27. Dr. W. Junk, The Hague, The Netherlands. Dawson MN (2001) Phylogeography in coastal marine animals: a solution from California? Journal of Biogeography, 28, 723–736. Deutsch JC (1997) Colour diversification in Malawi cichlids: evidence for adaptation, reinforcement or sexual selection? Biological Journal of the Linnean Society, 62, 1–14. Fleming CA (1979) The Geological History of New Zealand and Its Life. Auckland University Press, Auckland, New Zealand. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3, 294 –299. Fryda J, Blodgett RB (2001) The oldest known heterobranch gastropod, Kusokwimia gen. nov., from the Early Devonian of west-central Alaska, with notes on the early phylogeny of higher gastropods. Vestnik Ceského Geologickeho Ústavu, 76, 39 –54. Giribet G, Wheeler GC (2002) On bivalve phylogeny: a high-level analysis of the Bivalvia (Mollusca) based on combined morphology and DNA sequence data. Invertebrate Biology, 121, 271–324. Grewe PM, Smolenski AJ, Ward RD (1994) Mitochondrial DNA diversity in jackass morwong (Nemadactylus macropterus, Teleostei) from Australian and New Zealand waters. Canadian Journal of Fisheries and Aquatic Sciences, 51, 1101–1109. Hellberg ME, Vacquier VD (1999) Rapid evolution of fertilization selectivity and lysine cDNA sequences in teguline gastropods. Molecular Biology and Evolution, 16, 839 – 848. Huelsenbeck JP, Ronquist F (2001) mrbayes: Bayesian inference of phylogeny. Bioinformatics, 17, 754 –755. Kimura M (1980) A simple method for estimating evolutionary rate of base substitutions through comparative studies of nucleotide sequences. Journal of Molecular Evolution, 16, 111–120. Knight JB, Cox LR, Keen AM, Batten RL, Yochleson EL, Robertson R (1960) Systematic descriptions (Archaeogastropoda). In: Treatise on Invertebrate Paleontology, Part 1, Mollusca 1 (ed. Moore RC). Geological Society of America and University of Kansas Press, Lawrence, Kansas. Laws CR (1936) The Waitotaran faunule at Kaawa Creek, part 2. Transactions of the Royal Society of New Zealand, 66, 99 –124. Lessios HA, Kane J, Robertson DR (2003) Phylogeography of the pantropical sea urchin Tripneustes: contrasting patterns of population structure between oceans. Evolution, 57, 2026–2036. Ludbrook NH (1984) Quaternary molluscs of South Australia. Department of Mines and Energy, South Australia Handbook, 9, 1–327. Mantel N (1967) The detection of disease clustering and a generalized regression approach. Cancer Research, 27, 209 –220. McDowall RM (1978) Generalised tracks and dispersal in biogeography. Systematic Zoology, 27, 88 –104. © 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
Morley MS (2004) A Photographic Guide to Seashells of New Zealand. New Holland Publishers, Auckland, New Zealand. Morton J, Miller M (1968) The New Zealand Sea Shore. Collins, London, United Kingdom. Myers MJ, Meyer CP, Resh VH (2000) Neritid and thiarid gastropods from French Polynesian streams: how reproduction (sexual, parthenogenetic) and dispersal (active, passive) affect population structure. Freshwater Biology, 4, 535–545. Neil KM (2001) Microhabitat segregation of co-existing gastropod species. Veliger, 44, 294–300. O’Hara TD, Poore GCB (2000) Patterns of distribution for southern Australian marine echinoderms and decapods. Journal of Biogeography, 27, 1321–1335. O’Loughlin PM, Waters JM, Roy MS (2003) A molecular and morphological review of the asterinid, Patiriella gunnii (Gray) (Echinodermata: Asteroidea). Journal of the Royal Society of New Zealand, 60, 181–195. Ovenden JR, Brasher DJ, White RWG (1992) Mitochondrial DNA analyses of the red rock lobster Jasus edwardsii supports an apparent absence of population subdivision throughout Australasia. Marine Biology, 112, 319–326. Peakall R, Smouse PE (2001) GENALEX V5: Genetic analysis in Excel. Population genetic software for teaching and research. Australian National University, Canberra, Australia. Pole M (1994) The New Zealand flora — entirely long-distance dispersal? Journal of Biogeography, 21, 625–635. Posada D, Crandall KA (1998) modeltest: testing the model of DNA substitution. Bioinformatics, 14, 817–818. Powell AWB (1979) New Zealand Mollusca: Marine, Land and Freshwater. Collins, Auckland. Rehder HA (1980) The marine mollusks of Easter Island (Isla de Pascua) and Sala y Gómez. Smithsonian Contributions to Zoology Number 289. Smithsonian Institution, Washington DC. Richmond MH (1990) Tasmanian Sea Shells Common to Other Australian States. Richmond Printers, Tasmania, Australia. Rosen DE (1978) Vicariant patterns and historical explanation in biogeography. Systematic Zoology, 27, 159–188. Sanmartin I, Ronquist F (2004) Southern Hemisphere biogeography inferred by event-based models: plant versus animal patterns. Systematic Biology, 53, 216–243. Schneider S, Kueffer JM, Roessli D, Excoffier L (2000) ARLEQUIN, Version 2.000: a software for population genetic data analysis. Genetics and Biometry Laboratory, Department of Anthropology, University of Geneva, Switzerland. Sokolova IM, Berger VJ (2000) Physiological variation related to shell colour polymorphism in White Sea Littorina saxatilis. Journal of Experimental Marine Biology and Ecology, 245, 1–23. Sponer R, Roy MS (2002) Phylogeographic analysis of the brooding brittle star Amphipholis squamata (Echinodermata) along the coast of New Zealand reveals high cryptic genetic variation and cryptic dispersal potential. Evolution, 56, 1954–1967. Star B, Apte S, Gardner JPA (2003) Genetic structuring among populations of the greenshell mussel Perna canaliculus revealed by analysis of randomly amplified polymorphic DNA. Marine Ecology Progress Series, 249, 171–182. Swofford DL (1998) PAUP*: Phylogenetic Analysis Using Parsimony (*and Other Methods) 4.0b10. Sinauer Associates, Sunderland, Massachusetts. Underwood AJ (1974) The reproductive cycles of and geographical distribution of some common eastern Australian prosobranchs (Mollusca: Gastropoda). Australian Journal of Marine and Freshwater Research, 25, 63–88.
2802 J . M . W A T E R S E T A L . Underwood AJ (1975) Comparative studies on the biology of Nerita atramentosa Reeve, Bembicium nanum (Lamarck) and Cellana tramoserica (Sowerby) (Gastropoda: Prosobranchia) in S.E. Australia. Journal of Experimental Marine Biology and Ecology, 18, 153–172. Walsh PS, Metzger DA, Higuchi R (1991) Chelex 100 as a medium for simple extraction of DNA for PCR-based typing of forensic material. BioTechniques, 10, 506 –513. Waters JM, Roy MS (2003) Marine biogeography of southern Australia: phylogeographical structure in a temperate sea-star. Journal of Biogeography, 30, 1787–1796. Waters JM, Roy MS (2004a) Phylogeography of a high-dispersal New Zealand sea-star: does upwelling block gene-flow? Molecular Ecology, 13, 2797–2806. Waters JM, Roy MS (2004b) Out of Africa: the slow train to Australasia. Systematic Biology, 53, 18 –24.
Waters JM, Dijkstra LH, Wallis GP (2000) Biogeography of a Southern Hemisphere freshwater fish: how important is marine dispersal? Molecular Ecology, 9, 1815–1821. Waters JM, O’Loughlin PM, Roy MS (2004) Cladogenesis in a starfish species complex from southern Australia: evidence for vicariant speciation? Molecular Phylogenetics and Evolution, 32, 236–245. Winkworth RC, Wagstaff S, Glenny D, Lockhart PJ (2002) Plant dispersal N.E.W.S. from New Zealand. Trends in Ecology & Evolution, 17, 514–520.
Jon Waters’s research program focuses on the biogeography and evolution of Southern Hemisphere marine and freshwater biota. Hamish Spencer has ongoing research interests in molluscan systematics and evolutionary genetics.
© 2005 Blackwell Publishing Ltd, Molecular Ecology, 14, 2789–2802
