Hostname: page-component-745bb68f8f-v2bm5 Total loading time: 0 Render date: 2025-02-05T23:10:21.593Z Has data issue: false hasContentIssue false
  • English
  • Français

The Sepik River (Papua New Guinea) is not a dispersal barrier for lowland rain-forest frogs

Published online by Cambridge University Press:  11 September 2013

Chris Dahl ,
Stephen J. Richards  and
Vojtech Novotny
Chris Dahl*
Affiliation:
University of South Bohemia and Biology Centre of the Academy of Sciences of the Czech Republic, Branisovska 31 370 05 Ceske Budejovice, Czech Republic
Stephen J. Richards
Affiliation:
Herpetology Department, South Australian Museum, North Terrace, Adelaide and Center for Applied Biodiversity Science, Conservation International, Atherton, Australia
Vojtech Novotny
Affiliation:
University of South Bohemia and Biology Centre of the Academy of Sciences of the Czech Republic, Branisovska 31 370 05 Ceske Budejovice, Czech Republic
*
1Corresponding author. Email: cd.rokrok@gmail.com
Rights & Permissions [Opens in a new window]

Abstract:

Major tropical rivers have been suggested to be important dispersal barriers that increase the beta diversity of animal communities in lowland rain forests. We tested this hypothesis using assemblages of frogs in the floodplains of the Sepik River, a major river system in Papua New Guinea. We surveyed frogs at five sites within a continuous 150 × 500-km area of lowland rain forest bisected by the Sepik, using standardized visual and auditory survey techniques. We documented 769 frogs from 44 species. The similarity in species composition decreased with logarithm of geographical distance between the sites, which ranged from 82 to 465 km. The similarity decay did not depend on whether or not the compared sites were separated by the Sepik River or whether the species were aquatic or terrestrial breeders. Likewise, a DCA ordination of frog assemblages did not show separation of sites by the river as a significant factor explaining their composition. Our results suggest that even major rivers, such as the Sepik, may not act as dispersal barriers. Rivers may not limit the distribution of frogs and therefore have a limited effect on determining frog species abundance and assemblage structure in rain forests.

Keywords

amphibiansbeta diversityfrogsriverine barrier hypothesisspecies diversity
Type
Research Article
Information
Journal of Tropical Ecology , Volume 29 , Issue 6 , November 2013 , pp. 477 - 483
Copyright
Copyright © Cambridge University Press 2013 

INTRODUCTION

Major tropical rivers may act as barriers to dispersal for some animal taxa, thus promoting genetic divergence in their populations and ultimately their speciation. The river barrier hypothesis, proposed by Wallace (Reference WALLACE1852) to explain the distribution of monkeys in the Amazon, and later suggested as an explanation of the extraordinary species diversity in the Amazon in general, implies that (1) rivers cause increased genetic divergence between conspecific populations, (2) geographical distribution of species is constrained by rivers, and (3) species on the same side of the river tend to be closely related, forming monophyletic groups. These effects should depend on the size of the river, increasing thus from the headwaters to the river estuary. The river barrier hypothesis has been tested using (1) molecular markers in populations (mammals: Patton et al. Reference PATTON, DASILVA and MALCOLM1994, Peres et al. Reference PERES, PATTON and DASILVA1996; birds: Aleixo Reference ALEIXO2004; frogs: Funk et al. Reference FUNK, CALDWELL, PEDEN, PADIAL, DE LA RIVA and CANNATELLA2007, Gascon et al. Reference GASCON, LOUGHEED and BOGART1998, Noonan & Wray Reference NOONAN and WRAY2006, Zhao et al. Reference ZHAO, DAI and FU2009; insects: Tantrawatpan et al. Reference TANTRAWATPAN, SAIJUNTHA, PILAB, SAKDAKHAM, PASORN, THANONKEO, THIHA, SATRAWAHA and PETNEY2011), (2) distribution of geographical ranges of species (mammals: Harcourt & Wood Reference HARCOURT and WOOD2012; birds: Hayes & Sewlal Reference HAYES and SEWLAL2004; insects: Knopp et al. Reference KNOPP, RAHAGALALA, MIINALA and HANSKI2011), and (3) phylogeographical analysis (frogs: Symula et al. Reference SYMULA, SCHULTE and SUMMERS2003; insects: Hall & Harvey Reference HALL and HARVEY2002). While numerous studies have not found rivers acting as barriers (Gascon et al. Reference GASCON, LOUGHEED and BOGART1998, Knopp et al. Reference KNOPP, RAHAGALALA, MIINALA and HANSKI2011, Patton et al. Reference PATTON, DASILVA and MALCOLM1994, Symula et al. Reference SYMULA, SCHULTE and SUMMERS2003), others documented an increased diversification across the river at least in some circumstances or taxa (Aleixo Reference ALEIXO2004, Hayes & Sewlal Reference HAYES and SEWLAL2004, Noonan & Wray Reference NOONAN and WRAY2006, Peres et al. Reference PERES, PATTON and DASILVA1996, Tantrawatpan et al. Reference TANTRAWATPAN, SAIJUNTHA, PILAB, SAKDAKHAM, PASORN, THANONKEO, THIHA, SATRAWAHA and PETNEY2011, Zhao et al. Reference ZHAO, DAI and FU2009). As summarized by Haffer (Reference HAFFER1997) for Amazonia, rivers are not unimportant, but have been overrated as barriers.

Almost all studies focused geographically on rivers in Amazonia (but see Tantrawatpan et al. Reference TANTRAWATPAN, SAIJUNTHA, PILAB, SAKDAKHAM, PASORN, THANONKEO, THIHA, SATRAWAHA and PETNEY2011, Zhao et al. Reference ZHAO, DAI and FU2009) and considered their importance for the origin of species (Haffer Reference HAFFER1997). Here, we broaden the analysis to examine whether rivers, acting as dispersal barriers, can also impact on the quantitative composition of communities. In particular, barriers may slow down, rather than completely prevent, dispersal which may lead to metapopulation dynamics, where each population could follow its own population dynamics, and to some extent function independently from other populations (Hanski Reference HANSKI1999). Rivers acting as barriers could thus also cause heterogeneity in quantitative community composition, and not only influence the presence and absence of species. Their effect will be examined here using frog assemblages in the lowland rain forests surrounding the Sepik, one of the major tropical rivers outside the Amazon.

METHODS

The study was situated in northern New Guinea, within lowlands bisected by the Sepik, a major tropical river (Figure 1). The Sepik River is 1100km long, and has a mean discharge of 7000 m3 s−1 (Mitchell et al. Reference MITCHELL, PETR and VINER1980). It is the largest river system in Papua New Guinea in terms of area drained (78000km2). It dissects approximately 7500km2 of relatively uniform lowland rain forest on its floodplains (Novotny et al. Reference NOVOTNY, MILLER, BASSET, JANDA, SETLIFF, HULCR, STEWART, AUGA, MOLEM, MANUMBOR, TAMTIAI, MOGIA and WEIBLEN2007). The study area is part of a complex tectonic region at the convergence of the Australian and Pacific plates. The northern lowlands in New Guinea originated as a result of the accretion of volcanic arc terranes to the central cordillera which borders our study area in the south. The Bewani and Torricelli ranges bordering the study area in the north accreted 30–35 million years ago (Davies et al. Reference DAVIES, PEREMBO, WINN, KENGEMAR and Hancock1997, Pigram & Davies Reference PIGRAM and DAVIES1987). The last accretion event involved the Adelbert and Finisterre block and took place about two million years ago (Abbott Reference ABBOTT1995).

Figure 1. Map of Papua New Guinea showing the study area with five surveyed sites, two on the northern and three on the southern side of the Sepik River.

The study area comprises a 500 × 150-km area of lowland terrain with continuous rain forest. The Sepik River, which is up to 1 km wide, accompanying floodplain swamps, lakes and grasslands, up to 70 km wide, represent the only large discontinuity in the rain forest ecosystem of the study area (Reiner & Robbins Reference REINER and ROBBINS1964). We surveyed five sites (03°24′–05°14′S, 141°05′–145°12′E), including two sites on the north side of the river and three sites south of the Sepik River (Figure 1). All sites were located in the lowland rain forest (40–250 m asl) with a mean annual rainfall of 2000–4000 mm, a moderate dry season from July to September, and a mean air temperature of ~26 °C (McAlpine et al. Reference MCALPINE, KEIG and FALLS1983).

Each site was surveyed once, during the wet season, between May 2004 and March 2005. All habitats within an approximately 5-km2 area, including primary and secondary forest vegetation, swamps and stagnant water bodies, were surveyed at each site between 18h30 and 03h30 for 13–20 consecutive nights. This sampling effort was sufficient to survey local assemblages (Dahl et al. Reference DAHL, NOVOTNY, MORAVEC and RICHARDS2009) and its results are used in the present analysis (Appendix 1). We detected frogs visually using a headlamp and also by their advertisement calls. The frog calls were recorded, specimens were photographed and vouchers deposited at the University of Papua New Guinea (Port Moresby) and the South Australian Museum (Adelaide).

The frog species were classified into aquatic breeders, those having a tadpole stage, and terrestrial breeders that are independent of water as they have direct embryonic development from egg to a fully formed hatchling frog (Anstis et al. Reference ANSTIS, PARKER, HAWKES, MORRIS and RICHARDS2011, Austin et al. Reference AUSTIN, HAYDEN, BIGILALE, DAHL and ANAMINIATO2008, Zweifel & Tyler Reference ZWEIFEL and TYLER1982). These two ecological groups were analysed separately.

DATA ANALYSIS

The similarity of frog assemblages was measured using the Chao–Jaccard similarity index which measures the proportion of shared species corrected for possible bias owing to incomplete sampling of rare species. We also calculated the Bray–Curtis similarity index which responds to differences in abundance of individual species, not only their presence and absence. Both similarity measures were calculated using EstimateS. The possible change in species composition in frog assemblages from opposite sides of the river was explored by the Detrended Correspondence Analysis (with species abundance log transformed and rare species down-weighted) (Leps & Smilauer Reference LEPS and SMILAUER2003). The analysis was implemented in the program Canoco 4.5. The distribution of frog species at the same or different sides of the river was tested against the null hypothesis for each species retained and the number of sites where it was sampled, but we randomized their position with regards to the Sepik River.

RESULTS

We sampled a total of 769 individual frogs from 44 species (Appendix 1). Species richness ranged from 18 to 26 species per site. Seventeen species (39%) were recorded from only a single site, while 12 species (27%) were widespread; occurring at four–five sites and thus inevitably on both sides of the Sepik River. The seven species present at two sites had 40% probability of being limited to a single side of the river by chance, for the eight species present at three sites the probability was 10%. The combined random distribution expectation for species found at two or three sites was 3.6 species limited to a single side of the river, which was not significantly different from three species which exhibited this distribution (χ2 = 0.132, df = 1, P > 0.7, Figure 2).

Figure 2. The distribution of frog species (limited to the northern or southern Sepik River bank, or on both sides of the river) recorded from one, two, three, and more than three sites. The observed numbers of species (obs) are compared to the numbers expected (exp) for random distribution of species which retains the number of sites from which each species was recorded but randomizes their position with respect to the Sepik river.

Ten from the total of 18 species breeding in water were recorded from a single side of the river, as well as 11 from the 26 terrestrially breeding species. The proportion of species with restricted distribution was not significantly different between aquatic and terrestrial breeders (56% and 42% respectively, Fisher's exact test, P > 0.05).

The similarity of species composition (Chao–Jaccard index) between sites decreased with the logarithm of their distance. Further, the Bray–Curtis index which measures the similarity in the abundance of species decreased linearly with geographical distance (Figure 3). The similarity values for the paired sites on the same side of the river and those paired sites from the opposite sides of the river followed the same decreasing similarity of the distance relationship, although our data are too limited for formal statistical tests.

Figure 3. The relationship between similarity of frog assemblages, Chao–Jaccard (a) and Bray–Curtis (b) indices, and their geographical distance in Papua New Guinea. The similarity decay with distance function (logarithmic for Chao–Jaccard and linear for Bray–Curtis indices) was fitted to all pairs of sites, including those confined to the northern (N) or southern (S) side of the river and those including sites from the opposite sides of the river (N–S).

The quantitative composition of frog assemblages as depicted by DCA did not reveal any dichotomy between the northern and the southern sides of the river (Figure 4).

Figure 4. The DCA ordination of frog assemblages showing similarity relationships between study sites on the northern (solid squares) and southern (solid triangles) banks of the Sepik river. Empty squares show distributional optima of all frog species. Centroids show individual study sites: UTA = Utai, WAM = Wamangu, YAP = Yapsiei, MOR = Morox, WAN = Wanang.

DISCUSSION

We tested the riverine barrier hypothesis by surveying frogs in all habitat types, viz. the forest floor, arboreal habitats and forest streams at a series of sites on both sides of the Sepik River. Although the assemblages exhibited distinct geographical patterns, as indicated by their similarity decay with increasing geographical distance, we found there was no evidence that the Sepik River acts as a barrier to frog dispersal in northern Papua New Guinea. These results support the conclusions of Novotny et al. (Reference NOVOTNY, MILLER, BASSET, JANDA, SETLIFF, HULCR, STEWART, AUGA, MOLEM, MANUMBOR, TAMTIAI, MOGIA and WEIBLEN2007) who found that the Sepik River was not a dispersal barrier for herbivorous insects (Lepidoptera). This conclusion was likely expected for the aquatic breeders but interestingly applied to terrestrial-breeding frog species constituting the majority of species in New Guinea's forest.

Our study area represents a highly dynamic landscape, which is characteristic for New Guinea (Johns Reference JOHNS1986). Most areas with our study sites between the central and the northern ranges submerged from the Early Miocene until the Pliocene epoch (Davies et al. Reference DAVIES, PEREMBO, WINN, KENGEMAR and Hancock1997). The oceanic incursions continued during the periods of elevated sea level, including a sea that stretched 100 km inland and has separated our Wamangu and Wanang sites as recently as 6000 y ago (Swadling Reference SWADLING1997). The vegetation has also changed over time; a mosaic of broadleaved open and closed forests covered the study area during a cooler and drier period about 17000 y ago (Nix & Kalma Reference NIX, KALMA and Walker1972). The rapid changes in the course of the river, the extent of flowing water, the stagnant water and the swampy habitats, as well as the sea incursions suggest that our analysis tests the recent dispersal ability of frog species in a highly variable landscape, rather than ancient barriers to speciation.

Other tropical studies also failed to find major effects of rivers on the species composition of frogs (Gascon et al. Reference GASCON, MALCOLM, PATTON, DA SILVA, BOGART, LOUGHEED, PERES, NECKE and BOAG2000) or on the intraspecific genetic differentiation among their populations (Gascon et al. Reference GASCON, LOUGHEED and BOGART1998, Gehring et al. Reference GEHRING, PABIJAN, RANDRIANIRINA, GLAW and VENCES2012, Lougheed et al. Reference LOUGHEED, GASCON, JONES, BOGART and BOAG1999, Zhao et al. Reference ZHAO, DAI and FU2009). However, rivers can still be important barriers to other taxa, such as birds (Briggs Reference BRIGGS1974, Hayes & Sewlal Reference HAYES and SEWLAL2004), or frogs in certain ecological situations. For instance, rivers are barriers to gene flow to the frog species Rana kukunoris at high elevations (Li et al. Reference LI, CHEN, TU and FU2009). Further, Fouquet et al. (Reference FOUQUET, LEDOUX, DUBUT, NOONAN and SCOTTI2012) detected restricted gene flow across a river for a terrestrially breeding species Adenomera andreae with limited dispersal ability.

In conclusion, it appears that Papua New Guinean rivers may not be important barriers limiting frog dispersal, a pattern also prevalent elsewhere. Our conclusions are preliminary since our study was based on a limited number of sites. On the other hand, we have expanded previous analytical approaches to explore the effect of rivers on quantitative community composition, which potentially affects the metapopulation dynamics at the species level. Both qualitative and quantitative results suggested that the river had no effect. In contrast mountain ridges, and in particular the central cordillera that divides the island's lowlands into northern and southern parts, played a major role in determining the frog distributions, as suggested by the differences in the frog faunas between the Papua New Guinea's northern and southern lowlands (Allison Reference ALLISON, Keast and Miller1996, Richards Reference RICHARDS2002).

ACKNOWLEDGEMENTS

We acknowledge funding support from Conservation International to S.J.R. and C.D., the Czech Ministry of Education (CZ.1.07/2.3.00/20.0064, LH11008) to V.N., the UK Darwin Initiative to C.D. and V.N., and the US National Science Foundation to C.D. and V.N. (DEB0211591) and to Christopher C. Austin (DEB0445213). We thank the PNG Wildlife Conservation Society and the PNG Institute of Biological Research for logistical support, Jimmy Moses for the study map in Figure 1, the numerous field assistants for their help and the forest owners for the permission to work on their lands.

Appendix 1. The number of individual frogs recorded for each species and study site. Each species is characterized by its family, mode of reproduction (Rep., Aqua = Aquatic or Terr = Terrestrial), the number of individuals at each study site (Uta = Utai, Wam = Wamangu, Yap = Yapsiei, Mor = Morox, Wan = Wanang), the total number of individuals (Sum) and the number of sites.

References

LITERATURE CITED

ABBOTT, L. D. 1995. Neogene tectonic reconstruction of the Adelbert-Finisterre-New Britain collision, northern Papua New Guinea. Journal of Asian Earth Sciences 11:33–51.CrossRefGoogle Scholar
ALEIXO, A. 2004. Historical diversification of a Terra-firme forest bird superspecies: a phylogeographic perspective on the role of different hypotheses of Amazonian diversification. Evolution 58:1303–1317.Google ScholarPubMed
ALLISON, A. 1996. Zoogeography of amphibians and reptiles of New Guinea and the Pacific region. Pp. 407–436 in Keast, A. & Miller, S. E. (eds.). The origin and evolution of Pacific Island biotas, New Guinea to eastern Polynesia: patterns and processes. SPB Academic Publishing, Amsterdam.Google Scholar
ANSTIS, M., PARKER, F., HAWKES, T., MORRIS, I. & RICHARDS, S. J. 2011. Direct development in some Australopapuan microhylid frogs of the genera Austrochaperina, Cophixalus and Oreophryne (Anura: Microhylidae) from northern Australia and Papua New Guinea. Zootaxa 3052:1–50.CrossRefGoogle Scholar
AUSTIN, C. C., HAYDEN, C. J., BIGILALE, I., DAHL, C. & ANAMINIATO, J. 2008. Checklist and comments on the terresterial amphibian and reptile fauna from Utai northwestern Papua New Guinea. Herpetological Review 39:40–46.Google Scholar
BRIGGS, J. C. 1974. Operation of zoogeographic barriers. Systematic Zoology 23:248–256.CrossRefGoogle Scholar
DAHL, C., NOVOTNY, V., MORAVEC, J. & RICHARDS, S. J. 2009. Beta diversity of frogs in the forests of New Guinea, Amazonia and Europe: contrasting tropical and temperate communities. Journal of Biogeography 36:896–904.CrossRefGoogle Scholar
DAVIES, H. L., PEREMBO, R. C. B., WINN, R. D. & KENGEMAR, P. 1997. Terranes of the New Guinea orogen. Pp. 61–66 in Hancock, G. (ed.). Proceedings of the Geology Exploration and Mining Conference, Madang. Australasian Institute of Mining and Metallurgy, Melbourne.Google Scholar
FOUQUET, A., LEDOUX, J. B., DUBUT, V., NOONAN, B. P. & SCOTTI, I. 2012. The interplay of dispersal limitation, rivers, and historical events shapes the genetic structure of an Amazonian frog. Biological Journal of the Linnean Society 106:356–373.CrossRefGoogle Scholar
FUNK, W. C., CALDWELL, J. P., PEDEN, C. E., PADIAL, J. M., DE LA RIVA, I. & CANNATELLA, D. C. 2007. Tests of biogeographic hypotheses for diversification in the Amazonian forest frog, Physalaemus petersi. Molecular Phylogenetics and Evolution 44:825–837.CrossRefGoogle ScholarPubMed
GASCON, C., LOUGHEED, S. C. & BOGART, J. P. 1998. Patterns of genetic population differentiation in four species of Amazonian frogs: a test of the riverine barrier hypothesis. Biotropica 30:104–119.CrossRefGoogle Scholar
GASCON, C., MALCOLM, J. R., PATTON, J. L., DA SILVA, M. N. F., BOGART, J. P., LOUGHEED, S. C., PERES, C. A., NECKE, L. S. & BOAG, P. T. 2000. Riverine barriers and the geographic distribution of Amazonian species. Proceedings of the National Academy of Sciences USA 97:13672–13677.CrossRefGoogle ScholarPubMed
GEHRING, P. S., PABIJAN, M., RANDRIANIRINA, J. E., GLAW, F. & VENCES, M. 2012. The influence of riverine barriers on phylogeographic patterns of Malagasy reed frogs (Heterixalus). Molecular Phylogenetics and Evolution 64:618–632.CrossRefGoogle ScholarPubMed
HAFFER, J. 1997. Alternative models of vertebrate speciation in Amazonia: an overview. Biodiversity and Conservation 6:451–476.CrossRefGoogle Scholar
HALL, J. P. W. & HARVEY, D. J. 2002. The phylogeography of Amazonia revisited: new evidence from roidinid butterflies. Evolution 56:1489–1497.Google ScholarPubMed
HANSKI, I. 1999. Metapopulation ecology. Oxford University Press, Oxford. 328 pp.CrossRefGoogle Scholar
HARCOURT, A. H. & WOOD, M. A. 2012. Rivers as barriers to primate distributions in Africa. International Journal of Primatology 33:168–183.CrossRefGoogle Scholar
HAYES, F. E. & SEWLAL, J. N. 2004. The Amazon River as a dispersal barrier to passerine birds: effects of river width, habitat and taxonomy. Journal of Biogeography 31:1809–1818.CrossRefGoogle Scholar
JOHNS, R. J. 1986. The instability of the tropical ecosystem in New Guinea. Blumea 31:341–361.Google Scholar
KNOPP, T., RAHAGALALA, P., MIINALA, M. & HANSKI, I. 2011. Current geographical ranges of Malagasy dung beetles are not delimited by large rivers. Journal of Biogeography 38:1098–1108.CrossRefGoogle Scholar
LEPS, J. & SMILAUER, P. 2003. Multivariate analysis of ecological data using CANOCO. Cambridge University Press, Cambridge. 267 pp.CrossRefGoogle Scholar
LI, R., CHEN, W., TU, L. & FU, J. 2009. Rivers as barriers for high elevation amphibians: a phylogeographic analysis of the alpine stream frog of the Hengduan Mountains. Journal of Zoology 277:309–316.CrossRefGoogle Scholar
LOUGHEED, S. C., GASCON, C., JONES, D. A., BOGART, J. P. & BOAG, P. T. 1999. Ridges and rivers: a test of competing hypotheses of Amazonian diversification using a dart-poison frog (Epipedobates femoralis). Proceedings of the Royal Society of London 266:1829–1835.CrossRefGoogle Scholar
MCALPINE, J. R., KEIG, G. & FALLS, R. 1983. Climate of Papua New Guinea. The Australian National University, Australia. 200 pp.Google Scholar
MITCHELL, D. S., PETR, T. & VINER, A. B. 1980. The water-fern Salvinia molesta in the Sepik River, Papua New Guinea. Environmental Conservation 7:115–122.CrossRefGoogle Scholar
NIX, H. A. & KALMA, J. D. 1972. Climate as a dominant control in the biogeography of northern Australia and New Guinea. Pp. 61–92 in Walker, D. (ed.). Bridge and barrier: the natural and cultural history of Torres Straight. Australian National University, Canberra.Google Scholar
NOONAN, B. P. & WRAY, K. P. 2006. Neotropical diversification: the effects of a complex history on diversity within the poison frog genus Dendrobates. Journal of Biogeography 33:1007–1020.CrossRefGoogle Scholar
NOVOTNY, V., MILLER, S. E., BASSET, Y., JANDA, M., SETLIFF, G. P., HULCR, J., STEWART, A. J. A., AUGA, J., MOLEM, K., MANUMBOR, M., TAMTIAI, E., MOGIA, M. & WEIBLEN, G. D. 2007. Low beta diversity of herbivorous insects in tropical forests. Nature 448:692–696.CrossRefGoogle ScholarPubMed
PATTON, J. L., DASILVA, M. N. F. & MALCOLM, J. R. 1994. Gene genealogy and differentiation among arboreal spiny rats (Rodentia, Echimydae) of the Amazon Basin – a test of the riverine barrier hypothesis. Evolution 48:1314–1323.CrossRefGoogle Scholar
PERES, C. A., PATTON, J. L. & DASILVA, M. N. F. 1996. Riverine barriers and gene flow in Amazonian saddle-back tamarins. Folia Primatologica 67:113–124.CrossRefGoogle ScholarPubMed
PIGRAM, C. J. & DAVIES, H. L. 1987. Terranes and the accretion history of the New Guinea orogen. Journal of Australian Geology and Geophysics 10:193–211.Google Scholar
REINER, E. J. & ROBBINS, R. G. 1964. The middle Sepik Plains, New Guinea: a physiographic study. Geographical Review 54:20–44.CrossRefGoogle Scholar
RICHARDS, S. J. 2002. Rokrok: an illustrated guide to the frogs of the Kikori River Basin. WWF, Port Moresby. 35 pp.Google Scholar
SWADLING, P. 1997. Changing shorelines and cultural orientations in the Sepik-Ramu, Papua New Guinea: implications for Pacific prehistory. World Archaeology 29:1–14.CrossRefGoogle Scholar
SYMULA, R., SCHULTE, R. & SUMMERS, K. 2003. Molecular systematics and phylogeography of Amazonian poison frogs of the genus Dendrobates. Molecular Phylogenetics and Evolution 26:452–475.CrossRefGoogle ScholarPubMed
TANTRAWATPAN, C., SAIJUNTHA, W., PILAB, W., SAKDAKHAM, K., PASORN, P., THANONKEO, S., THIHA, SATRAWAHA, R. & PETNEY, T. 2011. Genetic differentiation among populations of Brachytrupes portentosus (Lichtenstein 1796) (Orthoptera: Gryllidae) in Thailand and the Lao PDR: the Mekong River as a biogeographic barrier. Bulletin of Entomological Research 101:687–696.CrossRefGoogle ScholarPubMed
WALLACE, A. R. 1852. On the monkeys of the Amazon. Proceedings of the Zoological Society of London'18; 20:107–110.Google Scholar
ZHAO, S., DAI, Q. & FU, J. 2009. Do rivers function as genetic barriers for the plateau wood frog at high elevations? Journal of Zoology 279:270–276.CrossRefGoogle Scholar
ZWEIFEL, R. G. & TYLER, M. J. 1982. Amphibia of New Guinea. Monographiae Biologicae 41:759–798.CrossRefGoogle Scholar
Figure 0

Figure 1. Map of Papua New Guinea showing the study area with five surveyed sites, two on the northern and three on the southern side of the Sepik River.

Figure 1

Figure 2. The distribution of frog species (limited to the northern or southern Sepik River bank, or on both sides of the river) recorded from one, two, three, and more than three sites. The observed numbers of species (obs) are compared to the numbers expected (exp) for random distribution of species which retains the number of sites from which each species was recorded but randomizes their position with respect to the Sepik river.

Figure 2

Figure 3. The relationship between similarity of frog assemblages, Chao–Jaccard (a) and Bray–Curtis (b) indices, and their geographical distance in Papua New Guinea. The similarity decay with distance function (logarithmic for Chao–Jaccard and linear for Bray–Curtis indices) was fitted to all pairs of sites, including those confined to the northern (N) or southern (S) side of the river and those including sites from the opposite sides of the river (N–S).

Figure 3

Figure 4. The DCA ordination of frog assemblages showing similarity relationships between study sites on the northern (solid squares) and southern (solid triangles) banks of the Sepik river. Empty squares show distributional optima of all frog species. Centroids show individual study sites: UTA = Utai, WAM = Wamangu, YAP = Yapsiei, MOR = Morox, WAN = Wanang.

Figure 4

Appendix 1. The number of individual frogs recorded for each species and study site. Each species is characterized by its family, mode of reproduction (Rep., Aqua = Aquatic or Terr = Terrestrial), the number of individuals at each study site (Uta = Utai, Wam = Wamangu, Yap = Yapsiei, Mor = Morox, Wan = Wanang), the total number of individuals (Sum) and the number of sites.