We investigated the large and small scale evolutionary relationships of the endemic Western Australian subterranean shrimp genus Stygiocaris (Atyidae) using nuclear and mitochondrial genes. Stygiocaris is part of the unique cave biota of the coastal, anchialine, limestones of the Cape Range and Barrow Island, most of whose nearest evolutionary relations are found in coastal caves of the distant North Atlantic. The dominance of atyids in tropical waters and their food resources suggest they are pivotal in understanding these groundwater ecosystems.
Our nuclear and mitochondrial analyses all recovered the Mexican cave genus Typhlatya as the sister taxon of Stygiocaris, rather than any of the numerous surface and cave atyids from Australia or the Indo-Pacific region. The two described Stygiocaris species were recovered as monophyletic, and a third, cryptic, species was discovered at a single site, which has very different physiochemical properties from the sites hosting the two described species.
Our findings suggest that Stygiocaris and Typhlatya may descend from a common ancestor that lived in the coastal marine habitat of the ancient Tethys Sea, and were subsequently separated by plate tectonic movements. This vicariant process is commonly thought to explain the many disjunct anchialine faunas, but has rarely been demonstrated using phylogenetic techniques. The Cape Range's geological dynamism, which is probably responsible for the speciation of the various Stygiocaris species, has also led to geographic population structure within species. In particular, Stygiocaris lancifera is split into northern and southern groups, which correspond to population splits within other sympatric subterranean taxa.
Citation: Page TJ, Humphreys WF, Hughes JM (2008) Shrimps Down Under: Evolutionary Relationships of Subterranean Crustaceans from Western Australia (Decapoda: Atyidae: Stygiocaris). PLoS ONE 3(2): e1618. doi:10.1371/journal.pone.0001618
Academic Editor: Sudhindra Gadagkar, University of Dayton, United States of America
Received: October 28, 2007; Accepted: January 18, 2008; Published: February 20, 2008
Copyright: © 2008 Page et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Collections of Stygiocaris were made under funding from the Butler Trust (WA Museum), Western Australian Museum, National Estate Grants Scheme, Australian National Parks and Wildlife Service, West Australian Petroleum Pty Ltd and the Department of Defence. Field work in the Gibson Desert was supported by the Waterhouse Club, South Australian Museum. Lab work was funded by the Australian Rivers Institute, Griffith University. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
By their very nature, caves and other subterranean environments are poorly known, and yet they are home to large numbers of endemic species and unique relictual taxa from an earlier age  (“wrecks of ancient life” in the words of Charles Darwin ). The various evolutionary and geographic patterns displayed by subterranean biota can illuminate the complex processes and histories that have resulted in this biodiversity , , in much the same way as has been demonstrated for isolated oceanic islands . The isolation and strong selective pressures inherent in the adoption of an underground life can lead in polar opposite directions, namely both genetic divergence and morphological convergence , which can greatly confound interpretation –. Darkness, low energy inputs and many other common factors in subterranean environments often lead to very different animals evolving similar traits, such as atrophied eyes and a translucent body . Once these “troglomorphies” have arisen, a species is presumed to have a limited dispersal ability owing to its highly structured and isolated environment and the very narrow range of habitat to which it is adapted , .
Australia was once thought to have few areas of interest for those with a bent towards the underworld , , but this has changed dramatically in recent decades, particularly in Western Australia , , . The Precambrian landscape of inland Western Australia's Pilbara and Yilgarn Cratons (the “Western Shield”, Fig. 1) is one of the oldest and most stable on earth . Beneath this parched, ancient landscape, many independent “calcrete” (terrestrial limestone) aquifers have formed by precipitation of thin layers of carbonates along old drainage lines . A number of molecular studies of various taxa (diving beetles, amphipods, bathynellacea , , , ) of this “subterranean archipelago”  have confirmed the heterogeneous evolutionary origin of its endemic fauna and the isolated island nature of the environment referred to above.
Figure 1. Map of collection locations of subterranean atyids in the Australian region.doi:10.1371/journal.pone.0001618.g001
A highly diverse and distinct subterranean fauna is also found on the western edge of the Pilbara Craton, in the younger limestones of the Cape Range peninsula and Barrow Island (Fig. 2) . The evolutionary relationships of this biota have not yet been studied using DNA sequences, but earlier small-scale studies have used allozymes – and found the presence of cryptic species and localised geographic structuring. In some ways, the Cape Range area is potentially even more interesting than the calcretes of the Western Shield, because its flooded coastal limestone caves and fissures (karst) are rare in Australia, and because, in contrast to the geological stability of the Western Shield, the Cape Range area has been geologically dynamic , , which can lead to high levels of biodiversity .
Figure 2. Map of Stygiocaris collection locations in the Cape Range peninsula and Barrow Island, northwestern Australia.
See Tables 1 for site details.doi:10.1371/journal.pone.0001618.g002
Today a low, arid, mountain range (~300 m high) runs north-south along the spine of the tropical Cape Range peninsula (80 kms by 20 kms) (Fig. 2), fringed on the western and northern sides by the Ningaloo Coral Reef. Until the Tertiary, this whole area was covered with a warm, shallow sea in which limestones were formed against the coastline of the Western Shield . Upfolding of the limestones in the Miocene formed a series of anticlines, two of which comprise the Cape Range, originally possibly an island –, and Barrow Island, 170 km to the northeast (Fig. 2) . Exposure of the raised limestone to solution by mildly acidic rainwater formed caves and gorges, characteristic of karst terrain , . This was especially so in the middle of the limestone sequences, the Tulki Limestone, that is now highly cavernous . Lower sea levels during the Pleistocene (50–150 m lower than present) would have exposed a 12 km wide plain to the west of Cape Range, and a continuous plain between northern Cape Range and Barrow Island. This plain was intersected by several rivers (Ashburton, Cane, Robe) which were similar in nature to those bordering the Cape Range peninsula today .
The subterranean habitats of the Cape Range area are largely of two types . Firstly, there are those within the range itself, which are either dry or have small perched aquifers . The biota here is relictual, largely related to distant Australian terrestrial humid-forest species which probably retreated underground as the climate became arid in the Miocene , , as did the fauna of the Western Shield . The second habitat is an extensive flooded underground karst wetland, found on the narrow, flat coastal plain (up to 2 km wide; Fig. 3a) of the peninsula and extending under the foothills. This habitat also occurs on Barrow Island . This coastal fauna is aquatic and unrelated to that of the range , and elements of it also occur on the poorly known Pilbara coastal strip to the north-east , . The coastal aquifer is in the form of a wedge of sea water which intrudes beneath freshwater. This aquifer probably varies with location (major gorges discharge to the coast , ) and with the episodic recharge of the aquifer in this region affected by tropical cyclones , . In consequence, salinity levels tend to be lower at the top of the water column and with increasing distance from the coast but vary between freshwater (cave C-215) to full seawater (deeper parts of Bundera Sinkhole, C-28). Tidal fluctuation in the groundwater progressively decreases inland of the coast but may still be 10% of ocean tides 1.7 km inland .
Figure 3. Photos of a) Cape Range and b) Stygiocaris stylifera.
Aerial photo by WFH, and shrimp photo courtesy of Dr. Danny Tang.doi:10.1371/journal.pone.0001618.g003
Coastal waterbodies with subterranean connections to the sea and tidal influences are known as “anchialine” , and are found throughout the tropics and sub-tropics, often on arid coasts . These habitats are characterised by variable salinity and light, and can be thought of as “groundwater estuaries” . The fauna found in anchialine habitats are often highly disjunct, relictual taxa, often thought to descend from shallow marine populations of the Tethys Sea, which once separated the ancient supercontinents of Gondwana and Laurasia , , , implying vicariance by continental drift. The Tethyan Seaway between the Mediterranean and Arabian Seas allowed frequent marine faunal interchange between the Atlantic and Indo-West Pacific regions until at least 19 million years ago , and possibly as recently as 10 million years ago .
The coastal plain habitat of Cape Range supports a fauna of a type unknown elsewhere in the southern hemisphere, also occurring in similar anchialine habitats on either side of the North Atlantic (Yucatán, Mexico, islands of northern Caribbean, and Canary Islands). Although most of this fauna is known only in Australia from the Bundera Sinkhole, the sole accessible deep anchialine system, elements of the fauna occur widely in the more superficial waters of the linear Cape Range coastal plain, Barrow Island and the Pilbara coast (e.g. Stygiocaris Holthuis, 1960 (Atyidae), Halosbaena Stock, 1976 (Thermosbaenacea), Haptolana Bowman, 1966 (Cirolanidae), various amphipods –, , and Australia's only subterranean fishes, a blind eel and gudgeon ). In contrast to the origins of the taxa from both the Cape Range itself and from the Western Shield, the evolutionary origins of the taxa from the coastal plain habitat are more obscure and likely to be marine.
Crustaceans are common in subterranean habitats , , , and shrimps from the family Atyidae are frequently found in both major types of anchialine habitat, namely those of continental coasts (“remipede-type”) and seamount islands (“procarid-type”) , . The Atyidae are an ancient group of decapod shrimps comprising dozens of genera spread throughout the globe and largely occurring in freshwaters . Although many atyid species have brackish water larval phases , there are no known extant marine relatives , which is a common situation for anchialine species , , .
The Cape Range coastal plain and Barrow Island host two endemic species of atyid shrimp, Stygiocaris lancifera Holthuis, 1960 and S. stylifera Holthuis, 1960 (Fig. 3b). Stygiocaris is potentially an ideal window into the evolution and biogeography of this biota because 1) it is widespread within the coastal plain and Barrow Island , 2) the two species are largely separated east and west by the range (Fig. 2), and so small scale microevolutionary and phylogeographic patterns within the area may be discernible, and 3) large-scale evolutionary and biogeographic patterns (which are rarely studied for groundwater species ) may also be reconstructed because, although Stygiocaris is endemic to the area, its hypothesized closest evolutionary relatives are all geographically remote, namely Typhlopatsa Holthuis, 1956 from Madagascar  and other genera from the atyid sub-family Typhlatyinae ,  (Antecaridina Edmondson, 1954: Indo-Pacific ; Typhlatya Creaser, 1936: Caribbean, Europe , ).
Molecular phylogenetic and clock analyses can help to cut through the confusion of morphological convergence to help to resolve the evolutionary and biogeographic history of subterranean taxa –. As subterranean atyids have been the subject of a number of molecular studies in Europe , the Caribbean , , the Pacific ,  and Australia , as have surface atyids (e.g. –), there is an excellent context within which to place new data to test evolutionary hypotheses of the Western Australian stygobites.
Previous studies of Australian  and European atyid  genera have found that subterranean species are often related to local surface species rather than other distant “congeneric” cave dwellers. Molecular studies of anchialine atyids have inferred at least occasional marine dispersal within an evolutionary timeframe , ,  and a study of an anchialine snail found large amounts of recent gene flow over large ranges (>200 km) . This implies that an ancient “Tethyan” vicariant explanation may not be required to explain the presence of Stygiocaris in northwestern Australia. Therefore, we hypothesized that Stygiocaris will be an evolutionary offshoot of one of the many surface or anchialine species of the Indo-Pacific or Australia, particularly as coastal Western Australia hosts many undescribed surface atyids ( and TJP unpublished data). At the smaller scale, we hypothesized that the geological dynamism of the area will be reflected in the local structuring of Stygiocaris populations into isolated, geographically distinct biological units, much as it has done for other groundwater fauna of the area .
Dense populations of Stygiocaris (138 m2) have been recorded in favourable caves  and they access a wide range of food resources . Atyids are a major component of many surface waters , where they consume fine organic matter and biofilms . These are characteristic energy sources in groundwater, and so Stygiocaris are expected to be similarly pivotal to understanding the dynamics of the anchialine system.
We collected specimens of the Stygiocaris species from throughout their distributions on the Cape Range peninsula and Barrow Island, Western Australia (Fig. 2, Table 1). Specimens were collected at various sites using a number of methods. We used hand or plankton haul nets (mesh size 125–350 µm) within caves, historical pastoral wells, oil field anode protection bores, water supply and monitoring bores in town, and Defence establishment water supply aquifers. Within the Bundera Sinkhole (an anchialine cave ), cave divers towed a net below the hydrogen sulphide layers. As these sites are remote and difficult to access, collection numbers of Stygiocaris are generally small  and only a limited number could be analysed per site (Table 1).
Table 1. Western Australian Stygiocaris specimens and sequences.doi:10.1371/journal.pone.0001618.t001
For context, we also included specimens of all the known subterranean atyid shrimp species in Australia (Fig. 1, Table 2). A number of these taxa have not been previously sequenced, including the Western Australian genus Pycneus Holthuis, 1986 (Gibson Desert), the Western Australian species Caridina spelunca Choy, 1996, and the Northern Territory species Parisia gracilis Williams, 1964. Also included were a number of anchialine species from the Indo-Pacific (Antecaridina spp., Halocaridina Holthuis, 1963), and various epigean and subterranean species from throughout the world. Where possible, we integrated published sequences of numerous atyid species into our datasets (Table 2). Specimens of various taxa were kindly provided to us by many museums, institutions and individuals (Tables 1, 2).
Table 2. Additional specimens and sequences included in Worldwide 16S and H3 analyses.doi:10.1371/journal.pone.0001618.t002
Genomic DNA was extracted, amplified and sequenced as per . Two mitochondrial genes and one nuclear gene were targeted. The mitochondrial large subunit 16S ribosomal DNA (16S) was chosen because it is effective for both higher and lower systematic level phylogenetics  and is the best represented gene on GenBank for the Atyidae (August 2007). 16S was sequenced for all specimens as per  (Tables 1 and 2 for all GenBank accession numbers). We also sequenced a subset of the Stygiocaris specimens for the more quickly-evolving mitochondrial gene cytochrome oxidase subunit I (COI), which is effective at discriminating at the population and species-level , and is the locus favoured in the push for “DNA barcoding” (see ). For the COI amplification, we used primers CR-COI-F (5′-CWA CMA AYC ATA AGA YAT TGG-3′) and CR-COI-R (5′-GCR GAN GTR AAR TAR GCT CG-3′) . For a conserved nuclear gene, we sequenced Histone (H3) as per , who show it is informative for deep-level decapod phylogenetics. We sequenced the H3 gene for Stygiocaris spp. and any atyid species that fell within a higher-level 16S clade with it.
We assembled a number of separate datasets with various combinations of genes and taxa to investigate different phylogenetic levels (Table 3). Sequences were trimmed separately for each dataset so sequences for all terminal taxa were the same length. Two of the datasets (“Atyid 16S”, “Stygiocaris 16S”) include only 16S sequences. The Atyid 16S dataset includes 13 genera of atyids (10 of which have subterranean species) from throughout the world, and includes all eight genera found in Australia. The Stygiocaris 16S dataset includes sequences from all of our specimens of Stygiocaris from nine sampling sites. The 16S sequences from these two datasets were aligned using ClustalX version 1.81  at default settings, and Gblocks version 0.91b  was used to identify poorly aligned sites, which were excluded from analyses. The sequences from the remaining datasets were aligned with ClustalX as above, with no sites excluded.
Table 3. Different datasets, molecular models and tree scores for analyses conducted in this study.doi:10.1371/journal.pone.0001618.t003
The “Histone” dataset incorporates conserved nuclear H3 sequences from eight atyid genera, which were also analysed in combination with the relevant 16S sequence (“Combined 16S/H3”). “Stygiocaris Combined” includes H3, 16S and COI sequences from the three Stygiocaris species (and major intraspecific groupings) with Antecaridina sp. East Timor as an outgroup. Each gene region was analysed separately by gene as well as combined. We aligned H3 and COI sequences without gaps. Two further datasets of all available Stygiocaris 16S and Stygiocaris COI sequences respectively were also assembled to create haplotype networks and to derive genetic divergence estimates within and between species.
We used Modeltest version 3.06  to select the Akaike Information Criterion best-fit model of evolution for each dataset separately. For the Stygiocaris Combined and Combined 16S/H3 Datasets, an appropriate model was selected for each gene, as well as for all genes combined. Three forms of phylogenetic analysis were employed. We used PHYML version 2.4.4  for maximum likelihood analyses, MrBayes version 3.1.2  for Bayesian analyses (parameters: 2 million generations, trees sampled every 100 cycles, datasets partitioned by gene where appropriate, 50% burn in, two runs of four chains heated to 0.2), and PAUP* version 4.0 b10  for parsimony analyses (full heuristic with 100 random repetitions). Maximum likelihood and parsimony analyses were bootstrapped 1000 times.
Phylogenetic hypotheses of the sister taxon of Stygiocaris in the Combined 16S/H3 dataset were investigated using the Shimodaira–Hasegawa (S–H) test in PAUP (1000 replicates of resampling of estimated log-likelihood test distribution) and Bayes Factors  in MrBayes (constrained versus unconstrained harmonic means of likelihood values ).
Haplotype networks were constructed separately for all Stygiocaris 16S and COI sequences using TCS version 1.21 . COI and 16S sequence divergences within and between Stygiocaris species were calculated using a correction for within-group polymorphism  (and ±S.E.). Distance matrices were constructed in PAUP* using both uncorrected and corrected divergences (using the suggested models of molecular evolution from Modeltest).
Molecular clock calculations
Because of the geological dynamism of the Cape Range area, there is likely a close relationship between geological and biological events , . If we accept the likelihood that the emergence of the Cape Range Anticline in the Miocene isolated the ancestors of Stygiocaris lancifera and S. stylifera, leading to their speciation , then we can use this event as a calibration point to estimate rates of molecular divergence for these taxa. This geological event is dated to the Miocene 7–10 million years ago (MYA) , .
As molecular clock calculations are often contentious, in particular for cave species , we used two methods of calculating molecular divergence rates. Firstly, we used a simple distance method, using the various COI and 16S sequence divergences between S. lancifera and S. stylifera referred to just above. We assumed the S. lancifera/stylifera split occurred 7–10 MYA and applied the derived rates from this single node to the divergence between S. lancifera/stylifera and S. sp. Bundera. Secondly, we used a relaxed (uncorrelated lognormal) molecular clock method  as implemented in BEAST version 1.4.6 . We did two independent runs of BEAST (chain length of 10,000,000; sampled every 1000; Yule speciation process; 10% burn in) and combined the results with Tracer version 1.4  to calculate the time to most recent common ancestor (tmrca) for two nodes, Typhlatya/Stygiocaris spp. and S. lancifera/stylifera/sp. Bundera, by fixing the S. lancifera/stylifera tmrca to fall within the 7–10 million year range.
Pycnisia bunyip Suzuki & Davie, 2003 did not produce usable sequences. The Modeltest-derived models and tree scores from maximum likelihood, Bayesian and parsimony analyses for all phylogenetic datasets appear in Table 3.
Higher-level relationships of Australian subterranean species
At the largest phylogenetic scale (Fig. 4), the atyids fell into two higher-level clades, the “Atyinae” and “non-Atyinae” (sensu , ). Australian subterranean species were found within both groups. There were three distinct taxa within Stygiocaris, implying the presence of a cryptic species (Stygiocaris sp. Bundera). The sister to Stygiocaris was the Mexican cave shrimp Typhlatya pearsei Creaser, 1936. The anchialine Halocaridina and Antecaridina (which also has an undescribed species) were recovered within a clade containing both Stygiocaris and Typhlatya. Interestingly, all of the species in this “Typhlatyinae” clade (sensu , ) (Fig. 4) have subterranean proclivities. The other major clade within the Non-Atyinae (“Paratyinae” sensu , ) has both cave and surface species, although, interestingly, both of these two surface species of Paratya Miers, 1882 have also been reported from caves , .
Figure 4. Maximum likelihood phylogram of mitochondrial Atyid 16S dataset.
Australian subterranean species in bold. Branches with support >75% for all forms of analysis have thicker lines (Maximum Likelihood bootstrap values/Bayesian posterior probabilities above node, and Parsimony bootstrap values below).doi:10.1371/journal.pone.0001618.g004
All other Australian subterranean species fall within the “Caridina-like” group. As previously found , the cave species Parisia unguis Williams, 1964 and Pycnisia raptor Bruce, 1992 form a strong clade. Similarly, the other species of Parisia Holthuis, 1956 in Australia, Parisia gracilis, forms a clade with another cave genus, Pycneus, but both sets of clades group with completely separate surface species of Caridina H. Milne Edwards, 1837. This implies an independent colonisation of the underworld in Australia and morphological convergence. The DNA sequence of the Western Australian cave species, Caridina spelunca, is nearly identical to a previously unidentified surface species from the same area (Caridina sp. WA1; ), implying that these are conspecific and that Caridina spelunca is a troglophile, namely, a facultative subterranean inhabitant (as suggested by ).
Alternative explanations for some of these mitochondrial relationships are long-branch attraction and mitochondrial introgression , and so we also sequenced the highly conserved nuclear Histone (H3) gene for a sub-set of taxa. This nuclear Histone dataset (Fig. 5) recovered very similar relationships to the larger mitochondrial 16S dataset (Fig. 4), which implies that the evolutionary relationships recovered with 16S may accurately reflect species history, and not merely the organelle history. Stygiocaris stylifera and S. lancifera are closely related (sharing a haplotype) relative to Stygiocaris sp. Bundera, which is distinct. In this nuclear dataset, Typhlatya was again recovered as the sister to Stygiocaris, and both again formed a clade with Halocaridina and Antecaridina. As in the 16S dataset, Parisia gracilis, Pycneus and Caridina sp. LE are very closely related (with P. gracilis and Pycneus sharing an H3 haplotype). When the H3 and 16S sequences were combined and analysed as above (not displayed), the topology was identical to the H3-only dataset (Fig. 5). Support values for the combined H3/16S sequences were higher than in the analyses of H3-only for the Stygiocaris/Typhlatya relationship (Maximum Likelihood/Bayesian/Parsimony: 78/79/75), and comparable for the Stygiocaris/Typhlatya/Halocaridina relationship (66/75/88).
Figure 5. Maximum likelihood phylogram of nuclear Histone dataset.
Australian subterranean species in bold. Branches with support >75% for all forms of analysis have thicker lines (Maximum Likelihood bootstrap values/Bayesian posterior probabilities above node, and Parsimony bootstrap values below).doi:10.1371/journal.pone.0001618.g005
Every best-scoring topology from all three forms of analysis of the three relevant datasets (Atyid 16S, Histone, Combined 16S/H3) recovered Typhlatya as the sister taxon to Stygiocaris. But, as Halocaridina or Antecaridina spp. are also potential sisters of Stygiocaris in place of Typhlatya (see Figs. 4, 5), we calculated tree likelihoods for topologies constrained to either Halocaridina or Antecaridina as a sister to Stygiocaris. We then compared them to unconstrained trees (in which Typhlatya was always sister to Stygiocaris). Using Bayes Factors, the evidence against a hypothesis of Halocaridina as a sister was “Substantial” (; 2×differences in logs = 6.12) (unconstrained harmonic mean marginal likelihood = –4122.84; Halocaridina constrained likelihood = –4125.90). The evidence against an Antecaridina sister was “Very Strong” (2×logs = 15.10) (Antecaridina constrained likelihood = –4130.39). The S-H test could not reject either Halocaridina or Antecaridina as potential sisters to Stygiocaris at the 0.05 level but would reject them both at the 0.10 level (Halocaridina P = 0.069; Antecaridina P = 0.085).
The two described Stygiocaris species were recovered as a clade. A third Stygiocaris taxon was only found at site C-28 (Bundera Sinkhole). The three species are ~13% divergent from each other at COI (~5% at 16S)(uncorrected). The relationship between the three Stygiocaris species is not clear in the two 16S datasets (Figs. 4, 6), and so we also sequenced representatives of the major intraspecific groupings within each of the three species for the conserved nuclear H3 gene and the more rapidly evolving mitochondrial COI gene. This Stygiocaris combined dataset unequivocally recovers S. lancifera and S. stylifera as sister taxa relative to Stygiocaris sp. Bundera in each individual gene analysis and when the three analyses are combined (Fig. 7).
Figure 6. Maximum likelihood phylogram of Stygiocaris 16S dataset.
Branches with support >75% for all forms of analysis have thicker lines (Maximum Likelihood bootstrap values/Bayesian posterior probabilities above node, and Parsimony bootstrap values below).doi:10.1371/journal.pone.0001618.g006
Figure 7. Maximum likelihood phylogram of Stygiocaris Combined dataset (H3/16S/COI).
Branches with support >75% for all forms of analysis have thicker lines (Maximum Likelihood bootstrap values/Bayesian posterior probabilities above node, and Parsimony bootstrap values below). Also, individual gene maximum likelihood phylograms. Outgroup (Antecaridina sp. East Timor) not displayed.doi:10.1371/journal.pone.0001618.g007
Intraspecific groupings within Stygiocaris species
There is significant intraspecfic diversity within S. stylifera and S. lancifera visible in the Stygiocaris 16S, COI and combined datasets (Figs. 6, 7). Within S. stylifera there are three subspecific groups, which are ~6% divergent from each other in their COI sequences (~2% 16S). One of these three groups was only found on the eastern side of the Cape Range in the Defence Bores. The second group was only found at C-25 on the western side of the Cape, and the third group at both C-25 and Barrow Island.
There were two groups within S. lancifera which are ~2% divergent at COI (~1% 16S). These groups are geographically structured (Fig. 8), with a southern (C-215, C-149, C-274) and northern group (C-273, C-25).
Figure 8. COI and 16S haplotype networks for Stygiocaris lancifera placed in Cape Range geographic context.doi:10.1371/journal.pone.0001618.g008
Rates of molecular evolution and divergence estimates
There is a range of molecular clock calculations available, depending on whether one uses more modern relaxed clock methods or fairly simplistic distance estimates, which will vary based on whether one uses model-corrected or uncorrected distances for either COI or 16S sequences. When we marry the distance estimates with the geological estimates, we can derive rates of molecular evolution for particular nodes. These give a wide range of possible divergence rates per million years (±S.E.) for the S. lancifera/stylifera node (COI: 1.33–5.16%; 16S: 0.55–1.03%). Applying these rates, the common ancestor of S. lancifera/stylifera and Stygiocaris sp. Bundera diverged 6.15–12.30 MYA (±S.E.) for COI and 9.39–16.61 MYA for 16S.
Using a relaxed clock method on the Atyid 16S dataset, the time to most recent common ancestor for S. lancifera/stylifera/sp. Bundera is comparable to that above, with a mean of 11.08 MYA (7–17.94 MYA for 95% highest posterior density [HPD]; effective sample size [ESS] = 1612.12). The mean tmrca for Stygiocaris spp./Typhlatya is 24.53 MYA (11.11–41.54 MYA 95% HPD; 1449.00 ESS).
The mean 16S divergence rate resulting from the relaxed method is 0.75% per million years (0.33%–1.25% for 95% HPD; 1126.55 ESS), which is also comparable to the point estimates above for the S. lancifera/stylifera node. Commonly used general interspecific rates for crustaceans are 0.65%  and 0.9%  for 16S and 1.25%  or 1.4%  for COI. Our interspecific COI rates are considerably slower than recently inferred intraspecific rates for populations of the related atyid (Halocaridina, ~20% per million years ). This may be a result of hypothesized differences in rates of molecular divergence between more recent intraspecific datasets and more ancient interspecific ones (see , ).
Western Australian cave atyids at the global scale
An earlier study of Australian atyids  has shown that some subterranean species (Parisia unguis, Pycnsia raptor) descend from localised surface species rather than morphologically similar subterranean species from further afield, and that formal systematic classifications often do not agree with inferred evolutionary relationships. We found a similar pattern in the present study for two further cave species (Parisia gracilis, Pycneus moritans Holthuis, 1986). However, importantly, we did not find this pattern for Stygiocaris spp., which formed no clades with any Australian taxa. Stygiocaris formed a clade with subterranean species (largely from the subfamily Typhlatyinae, Fig. 4) found over a very large area, thus refuting our hypothesis that Stygiocaris would follow a pattern similar to other Australian subterranean atyids.
Interestingly the nearest relation to Stygiocaris in both our mitochondrial and nuclear data is the subterranean genus Typhlatya, whose centre of diversity is the Caribbean/North Atlantic/Mediterranean . This pattern fits closely with the hypothesis that certain widespread disjunct anchialine species have descended from marine species from the Tethys Sea, whose disjuncture can be explained by sea floor spreading due to plate tectonic movement , , . The severing of migration routes caused by the closure of the Tethys Seaway about 19 MYA cut the link between the previously closely related marine faunas on either side , . Our mean estimate of a common ancestor of Stygiocaris and Typhlatya at 24.53 MYA (11.11–41.54 MYA) is congruent with a “Tethyan Track” explanation .
Stygiocaris is by no means the only Cape Range representative of this pattern, as the Bundera Sinkhole (site C-28) hosts species of remipede crustaceans , thaumatocypridid ostrocods  and certain copepods  (and the thermosbaenacean genus Halosbaena from C-215, Poore & Humphreys 1992) found nowhere else in the Southern Hemisphere, whose sister taxa are instead in coastal caves of the North Atlantic (Caribbean Islands, Yucatán, Bermuda, Canary Islands) .
One interesting point is that the closure of the Tethys Seaway (19–10 million years , ) is older than the formation of the Cape Range (10–7 million years , ). This is likely explained by either 1) the ancestors of Stygiocaris colonising the coast of the nearby coast of Pilbara Craton first and later moving to the Cape Range area , , or 2) local marine species actively colonising or being stranded in newly emergent limestone habitat (“regression”; ) and their marine ancestors becoming extinct  or remaining unsampled . A similar scenario to the second explanation is envisioned for the Stygiocaris's sister taxon, Typhlatya, that are hypothesised to have lived in marine caves in the Caribbean before the formation of their current anchialine habitats .
A true evolutionary history of any species is unlikely to conform to a single, simple idea, such as Tethyan vicariance, and a more complex “mixed-model” approach is likely to better reflect reality, as found in many studies of subterranean crustaceans ,  . The Stygiocaris/Typhlatya relationship on its own is only weak evidence for an imprint of the Tethys, especially as other atyids have proven to disperse over great distances , , , , but similar “Tethyan” patterns amongst sympatric taxa imply that this idea must be considered at least feasible at the large scale. This will be testable as more molecular data from sympatric taxa becomes available . It is also likely a question of scale , , with Tethyan vicariance responsible for some global distributions, but overlaid with both long and short distance dispersal and vicariance , , .
Diversity of Stygiocaris species
If the evolutionary relatives of the genus Stygiocaris lie in a far off land, there is no doubt that the various Stygiocaris species themselves are endemic to northwestern Australia. The two described Stygiocaris species are largely found on opposite sides of the Cape Range, with S. lancifera on the western side and S. stylifera on the east and Barrow Island, with a few sites of sympatry in the northwest of the peninsula (Fig. 2) . Both species are highly troglobitic (transparent, reduced eyes, <20mm ; Fig. 3b). The two species are difficult to distinguish from each other , , , , but their evolutionary differentiation has been confirmed using allozymes , , DNA (present study) and recent morphological work (WFH and Dr Danny Tang).
The east/west split between species may be the result of isolation by the orogeny of the range, with limited secondary contact in the northwest, probably during epochs of lowered sea-levels . Secondary contact also likely explains the sympatry of some Typhlatya species in Mexico ,  and ostrocod species of the genus Danielopolina Kornicker & Sohn, 1976 in the Canary Islands  . The sympatric blind gudgeon Milyeringa veritas Whitley, 1945 also shows a split between east and west with a site of overlap in the north , . This divergence is at a smaller, intraspecific level, which may also imply a persistent dispersal barrier near the northern tip of the peninsula , limiting dispersal between east and west for most stygofauna.
Our molecular data suggest the presence of a third, cryptic, species of Stygiocaris at the Bundera Sinkhole (site C-28, Fig. 2). There is a proliferation of cryptic species in subterranean habitats both because of limited study due to difficult access and rampant morphological convergence in the face of strong, similar selective pressures , , . Cryptic species have been identified using molecular methods within other subterranean taxa in the Cape Range (millipedes , amphipods ) and Western Shield (amphipods , , Parabathynellidae ) and within other subterranean atyids (Typhlatya , ). This has meant that widespread groundwater “species” have proven to be complexes of cryptic species , highly compartmentalised by their underground landscapes . As for Stygiocaris sp. Bundera, these speciations have often been associated with very limited or undetected morphological differentiation . Species discovered with molecular methods can help to define species boundaries and foster further, targeted morphological study, which can go on to discover new morphological characters .
The Stygiocaris specimens from Bundera were originally identified as S. stylifera , but there has always been some doubt (WFH) due to their large size, Bundera's isolation from other Stygiocaris sites (>30 kms, Fig. 2), and the very high salinity water in which they occur (20,000–35,000 mg L−1 ; ), whereas other Stygiocaris species are only in fresh to brackish water (mean = 2064 mg L−1 TDS, sd 1753; range 290–7700, n = 26) . Bundera Cenote (sinkhole) is of particular interest as it is the only deep anchialine system in Australia (penetrating down to the underlying seawater) and the only continental one in the Southern Hemisphere . The water profile at Bundera is highly stratified, with warm sea water beneath a stable thermo-halocline . It contains Stygiocaris sp. Bundera (as well as the unique “Tethyan” community referred to above), which were only sampled well beneath the overlaying layer of brackish water, in water with a salinity similar to local seawater . This occurrence of different congeneric species in distinct physiochemical conditions is also found within Stygiocaris's sister, Typhlatya, which hosts both fully freshwater and brackish species , and copepods in European karst .
The brackish-seawater environment of Stygiocaris sp. Bundera may well represent the original lifestyle of the coastal marine Tethyan ancestor of Stygiocaris and Typhlatya, with some species subsequently invading and adapting to freshwater. The dispersal of freshwater species is limited by saline conditions, whereas Stygiocaris sp. Bundera may be limited by surrounding freshwater. There is considerable evidence, from the nature of the faunas and their distribution, that anchialine habitats are both geologically very old and persistent , ,  so providing a potentially persistent “Tethyan” time capsule of which Bundera is the present manifestation. These anchialine systems do not necessarily need to rely on allochthonous energy sources. Food web studies using stable isotopes of Stygiocaris (at Bundera  and Barrow Island ) and Typhlatya  have suggested and demonstrated respectively that they feed on sulphide-oxidising chemoautotrophic bacteria (which are common in anchialine systems , ), thus making these systems at least partially independent from the surface, similar to some deep-sea vent communities .
Intraspecific diversity at the local scale
Manifestations of the uplift and subsequent karst development in Cape Range are found in the subterranean fauna, both terrestrial and aquatic. The terrestrial cave fauna differs along the length of the range, and between the coast and the range, with one terrestrial species, a troglobitic micro-whipscorpion, found on the east coast plain and Barrow Island, like Stygiocaris . Norcapensis Bradbury & Williams, 1997, a melitid amphipod, inhabits perched groundwater at elevations of about 200 m in Cape Range, where three distinct allozyme populations occur ; melitids are a marine family that has invaded inland waters and those high in Cape Range probably uplifted with the range.
There is significant geographic intraspecific biodiversity within many groundwater species, visible within both described Stygiocaris species (Figs. 6, 7), as well as within other Cape Range stygofauna (amphipods , and gudgeons ), which confirms our second hypothesis. In particular, the populations within the linear habitat of S. lancifera are split into divergent northern and southern groups (Fig. 8) at a very similar point to geographic provinces of the other taxa mentioned above. This comparative phylogeographic approach strengthens the hypothesis that common events or dispersal barriers , such as sea-level changes , gorges cutting through to impermeable limestone layers , or hypersaline groundwater , has structured and isolated many of the populations into distinct geographic communities. This has conservation implications in that we will need to avoid any human mediated exchange of individuals between evolutionary distinct populations .
Conservation of groundwater communities
Our multi-scale molecular data have allowed us to identify unappreciated groundwater biodiversity at the alpha to gamma levels, ranging from localised population differences to ancient globe-spanning evolutionary relationships. Groundwater faunas generally are vulnerable to human impacts resulting variously from water extraction, the addition of substances to water, and changes to aquifer permeability. As these isolated and widely vicariant anchialine communities occur in tightly constrained coastal locations, they may be particularly vulnerable  or especially resilient , , , depending on the nature of the threat. This dichotomy of interpretation of the evidence, which needs to be addressed globally, may result from the extremely sharp gradients in physio-chemical conditions that may occur in anchialine habitats . Issues potentially or actually pertinent to the Cape Range/Barrow Island groundwater fauna include oil and gas field development , mining , water extraction or various uses and waste discharge , increasing salinity , and pollution .
Although Stygiocaris is only one component of the local biota, our data show that it is both an Australian element of the ancient Tethyan fauna and harbours cryptic species with very small ranges (possibly even a single site) and thus liable to extinction . Indeed, Bundera Sinkhole is listed as a specially protected community in Western Australia, and a number of the species, including S. lancifera, the remipede Lasionectes exleyi Yager & Humphreys 1996, and both species of cave fish, are specially protected under Commonwealth of Australian and/or Western Australian fauna legislation.
Subterranean habitats tend to have truncated , easily disrupted, food webs , within which Stygiocaris, may play a keystone role (and be a potential bioindicator ). This is because atyids are biofilm feeders that may act as a conduit between energy producing chemoautotrophic bacteria and vulnerable higher-trophic level vertebrate predators, such as the blind gudgeon . Because the basic biology of the diverse groundwater fauna of Australia is largely unknown , it is not possible to assess adequately their vulnerability to anthropogenic changes. As we add more information on their evolution and ecology, the arguments for the protection of all groundwater habitats are considerably strengthened .
Field support was provided variously by Brian Vine, Darren Brooks, Carl Close, Stefan Eberhard, Andrew Poole, and Julianne Waldock, Chris Watts and members of the Waterhouse Club, South Australian Museum. Collections were made under permits from the antecedents of the Western Australian Department of Environment and Conservation, and the Commonwealth Department of Environment and Water. We also thank four anonymous reviewers for greatly improving the manuscript
Conceived and designed the experiments: JH TP WH. Performed the experiments: TP. Analyzed the data: TP. Contributed reagents/materials/analysis tools: TP WH. Wrote the paper: JH TP WH.
- 1. Boulton AJ, Humphreys WF, Eberhard SM (2003) Imperilled subsurface waters in Australia: biodiversity, threatening processes and conservation. Aquatic Ecosystem Heath and Management 6: 41–54.
- 2. Darwin C (1859) On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life, 1st edition. London: reprinted by the Folio Society.
- 3. Hamilton-Smith E (2001) Maintenance of karst biodiversity, with an emphasis upon Australian populations. Records of the Western Australian Museum Supplement 64: 85–95.
- 4. Danielopol DL, Pospisil P, Rouch R (2000) Biodiversity in groundwater: a large-scale view. Trends in Ecology and Evolution 15: 223–224.
- 5. Cooper SJB, Bradbury JH, Saint KM, Leys R, Austin AD, et al. (2007) Subterranean archipelago in the Australian arid zone: mitochondrial DNA phylogeography of amphipods from central Western Australia. Molecular Ecology 16: 1533–1544.
- 6. Finston TL, Johnson MS, Humphreys WF, Eberhard SM, Halse SA (2007) Cryptic speciation in two widespread subterranean amphipod genera reflects historical drainage patterns in an ancient landscape. Molecular Ecology 16: 355–365.
- 7. Lefébure T, Douady CJ, Malard F, Gibert J (2007) Testing vicariance and cryptic diversity in a widely distributed groundwater amphipod (Niphargus rhenorhodanensis). Molecular Phylogenetics and Evolution 42: 676–686.
- 8. Proudlove G, Wood PJ (2003) The blind leading the blind: cryptic subterranean species and DNA taxonomy. Trends in Ecology and Evolution 18: 272–273.
- 9. Lefébure T, Douady CJ, Gouy M, Trontelj P, Briolay J, et al. (2006) Phylogeography of a subterranean amphipod reveals cryptic diversity and dynamic evolution in extreme environments. Molecular Ecology 15: 1797–1806.
- 10. Humphreys WF (2006) Aquifers: the ultimate groundwater-dependent ecosystems. Australian Journal of Botany 54: 115–132.
- 11. Trontelj P, Douady CJ, Fišer C, Gibert J, Gorički Š, et al. (in press) A molecular test for cryptic diversity in ground water: how large are ranges of macro-stygobionts? Freshwater Biology DOI: 10.1111/j.1365-2427.2007.01877.x.
- 12. Hamilton-Smith E (1967) The arthropoda of Australian caves. Journal of the Australian Entomological Society 6: 103–118.
- 13. Humphreys WF (in press) Rising from down under: developments in subterranean biodiversity in Australia from a groundwater fauna perspective. Invertebrate Systematics.
- 14. Humphreys WF (2001) Groundwater calcrete aquifers in the Australian arid zone: the context to an unfolding plethora of stygal biodiversity. Records of the Western Australian Museum Supplement 64: 63–83.
- 15. Leys R, Watts CHS, Cooper SJB, Humphreys WF (2003) Evolution of subterranean diving beetles (Coleoptera : Dytiscidae : Hydroporini, Bidessini) in the arid zone of Australia. Evolution 57: 2819–2834.
- 16. Guzik MT, Cooper SJB, Humphreys WF, Cho J-L, Austin A (in press) Phylogeography of the ancient Parabathynellidae (Crustacea: Bathynellacea) from the Yilgarn region of Western Australia. Invertebrate Systematics.
- 17. Humphreys WF (1993) Significance of the subterranean fauna in biogeographic reconstruction: examples from the Cape Range peninsula, Western Australia. Records of the Western Australian Museum Supplement 45: 165–192.
- 18. Humphreys WF, Adams M (1991) The subterranean aquatic fauna of the North West Cape Peninsula, Western Australia. Records of the Western Australian Museum 15: 383–411.
- 19. Adams M, Humphreys WF (1993) Patterns of genetic diversity within selected subterranean fauna of the Cape Range peninsula, Western Australia: systematic and biogeographic implications. Records of the Western Australian Museum Supplement 45: 145–164.
- 20. Humphreys WF, Adams M (2001) Allozyme variation in the troglobitic millipede Stygiochiropus communis (Diplopoda: Paradoxosomatidae) from the arid tropical Cape Range, northwestern Australia: population structure and implications for the management of the region. Records of the Western Australian Museum Supplement 64: 15–36.
- 21. Wyrwoll KH, Kendrick GW, Long JA (1993) The geomorphology and Late Cenozoic geomorphological evolution of the Cape Range-Exmouth Gulf Region. Records of the Western Australian Museum Supplement 45: 1–23.
- 22. Collins LB, Read JF, Hogarth JW, Coffey BP (2006) Facies, outcrop gamma ray and C-O isotopic signature of exposed Miocene subtropical continental shelf carbonates, North West Cape, Western Australia. Sedimentary Geology 185: 1–19.
- 23. Allen AD (1993) Outline of the geology and hydrogeology of Cape Range, Carnarvon Basin, Western Australia. Records of the Western Australian Museum Supplement 45: 25–38.
- 24. Humphreys WF (2001) The subterranean fauna of Barrow Island, northwestern Australia, and its environment. Memoires de Biospeologie 28: 107–127.
- 25. Sket B (1996) The ecology of anchialine caves. Trends in Ecology and Evolution 11: 221–225.
- 26. Humphreys WF (2001) Milyeringa veritas (Eleotridae), a remarkably versatile cave fish from the arid tropics of northwestern Australia. Environmental Biology of Fishes 62: 297–313.
- 27. Knott B (1993) Stygofauna from Cape Range peninsula, Western Australia: Tethyan relicts. Records of the Western Australian Museum Supplement 45: 109–127.
- 28. Humphreys WF (1999) Physio-chemical profile and energy fixation in Bundera Sinkhole, an anchialine remiped habitat in north-west Australia. Journal of the Royal Society of Western Australia 82: 89–98.
- 29. Humphreys WF, Danielopol DL (2006) Danielopolina (Ostracoda, Thaumatocyprididae) on Christmas Island, Indian Ocean, a sea mount island. Crustaceana 78: 1339–1352.
- 30. Sanz S, Platvoet D (1995) New perspectives on the evolution of the genus Typhlatya (Crustacea, Decapoda): first record of a cavernicolous Atyid in the Iberian Peninsula, Tyghlatya miravetensis n. sp. Contributions to Zoology 65: 79–99.
- 31. Harzhauser M, Kroh A, Mandic O, Piller WE, Gohlich U, et al. (2007) Biogeographic responses to geodynamics: A key study all around the Oligo-Miocene Tethyan Seaway. Zoologischer Anzeiger 246: 241–256.
- 32. McQuarrie N, Stock JM, Verdel C, Wernicke BP (2003) Cenozoic evolution of Neotethys and implications for the causes of plate motions. Geophysical Research Letters 30: art. no. 2036.
- 33. Lefébure T, Douady CJ, Gibert J (2006) Relationship between morphological taxonomy and molecular divergence within Crustacea: proposal of a molecular threshold to help species delimitation. Molecular Phylogenetics and Evolution 40: 435–447.
- 34. Maciolek JA (1983) Distribution and biology of Indo-Pacific insular hypogeal shrimps. Bulletin of Marine Science 33: 606–618.
- 35. Page TJ, von Rintelen K, Hughes JM (2007) Phylogenetic and biogeographic relationships of subterranean and surface genera of Australian Atyidae (Crustacea: Decapoda: Caridea) inferred with mitochondrial DNA. Invertebrate Systematics 21: 137–145.
- 36. Smith MJ, Williams WD (1981) The occurrence of Antecaridina lauensis (Edmondson) (Crustacea, Decapoda, Atyidae) in the Solomon Islands. An intriguing biogeographical problem. Hydrobiologia 85: 49–58.
- 37. Culver DC, Pipan T, Schneider K (in press) Vicariance, dispersal and scale in the aquatic subterranean fauna of karst regions. Freshwater Biology DOI:10.1111/j.1365-2427.2007.01856.x.
- 38. Holthuis LB (1960) Two new species of atyid shrimps from the subterranean waters of N.W. Australia (Decapoda, Natantia). Crustaceana 1: 47–57.
- 39. Holthuis LB (1993) The recent genera of the Caridean and Stenopodidean shrimps (Crustacea, Decapoda) with an appendix on the order Amphionidacea. Leiden: Nationaal Natuurhistorisch Museum.
- 40. Hunter R, Webb MS, Iliffe TM (2008) Phylogeny and historical biogeography of the cave-adapted shrimp genus Typhlatya (Atyidae) in the Caribbean Sea and western Atlantic. Journal of Biogeography 35: 65–75.
- 41. Trontelj P, Gorički S, Polak S, Verovnik R, Zakšek V, et al. (2007) Age estimates for some subterranean taxa and lineages in the Dinaric karst. Acta Carsologica 36: 183–189.
- 42. Pohlman JW, Cifuentes LA, Iliffe TM (2000) Food web dynamics and biogeochemistry of anchialine caves: a stable isotope approach. In: Wilken H, Culver DC, Humphreys WF, editors. Ecosystems of the world 30 Subterranean ecosystems. Amsterdam: Elsevier. pp. 345–357.
- 43. Moritz C, Cicero C (2004) DNA barcoding: promise and pitfalls. PLoS Biology 2: 1529–1531.
- 44. Zakšek V, Sket B, Trontelj P (2007) Phylogeny of the cave shrimp Troglocaris: evidence of a young connection between Balkans and Caucasus. Molecular Phylogenetics and Evolution 42: 223–235.
- 45. Webb MS (2003) Intraspecfic relationships among the stygobitic shrimp Typhlatya mitchelli, by analyzing sequence data from mitochondrial DNA [Masters Thesis]. College Station, TX: Texas A&M University.
- 46. Santos SR (2006) Patterns of genetic connectivity among anchialine habitats: a case study of the endemic Hawaiian shrimp Halocaridina rubra on the Island of Hawaii. Molecular Ecology 15: 2699–2718.
- 47. Craft JD, Russ AD, Yamamoto MN, Iwai TY Jr, Hau S, et al. (2008) Islands under islands: the phylogeography and evolution of Halocaridina rubra Holthuis, 1963 (Crustacea: Decapoda: Atyidae) in the Hawaiian archipelago. Limnology and Oceanography 53: 675–689.
- 48. Page TJ, von Rintelen K, Hughes JM (2007) An Island in the stream: Australia's place in the cosmopolitan world of Indo-West Pacific freshwater shrimp (Decapoda: Atyidae: Caridina). Molecular Phylogenetics and Evolution 43: 645–659.
- 49. Page TJ, Baker AM, Cook BD, Hughes JM (2005) Historical transoceanic dispersal of a freshwater shrimp: the colonization of the South Pacific by the Genus Paratya (Atyidae). Journal of Biogeography 32: 581–593.
- 50. von Rintelen K, von Rintelen T, Meixner M, Lüter C, Cai Y, et al. (2007) Freshwater shrimp-sponge association from an ancient lake. Biology Letters 3: 262–264.
- 51. Kano Y, Kase T (2004) Genetic exchange between anchialine cave populations by means of larval dispersal: the case of a new gastropod species Neritilia cavernocola. Zoologica Scripta 33: 423–437.
- 52. Humphreys WF (1994) The Subterranean fauna of the Cape Range coastal plain, northwestern Australia. Perth, WA: Report by the Western Australian Museum to the Australian Heritage Commission and the Western Australian Heritage Committee.
- 53. Yam RSW, Dudgeon D (2006) Production dynamics and growth of atyid shrimps (Decapoda: Caridina spp.) in 4 Hong Kong streams: the effects of site, season, and species. Journal of the North American Benthological Society 25: 406–416.
- 54. Burns A, Walker KF (2000) Biofilms as food for decapods (Atyidae, Paleomonidae) in the River Murray, South Australia. Hydrobiologia 437: 83–90.
- 55. Cook BD, Pringle CM, Hughes JM (in press) Molecular evidence for sequential colonization and taxon cycling in freshwater decapod shrimps on a Caribbean island. Molecular Ecology DOI:10.1111/j.1365-294X.2007.03637.x.
- 56. Porter ML, Pérez-Losada M, Crandall KA (2005) Model-based multi-locus estimation of decapod phylogeny and divergence times. Molecular Phylogenetics and Evolution 37: 355–369.
- 57. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24: 4876–4882.
- 58. Castresana J (2000) Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution 17: 540–552.
- 59. Posada D, Crandall KA (1998) Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817–818.
- 60. Guindon S, Gascuel O (2003) A simple, fast and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52: 696–704.
- 61. Huelsenbeck JP, Ronquist F (2001) MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755.
- 62. Swofford DL (2002) PAUP * version 4.0b5-Phylogenetic Analysis Using Parsimony (* and other methods). Sunderland, MA: Sinauer Associates.
- 63. Kass RE, Raftery (1995) Bayes factors. Journal of the American Statistical Association 90: 773–795.
- 64. Ronquist F, Huelsenbeck JP, van der Mark P (2005) MrBAYES 3.1 Manual. Florida State University. pp. 1–69.
- 65. Clement M, Posada D, Crandall KA (2000) TCS: a computer program to estimate gene genealogies. Molecular Ecology 9: 1657–1660.
- 66. Avise JC (1994) Molecular markers, natural history and evolution. New York: Chapman & Hall.
- 67. Trontelj P (2007) The age of subterranean crayfish species. A comment on Buhay & Crandall (2005): subterranean phylogeography of freshwater crayfishes shows extensive gene flow and surprisingly large population sizes. Molecular Ecology 16: 2841–2843.
- 68. Drummond AJ, Ho SYW, Phillips MJ, Rambaut A (2006) Relaxed Phylogenetics and Dating with Confidence. PLoS Biology 4: e88, doi:10.1371/journal.pbio.0040088.
- 69. Drummond AJ, Rambaut A (2007) BEAST: Bayesian evolutionary analysis by sampling trees. BMC Evolutionary Biology 7:214: doi:10.1186/1471-2148-1187-1214.
- 70. Rambaut A, Drummond AJ (2007) TRACER: MCMC Trace Analysis Tool Version v1.4. Oxford: University of Oxford.
- 71. Williams WD (1964) Subterranean freshwater prawns (Crustacea: Decapoda: Atyidae) in Australia. Australian Journal of Marine and Freshwater Research 15: 93–106.
- 72. Choy SC (1996) Caridina spelunca, a new species of freshwater shrimp (Crustacea: Decapoda: Atyidae) from a Western Australian cave. Records of the Western Australian Museum 18: 103–107.
- 73. Schubart CD, Cuesta JA, Diesel R, Felder DL (2000) Molecular phylogeny, taxonomy, and evolution of nonmarine lineages within the American grapsoid crabs (Crustacea: Brachyura). Molecular Phylogenetics and Evolution 15: 179–190.
- 74. Sturmbauer C, Levinton JS, Christy J (1996) Molecular phylogeny analysis of fiddler crabs: test of the hypothesis of increasing behavioral complexity in evolution. Proceedings of the National Academy of Sciences of the United States of America 93: 10855–10857.
- 75. Ketmaier V, Argano R, Caccone A (2003) Phylogeography and molecular rates of subterranean aquatic Stenasellid Isopods with a peri-Tyrrhenian distribution. Molecular Ecology 12: 547–555.
- 76. Knowlton N, Weigt LA (1998) New dates and new rates for divergence across the Isthmus of Panama. Proceedings of the Royal Society of London Series B-Biological Sciences 265: 2257–2263.
- 77. Ho SYW, Shapiro B, Phillips MJ, Cooper A, Drummond AJ (2007) Evidence for time dependency of molecular rate estimates. Systematic Biology 56: 515–522.
- 78. Emerson BC (2007) Alarm bells for the molecular clock? No support for Ho et al.'s model of time-dependent molecular rate estimates. Systematic Biology 56: 337–345.
- 79. Hrbek T, Meyer A (2003) Closing of the Tethys Sea and the phylogeny of Eurasian killifishes (Cyprinodontiformes : Cyprinodontidae). Journal of Evolutionary Biology 16: 17–36.
- 80. Jaume D, Boxshall GA, Humphreys WF (2001) New stygobiont copepods (Calanoida; Misophrioida) from Bundera Sinkhole, an anchialine cenote in north-western Australia. Zoological Journal of the Linnean Society 133: 1–24.
- 81. Danielopol DL, Baltanas A, Humphreys WF (2000) Danielopolina kornickeri sp n. (Ostracoda, Thaumatocypridoidea) from a western Australian anchialine cave: morphology and evolution. Zoologica Scripta 29: 1–16.
- 82. Stock JH (1986) Caribbean biogeography and a biological calendar for geological events. In: Gore RH, Heck KL, editors. Crustacean Biogeography: Crustacean Issues 3. Rotterdam: A. A. Balkema. pp. 195–203.
- 83. Iliffe TM (2000) Anchialine cave ecology. In: Wilkens H, Culver DC, Humphreys WF, editors. Ecosystems of the World, vol 30 Subterranean Ecosystems. Amsterdam: Elsevier. pp. 59–76.
- 84. Seymour JR, Humphreys WF, Mitchell JG (2007) Stratification of the microbial community inhabiting an anchialine sinkhole. Aquatic Microbial Ecology 50: 11–24.
- 85. Humphreys WF (2000) The hypogean fauna of the Cape Range peninsula and Barrow Island, northwestern Australia. In: Wilkens H, Culver DC, Humphreys WF, editors. Ecosystems of the World, vol 30 Subterranean Ecosystems. Amsterdam: Elsevier. pp. 581–601.
- 86. Iliffe TM, Jickells TD, Brewer MS (1984) Organic pollution of an inland marine cave from Bermuda. Marine Environment Research 12: 173–189.
- 87. Sket B (1986) Ecology of the mixohaline hypogean fauna along the Yugoslav coasts. Stygologia 2: 317–337.
- 88. Gibert J, Deharveng L (2002) Subterranean ecosystems: a truncated functional biodiversity. Bioscience 52: 473–481.