Large scale patterns and trophic structure of southern African rocky shores: the roles of geographic variation and wave exposure

In this study we revise the biogeographic delimitation, and large-scale patterns of community structure of the intertidal rocky shores of southern Africa. We use binary (presence/absence) and per-species biomass data collected at fifteen localities and thirty-seven different rocky sites, encompassing the shores of southern Namibia, South Africa and southern Mozambique. Multivariate analyses revealed that the shores of southern Africa (south of 25°) can be divided into three main biogeographic provinces: the west coast or Namaqua province, the south coast or Agulhas province and the east coast or Natal province. The biomass structure of the intertidal rocky shores communities of southern Africa varied at a large scale, corresponding to biogeographic differences, while local-scale variation accorded with the intensity of local wave action. The average biomass of west coast communities was on average significantly greater than that of the south and east provinces. At a local scale, the community biomass on exposed shores was an order of magnitude greater than on sheltered shores, within all biogeographic provinces. Semi-exposed shores exhibited intermediate average biomass. The trophic structure of these communities varied significantly with wave action: autotrophs, filter-feeders and invertebrate predators were more prevalent on wave exposed than sheltered shores, whereas grazers were more abundant on sheltered and semi-exposed shores. Exposed shores were consistently dominated by far fewer species than semi-exposed and sheltered shores, independently of biogeographic differences. Within all biogeographic provinces semi-exposed and sheltered shores were more diverse than exposed shores. West coast intertidal communities therefore had high levels of biomass, but were consistently species-poor. Several working hypotheses that could explain these large and small-scale patterns are presented.     

The Lower Mekong: an insurmountable barrier to amphibians in southern Indochina?

Geographical information system (GIS)‐based cluster analyses and comparative ecological approaches were applied to unravel the hitherto controversially discussed role of the Mekong River as a biogeographical barrier to the amphibian fauna in Indochina. Referring to the unique and independent history of the Lower Mekong, south of the Lao–Cambodian border, this approach focused on the amphibians of southern Indochina and adjacent parts of eastern Thailand. To date, 131 amphibian species have been recorded from this region. In concordance with previous studies, our GIS‐based analyses revealed that the amphibian fauna of the Cardamom Mountains, west of the Mekong, possessed no closer affinity to the herpetofauna of the Southern Annamites. In addition, a statistical comparison of ecological characters showed an occurrence of ecological differentiation among amphibians across the Lower Mekong. Our results suggest that the Lower Mekong serves as, or at least coincides with, a biogeographical barrier in southern Indochina. Here, we discuss this hypothesis, together with selected species spatial distribution patterns, their species‐specific ecological traits and the palaeogeographical background of the region. © 2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 114 , 905–914.     

Using the biogeographical distribution and diversity of seaweed species to test the efficacy of marine protected areas in the warm-temperate Agulhas Marine Province, South Africa

Aim To study the siting of marine protected areas (MPAs) with respect to the biogeographical distribution of seaweeds within the Agulhas Marine Province and to assess the effectiveness of current MPAs in including (conserving) seaweeds of the South African south coast. Location South Africa – the south coast between Cape Agulhas and the Eastern Cape/Kwazulu‐Natal border, and eight MPAs within that area. Methods We used interpolated seaweed distribution records from all available sources, in 50‐km coastal sections. Cluster analysis (Jaccard Average Linkage) of species presence/absence data provided measures of similarity between coastal sections and between MPAs. Complementarity analyses identified the sequence of ‘importance’ of sections/MPAs for conserving seaweed species. Results Species presence/absence data indicated two main groups, representing western (cooler water) and eastern (warmer water) biogeographical divisions, as well as several biogeographical subdivisions within each of these groups. Complementarity analysis yielded a sequence of ‘importance’ of coastal sections (in terms of the highest number of species included) that began with a section just east of central in the Agulhas Marine Province, around Port Alfred, where there is no MPA. This was followed by the easternmost section (warmest water), which contains the Pondoland MPA, and then by the westernmost (coolest water) section, containing the De Hoop MPA. Similar analysis of the actual species collected in MPAs showed a generally similar pattern. Main conclusions Seven current MPAs and one proposed coastal MPA in the Agulhas Marine Province appear to be well distributed and well sited to include (conserve) the full biogeographical range of seaweeds. However, if further MPAs are to be considered, the Port Alfred area is recommended for improved conservation. This study did not examine estuaries, which may require improved conservation efforts. Seaweed distribution data, which are often relatively complete, offer a good tool for planning the siting of coastal MPAs.     

Genetic structure of the common impala (Aepyceros melampus melampus) in South Africa: phylogeography and implications for conservation

We analysed 131 common impala (Aepyceros melampus melampus) samples from two provinces in South Africa (Limpopo and KwaZulu‐Natal) that are separated by the Drakensberg Mountain Range using sequences of the mitochondrial control region and seven polymorphic nuclear microsatellite loci. In line with earlier studies on bovid species, we found very high values of genetic diversity, particularly at the mtDNA locus with an overall nucleotide diversity of 3.6% and an overall haplotype diversity of 0.98. All statistical approaches confirmed a significant population differentiation between Limpopo and KwaZulu‐Natal, suggesting that areas of unsuitable habitat caused by the presence of the Drakensberg Range and the Indian Ocean coastal belt act as a barrier to gene flow. Only few individuals with signs of admixed origin were indicative of translocations or rare migration between the two provinces. Combination of our mtDNA data set with those of previous studies on impala from south‐western, southern and eastern Africa revealed the highest diversity in South Africa. This is in line with the hypothesis of a southern glacial refuge from which various African ungulate species spread northeast during the Holocene, although in the case of impala further analyses based on larger data sets will be necessary to definitively settle this question.     

Marine biogeography of southern Australia: phylogeographical structure in a temperate sea-star

Aim To test whether marine biogeographical patterns observed at the community level are also important within species. It is postulated that historical hydrogeographic barriers have driven in situ diversification. Location The intertidal and shallow subtidal zones of southern Australia, New Zealand and nearby islands. Australia's temperate marine communities are characterized by a high degree of endemism and show strong biogeographical structure along an east–west axis. Methods Phylogeographical analysis of the widespread asteriid sea‐star Coscinasterias muricata Verrill across southern Australia and New Zealand. Forty‐two samples from 27 locations were included in phylogenetic analyses of mitochondrial (CO1; control region) and nuclear (ITS2) DNA sequences. Results Analysis of mtDNA revealed a deep phylogenetic split within Australian C. muricata, strongly correlated with latitude. ‘Northern’ haplotypes (latitude ≤ 37.6° S, nine sites, 15 samples) were 7.3–9.4% divergent from ‘southern’ haplotypes (latitude ≥ 37.6° S, 19 sites, 27 samples), consistent with late Pliocene separation. Eastern and western representatives of the ‘northern’ clade were 0.5–1.0% divergent, probably reflecting Pleistocene isolation. The ‘southern’ clade of Australia is also represented in New Zealand, indicating Pleistocene oceanic dispersal. Nuclear DNA (ITS2) sequences yielded relatively little phylogenetic resolution, but were generally congruent with mtDNA‐based groupings. Main conclusions The phylogeographical pattern detected within Australian C. muricata closely resembles marine biogeographical groupings proposed on the basis of community and species distributions. Recurring evolutionary patterns may have been driven by the hydrographic history of southern Australia. Specifically, we suggest that Plio‐Pleistocene temperature change and the repeated opening and closure of Bass Strait promoted allopatric divergence and perhaps cryptic speciation in C. muricata.     

Biogeographical structure and affinities of the marine demersal ichthyofauna of Australia

Aim To investigate the biogeographical structure and affinities of the Australian marine demersal ichthyofauna at the scale of provinces and bathomes for the purposes of regional marine planning. Location Australia. Methods Patterns of distribution in the Australian fish fauna, at both intra‐regional and global scales, were examined using a science‐based, management framework dividing Australia’s marine biodiversity into 16 province‐level biogeographical units. Occurrences of 3734 species in eight depth‐stratified bathomes (from the coast to the mid‐continental slope) within each province were analysed to determine the structure and local affinities of their assemblages and their association with faunas of nearby regions and oceans basins. Results Strong geographic and depth‐related structure was evident. Fish assemblages in each province, and in each bathome of each province, were distinct, with the shelf‐break bathome more similar to the adjacent continental shelf bathome than to the upper slope bathome. Data based only on endemic species performed well as a surrogate of the entire dataset, yielding comparable patterns of similarity between provinces and bathomes. Tropical and temperate elements were better discriminated than elements of the Pacific and Indian oceans, with the central western province more similar to the tropical provinces (including those in the east), and the eastern province closer to southern temperate provinces. The fauna shares the closest regional affinities with those of the adjacent south‐west Pacific, western Pacific Rim, and elements of wide‐ranging Indo‐Pacific components. Elements unique to the Pacific and Indian oceans are poorly represented. Main conclusions The complex nature of Australia’s marine ichthyofauna is confirmed. A hierarchy of provinces and bathomes, used to ensure that Australia’s developing marine reserve network is both representative and comprehensive, is equally robust when based on all known Australian fish species or on only those species endemic to this continent. Latitude and depth are more important than oceanic influences on the composition of this fauna at these scales.     

‘After Africa’: the evolutionary history and systematics of the genus Charaxes Ochsenheimer (Lepidoptera: Nymphalidae) in the Indo‐Pacific region

The predominantly Afrotropical genus Charaxes is represented by 31 known species outside of Africa (excluding subgenus Polyura Billberg). We explored the biogeographic history of the genus using every known non‐African species, with several African species as outgroup taxa. A phylogenetic hypothesis is proposed, based on molecular characters of the mitochondrial genes cytochrome oxidase subunit I (COI) and NADH dehydrogenase 5 (ND5), and the nuclear wingless gene. Phylogenetic analyses based on maximum parsimony and Bayesian inference of the combined dataset implies that the Indo‐Pacific Charaxes form a monophyletic assemblage, with the exception of Charaxes solon Fabricius. Eight major lineages are recognized in the Indo‐Pacific, here designated the solon (+African), elwesi, harmodius, amycus, mars, eurialus, latona, nitebis, and bernardus clades. Species group relationships are concordant with morphology and, based on the phylogeny, we present the first systematic appraisal and classification of all non‐African species. A biogeographical analysis reveals that, after the genus originated in Africa, the evolutionary history of Charaxes in the Indo‐Pacific, in particular Wallacea, may be correlated with the inferred geological and climatic history of the region. We propose that Wallacea was the area of origin of all Charaxes (excluding C. solon) occurring to the east of Wallace's [1863] Line. The earliest Indo‐Pacific lineages appear to have diverged subsequent to the initial fragmentation of a palaeo‐continent approximately 13 million years ago. Further diversification in Indo‐Pacific Charaxes appears primarily related to climatic changes during the Pliocene and possibly as recently as the Pleistocene. Although both dispersal and vicariance have played important roles in the evolution of the genus within the region, the latter has been particularly responsible for diversification of Charaxes in Wallacea. © 2010 The Linnean Society of London, Biological Journal of the Linnean Society, 2010, 100 , 457–481.     

Faunal changes and geographic crypticism indicate the occurrence of a biogeographic transition zone along the southern East Pacific Rise

Aim Deep‐sea hydrothermal vents have now been reported along all active mid‐ocean ridges and back‐arc basins, but the boundaries of biogeographic entities remain questionable owing to methodological issues. Here we examine biogeographic patterns of the vent fauna along the East Pacific Rise (EPR) and determine the relative roles of regional and local factors on the distribution of biodiversity associated with mussel beds along a poorly explored zone, the southern EPR (SEPR). Location East Pacific Rise. Methods A species list of macrobenthic invertebrates along the EPR was compiled from the literature and supplemented with data recovered during the French research cruise BIOSPEEDO carried out in 2004 along the SEPR. Biogeographic patterns were assessed by combining the identification of morphological species with a molecular barcoding approach. A multivariate regression tree (MRT) analysis was performed to identify any geographic breaks, and an empirical distribution of species richness was compared with predictions provided by a mid‐domain effect model. Macrofaunal community structure associated with mussel beds along the SEPR was analysed in relation to environmental factors using cluster and canonical redundancy analyses. Results Sequencing of the cytochrome c oxidase subunit I gene revealed the occurrence of several cryptic species complexes along the EPR, with the equator separating the southern and northern clades. Furthermore, during the BIOSPEEDO cruise at least 10 still unnamed species were collected between 7°25′ S and 21°33′ S. The shift in community structure identified by MRT analysis was located south of 17°34′ S or south of 13°59′ S, depending on the data used, suggesting that the southern part of the SEPR (17°25′–21°33′ S) constitutes a biogeographic transition zone in the vent fauna along the EPR. At a regional scale, latitude combined with the type of venting was significantly correlated with the community structure associated with mussel beds. Main conclusions Together, the molecular data, in situ observations, and the distribution of species suggest that the high diversity of vent fauna species presently observed between 17°25′ S and 21°33′ S is probably a result of the overlap of several distinct biogeographic provinces. We argue that this area thus constitutes a biogeographic vent fauna transition zone along the EPR.     

Biogeographical analysis of two Polypodium species complexes (Polypodiaceae) in Mexico and Central America

We analyzed the geographical and elevational distributions of two Polypodium complexes from Mexico and Central America. Distribution data of nine species of the Polypodium colpodes complex and the Polypodium plesiosorum complex were obtained from almost 1500 herbarium specimens, field collections in Mexico and Costa Rica, and literature studies. The presence of each species was recorded for each Mesoamerican country, in 1° × 1° grid‐cells and biogeographical provinces. The rarity of species was also evaluated. Although the two complexes show extensive overlap, the P. colpodes complex is distributed mainly along the Pacific versant of Mexico and Central America, whereas the P. plesiosorum complex occurs mainly along the Atlantic versant. Those biogeographical provinces with maximum species diversity are Chiapas (seven species), Sierra Madre del Sur (six species), and the Trans‐Mexican Volcanic belt (six species). Grid‐cells with more species are located mainly in the mountains of central‐southern Mexico and northern Central America. Richness does not decrease or increase with latitude. Elevation distributions showed that most Polypodium species are concentrated in the montane interval and three species groups were recognized based on elevational preferences. Polypodium colpodes and P. plesiosorum are the most widely distributed species, whereas Polypodium castaneum and Polypodium flagellare are the only two species that possess the three attributes of rarity (narrow geographical distribution, high habitat specificity, and scarce local populations). Polypodium species of both complexes are present mainly in the montane regions of the study area and show some degree of geographical sympatry, especially in southern Mexico and northern Central America. This overlapping is explained by the elevation tolerance within montane systems and because most species inhabit three or more vegetation types. The distributional patterns of these complexes coincided with the three regional highlands of Mesoamerica, which are separated from each other by the Isthmus of Tehuantepec and by the lowlands of Nicaragua. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, ?? , ??–??.     

The Indochinese–Sundaic zoogeographic transition: a description and analysis of terrestrial mammal species distributions

Aim We describe the distributions of mammal species between the Indochinese and Sundaic subregions and examine the traditional view that the two faunas show a transition near the Isthmus of Kra on the Thai–Malay peninsula. Location Species distributions are described along a 2000‐km transect from 20° N (northernmost Thailand) to 1° N (Singapore). Methods For the 325 species of native non‐marine mammals occurring along the transect we used published records to provide a database of their distributional records by degree of latitude. Results Along the transect we found 128 Indochinese species with southern range limits, 121 Sundaic species with northern range limits, four un‐assignable endemics and 72 widespread species. In total, 152 southern and 147 northern range limits were identified, and their distribution provides no evidence for a narrow faunal transition near the Isthmus of Kra (10°30′ N) or elsewhere. Range limits of both bats and non‐volant mammals cluster in northernmost peninsular Malaysia (5° N) and 800 km further north, where the peninsula joins the continent proper (14° N). The clusters of northern and southern range limits are not concordant but overlap by 100–200 km. Similarly, the range limits of bats and non‐volant mammals cluster at slightly different latitudes. There are 30% fewer species and range limits in the central and northern peninsula (between 6 and 13° N), and 35 more widely distributed species have range gaps in this region. In addition, we found 70 fewer species at the southern tip of the peninsula (1° N) than at 3–4° N. Main conclusions The deficiencies of both species and species range limits in the central and northern peninsula are attributed to an area effect caused by repeated sea‐level changes. Using a new global glacioeustatic curve developed by Miller and associates we show that there were > 58 rapid sea‐level rises of > 40 m in the last 5 Myr that would have resulted in significant faunal compression and local population extirpation in the narrow central and northern parts of the peninsula. This new global sea‐level curve appears to account for the observed patterns of the latitudinal diversity of mammal species, the concentration of species range limits north and south of this area, the nature and position of the transition between biogeographical subregions, and possibly the divergence of the faunas themselves during the Neogene. The decline of species diversity at the southern end of the transect is attributed to a peninsula effect similar to that described elsewhere.     

Species range size distributions for some marine taxa in the Atlantic Ocean. Effect of latitude and depth

The range size distributions of 6643 species in ten different fish and invertebrate taxa dwelling in pelagic (latitudinal range sizes) and benthic (latitudinal and depth range sizes) habitats on both sides of the Atlantic Ocean (80°N−70°S) were studied. The objectives were to analyse: (1) the range size distribution patterns for the various taxa and whether they have right/left skewed or lognormal distributions; (2) the geographical species distributions, to ascertain whether the distribution ranges change with latitude (Rapoport's rule); and (3) the relationship between the depth ranges of benthic species and their maximum depth of occurrence and how depth range size distributions change with latitude. The pelagic taxa exhibited larger range sizes than did the benthic taxa, continental slope/rise species excepted. On the other hand, the boundaries between geographical provinces for both benthic taxa and pelagic taxa tended to occur in association with major oceanographic processes. The shape of the latitudinal range frequency distributions (LRFDs) of the pelagic organisms were distinctly left‐skewed, and the LRFDs for most taxa were significantly different from lognormal. There was no common pattern for the distributions of the benthic organisms, which were lognormal in Cephalopoda, Stomatopoda, and Crustacea Decapoda and tended to be left‐skewed and significantly different from lognormal in Pisces. The applicability of Rapoport's rule was not clearly inferable from the results, and the rule appears to be conditioned by the location of biogeographical boundaries and the endemism rate in the different biogeographical provinces. A clear increase in depth range size with maximum depth range was observable for benthic species, confirming previous studies. Species’ depth range distributions displayed a discernible latitudinal pattern, right‐skewed at high latitudes and left‐skewed at low latitudes. The location of biogeographical boundaries, and endemism rate by biogeographical province were considered to be the factors most useful in explaining species’ distribution patterns and their conformity or nonconformity to Rapoport's rule. © 2003 The Linnean Society of London, Biological Journal of the Linnean Society, 2003, 80 , 437–455.     

The distribution of the ectoparasite fauna of Sebastes capensis from the southern hemisphere does not correspond with zoogeographical provinces of free-living marine animals

Aim We examined the ectoparasite fauna of Sebastes capensis over almost all its geographical distribution range (Chilean, Argentinean and South African coasts) to determine (1) whether the ectoparasites of this host show a zoogeographical pattern and, if so, (2) how this pattern is related to known zoogeographical patterns for free‐living organisms. Location Fish were captured from 20, 24, 30, 33, 36, 40, 45 and 52° S along the Chilean coast; 11° S on the Peruvian coast; 43° S on the Argentina coast; and 34° S on the South African coast. Methods From April to September 2003 and from April to August 2004, 626 fish were captured. The parasites were collected using standard parasitological techniques. At the component community level, zoogeographical distribution patterns were evaluated using cluster analysis. At the infra‐community level, patterns of similarity in parasite composition among localities were investigated with multivariate discriminant analyses. Results The ectoparasite fauna of S. capensis consists of six species distributed along the whole of the Chilean coast. Four other species are distributed only within the transitional zone between the northern warm temperate region (Peruvian faunistic province), extending from Peru to the northern Chilean coast up to c. 30° S, and the cold temperate region (Magellanic faunistic province). The component communities from latitudes 30 to 40° S showed higher ectoparasite species richness, while localities on the margins of the geographical range showed lower species richness. Cluster analysis indicated a grouping of localities consistent with the transitional zone. Argentina and South Africa always emerged as separate localities. Main conclusions The ectoparasite communities of S. capensis do not follow a distributional pattern concordant with the known biogeographical zones for invertebrates and/or fish along the south‐eastern Pacific. Therefore their ectoparasite fauna is not useful as a zoogeographical indicator, although it does allow us to distinguish the transitional zone of the south‐eastern Pacific. On a more extended geographical scale, it is possible to distinguish the ectoparasite communities of S. capensis in the south‐eastern Pacific (as a whole) from those of Argentina and South Africa.     

Rapid expansion of lichen woodlands within the closed-crown boreal forest zone over the last 50 years caused by stand disturbances in eastern Canada

Aim Our two main goals are first to evaluate the resilience of the boreal forest according to latitude across the closed‐crown forest zone using the post‐disturbance distribution and cover of lichen woodlands and closed‐crown forests as a metric, and second to identify the disturbance factors responsible for the regeneration and degradation of the closed‐crown forest according to latitude since the 1950s. Location The study area extends between 70°00′ and 72°00′ W and throughout the closed‐crown forest zone, from its southern limit near 47°30′ N to its northern limit at the contact with the lichen woodland zone at around 52°40′ N. Methods Recent (1972–2002) and old (1954–1956) aerial photos were used to map the distribution of lichen woodlands across the closed‐crown forest zone. Forest disturbances such as fire, spruce budworm (Choristoneura fumiferana (Clemens)) outbreak, and logging were recorded on each set of aerial photos. Each lichen woodland and stand disturbance was validated by air‐borne surveys and digitized using GIS software. Results Over the last 50 years, the area occupied by lichen woodlands has increased according to latitude; that is, 9% of the area that was occupied by closed‐crown forests has shifted to lichen woodlands. Although logging activities have been concentrated in the same areas during the last 50 years, the area covered by logging has increased significantly. Outbreaks by the spruce budworm occurred predominantly in the southern (47°30′ N to 48°30′ N) and central (48°53′ N to 50°42′ N) parts of the study area, where balsam fir stands are extensive. In the northern part of the study area (51°–52°40′ N), extensive fires affected the distribution and cover of closed‐crown forests and lichen woodlands. Main conclusions Over the last 50 years, the area occupied by closed‐crown forests has decreased dramatically, and the ecological conditions that allow closed‐crown forests to establish and develop are currently less prevalent. Fire is by far the main disturbance, reducing the ability of natural closed‐crown forests to self‐regenerate whatever the latitude. Given the current biogeographical shift from dense to open forests, the northern part of the closed‐crown forest zone is in a process of dramatic change towards the dominance of northern woodlands.     

A biogeographical regionalization of Angolan mammals

相似文献   

Patterns of endemism in south-eastern Pacific benthic polychaetes of the Chilean coast

Aim In this study we evaluate patterns of endemism for benthic polychaete species along the southeastern Pacific coast of Chile. Our goals were (1) to describe latitudinal gradients of endemism and identify areas of high endemism, (2) to evaluate the effect of biogeographical limits on endemism patterns, and (3) to evaluate indirectly the role played by evolutionary dynamics on patterns of endemism. Location South‐eastern Pacific coast of Chile, ranging from Arica (18° S) to Cape Horn (56° S). Methods We used a list of 178 species of endemic, shallow benthic polychaetes to evaluate patterns of endemism. Parsimony analysis of endemicity (PAE) and the endemism index (EI) were used to evaluate hierarchical relationships of endemism between different latitudinal bands, and to identify areas with high degrees of endemism and differences in endemism. We evaluated the effect of biogeographical limits on endemic polychaete fauna by testing for the existence of geometric constraints (mid‐domain effect). The role of evolutionary dynamics on latitudinal patterns of endemism was evaluated with nestedness analysis (NA) using the temperature index. Results The PAE analysis indicated two large, separate areas of endemism: (1) the northern area between 18° S and 38° S, and (2) the southern area between 39° S and 56° S. The endemism index showed a maximum value (32 species) around 39°–41° S. Species‐richness curves of each 3° band of latitude showed a clear mid‐domain effect (69%), but the two maximum points of species richness at mid‐latitudes (36° S to 38° S and 39° S to 41° S) did not correspond to the mid‐domain peak in species richness, presenting a greater number of species than expected by the mid‐domain effect. The nestedness analysis showed that the number of genera reaches a maximum of 70 at mid‐latitudes (36°–41° S), decreasing towards both the northern and southern areas. The spatial distribution of the entire data set of endemic species showed a nested pattern (T° = 24.5°, P < 0.0001). Main conclusions Our results strongly support the existence of a latitudinal gradient of endemism for benthic polychaete species along the Chilean coast. The shape of this gradient is clearly non‐linear, with a marked peak of endemism occurring at mid‐latitudes (36°–41° S, endemism hotspot), which also corresponds to a peak in species richness. Furthermore, this hotspot is the midpoint separating two distinct areas of endemism to the north and south. We suggest that the observed pattern of endemism for benthic polychaete taxa of the Chilean coast can be explained by a combination of geometric constraints and historical mechanisms, such as the processes that affected the Chilean coast during the Neogene (e.g. ENSO, oxygen minimum zone, glaciations).     

Tethyan vicariance, relictualism and speciation: evidence from a global molecular phylogeny of the opisthobranch genus Bulla

Aim Our aims were: (1) to reconstruct a molecular phylogeny of the cephalaspidean opisthobranch genus Bulla, an inhabitant of shallow sedimentary environments; (2) to test if divergence times are consistent with Miocene and later vicariance among the four tropical marine biogeographical provinces; (3) to examine the phylogenetic status of possible Tethyan relict species; and (4) to infer the timing and causes of speciation events. Location Tropical and warm‐temperate regions of the Atlantic, Indo‐West Pacific, Australasia and eastern Pacific. Methods Ten of the 12 nominal species of Bulla were sampled, in a total sample of 65 individuals, together with cephalaspidean outgroups. Phylogenetic relationships were inferred by Bayesian analysis of partial sequences of the mitochondrial cytochrome c oxidase I (COI) and 16S rRNA and nuclear 28S rRNA genes. Divergence times and rates of evolution were estimated using uncorrelated relaxed‐clock Bayesian methods with fossil calibrations (based on literature review and examination of fossil specimens), implemented in beast . The geographical pattern of speciation was assessed by estimating the degree of overlap between sister lineages. Results Four clades were supported: Indo‐West Pacific (four species), Australasia (one species), Atlantic plus eastern Pacific (three species) and Atlantic (two species), with estimated mean ages of 35–46 Ma. Nominal species were monophyletic, but deep divergences were found within one Indo‐West Pacific and one West Atlantic species. Species‐level divergences occurred in the Miocene or earlier. The age of a sister relationship across the Isthmus of Panama was estimated at 7.9–32.1 Ma, and the divergence of a pair of sister species on either side of the Atlantic Ocean occurred 20.4–27.2 Ma. Main conclusions Fossils suggest that Bulla originated in the Tethys realm during the Middle Eocene. Average ages of the four main clades fall in the Eocene, and far pre‐date the 18–19 Ma closure of the Tethys Seaway. This discrepancy could indicate earlier vicariant events, selective extinction or errors of calibration. Similarly, the transisthmian divergence estimate far pre‐dates the uplift of the Panamanian Isthmus at about 3 Ma. Speciation events occurred in the Miocene, consistent with tectonic events in the central Indo‐West Pacific, isolation of the Arabian Sea by upwelling and westward trans‐Atlantic dispersal. Differences in habitat between sister species suggest that ecological speciation may also have played a role. The basal position of the Australasian species supports its interpretation as a Tethyan relict.     

Resilient forest faunal communities in South Africa: a legacy of palaeoclimatic change and extinction filtering?

Aim To examine the influence of climatic extinction filtering during the last glacial maximum (LGM; c. 18,000 yr bp ) and of the subsequent recolonization of forest faunas on contemporary assemblage composition in southern African forests. Location South Africa, Mozambique, Swaziland, Zimbabwe. Methods Data comprised presence/absence by quarter‐degree grid cell for forest‐dependent and forest‐associated birds, non‐volant mammals and frogs. Twenty‐one forest subregions were assigned to one of three previously identified forest types: Afrotemperate, scarp, and Indian Ocean coastal belt. Differences among forest types were examined through patterns and gradients of species richness and endemism, assemblage similarity, species turnover, and coefficients of species dispersal direction. The influence of contemporary environment on assemblage composition was investigated using partial canonical correspondence analysis. Several alternative biogeographical hypotheses for the recolonization of forest faunas were tested. Results Afrotemperate faunas are relatively species‐poor, have low species turnover, and are unsaturated and infiltrated by generalist species. In northern and central regions, communities are supplemented by recolonization from scarp forest refugia, and among frogs by autochthanous speciation in localized refugia. Scarp faunas are relatively species‐rich, contain many forest‐dependent species, have high species turnover, and overlap with coastal and Afrotemperate faunas. Coastal forests are relatively species‐rich with high species turnover. Main conclusions Afrotemperate communities were affected most by climatic extinction filtering events. Scarp forests were Afrotemperate refugia during the LGM and are a contemporary overlap zone between Afrotemperate and coastal forest. Coastal faunas derive from post‐LGM colonization along the eastern seaboard from tropical East African refugia. The greatest diversity is achieved in scarp and coastal forest faunas in northern KwaZulu–Natal province. This historical centre of diversity has influenced the faunal diversity of nearly all other forests in South Africa. The response of vertebrate taxa to large‐scale, historical processes is dependent on their relative mobility: forest birds best illustrate patterns resulting from post‐glacial faunal dispersal, while among mammals and frogs the legacy of climatic extinction filtering remains stronger.     

Integrating ecology with biogeography using landscape characteristics: a case study of subtidal habitat across continental Australia

Aim We aimed to redress a current limitation of local ecological studies (i.e. piecemeal information on specific taxa) by integrating existing ecological knowledge with quantifiable patterns in primary habitat (i.e. composition, distribution and cover) from local to continental scales. By achieving this aim, we sought to provide a biogeographical framework for the interpretation of variation in the ecology of, and threats to, subtidal rocky landscapes. Location The subtidal rocky coast of continental Australia, with longitudinal comparisons spanning > 4000 km of southern coast (115°03′ E–153°60′ E) between latitudes of 33°05′ S and 35°36′ S, and latitudinal comparisons across 26°40′ S to 37°08′ S of eastern Australia. Methods The frequency and size of patches of major benthic habitat were quantified to indicate contemporary function (ecology) and to establish patterns that may result from contrasting regional‐scale processes (biogeography). This was achieved by quantifying the composition and patchiness of key subtidal habitats across the continent and relating them to the known ecology of subsets of locations in each region. A nested design of several spatial scales (1000s, 100s, 10–1 km) was adopted to distinguish patterns at local through to biogeographical scales. Results We show biogeography (in terms of longitude and latitude) to have a fundamental influence on the patterns of abundance and composition of subtidal habitats across regional (1000s of kilometres) to local (10s of kilometres to metres) scales. Across the continent, the most fundamental patterns related to (1) the proportion of rock covered by kelp forests, as related to particular functional groups of herbivores, and (2) the small‐scale heterogeneity (metres) that characterizes these forests. Main conclusions We interpret these results within a framework of alternative processes known to maintain habitat heterogeneity across these regions (e.g. productivity versus consumption as shapers of habitat structure). These interpretations illustrate how regional differences in ecological patterns and processes can create contradictory outcomes for the management of natural resources. We suggest that researchers and managers of natural resources alike may benefit from understanding local issues (e.g. the effects of fishing and its synergies with water pollution) in their biogeographical contexts.     

Patterns in marine hydrozoan richness and biogeography around southern Africa: implications of life cycle strategy

Aim To examine patterns of marine hydrozoan richness around southern Africa and to test the hypothesis that patterns of biogeography become weaker with increasing dispersal ability. Location The coastline of southern Africa from 21° S, 14° E to 28° S, 33° E, extending from the intertidal zone seawards a distance of 200 nautical miles. Methods Published and unpublished information on the distribution of marine Hydrozoa was entered as presence/absence data onto a gridded coastline of the region. A similarity matrix between the species composition of grid squares was constructed using the Bray–Curtis index and visualized using non-metric multidimensional scaling ordinations. Separate analyses were conducted, and compared, on the three major life cycle groupings: holoplanktic, meroplanktic and benthic. Results Over 450 species of marine Hydrozoa have been reported from the region, and species richness increases eastwards, in a manner at odds with the distribution of sampling effort. There was a significant correlation between the geographic structures of the resemblance matrices generated for the three life cycle groupings. In other words, all three groups showed similar patterns of biogeography around the region, and these were broadly similar to those reported by others. However, there were differences between them that reflect the resolution at which the data were examined. At a level of 40% similarity, there was no biogeographic structure to the holoplanktic fauna, the meroplanktic taxa were simply sub-divided into cool- and warm-temperate/subtropical elements, and in the case of benthic taxa, the cool-water fauna was further split into a southern Namaqua and a depauperate northern Namib component. Even at a resolution of 70% similarity, the holopelagic taxa could be separated only into cool-temperate and warm-temperate/subtropical faunas. Main conclusions Holoplanktic taxa show comparatively less biogeographic structure than meroplanktic taxa, which in turn show less clearly defined biogeographic structure than benthic taxa. It is suggested that this is related to the interaction between oceanography and dispersive-stage duration. The role that the Agulhas Current plays in influencing the Benguela Current fauna is highlighted. This study has implications for conservation planning exercises based on protecting representative biotopes in different biogeographic regions.     

Unrecognized and imperilled diversity in an endemic barb (Smiliogastrini,Enteromius) from the Fouta Djallon highlands

The Upper Guinean Forests of Guinea, Sierra Leone and Liberia contain high levels of freshwater biodiversity. The Guinean Range and associated Fouta Djallon highlands separate two biogeographical provinces in the region and served as a refugium during past climatic fluctuations. While many species of freshwater fishes are restricted to one biogeographical province or the other, some are reported to occur on both sides of the divide. Here, we examine the molecular and morphological diversity of an endemic small African barb, Enteromius foutensis, reported to occur in both provinces. This integrative analysis revealed unrecognized diversity and suggests recent, or ongoing, events that facilitated geodispersal and subsequent divergence in the region. The molecular analysis revealed three divergent and well‐supported populations within E. foutensis. Accounting for allometric shape variation allowed us to observe diagnostic morphological differences among the populations. Enteromius foutensis sensu stricto is restricted to the Little Scarcies drainage in Guinea and northern Sierra Leone. Our study revealed two candidate species distinct from E. foutensis. One is likely a narrow endemic restricted to a small area in the Konkouré River basin; the other candidate species inhabits the upper Senegal and Gambie River drainages. How these patterns of diversity compare with other freshwater species from the Fouta Djallon highlands and the conservation status of these candidate species are also discussed.