SEARCH

SEARCH BY CITATION

Keywords:

  • 12S;
  • 16S;
  • 18S;
  • Black Sea;
  • Caspian Sea;
  • cytochrome c oxidase subunit I;
  • endemics;
  • marine Cladocera;
  • predatory Cladocera;
  • speciation;
  • zooplankton

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Members of the order Cladocera show remarkable morphological and ecological diversity. One of the most spectacular adaptive radiations in this group has involved species of the suborder Onychopoda, which have adopted a novel feeding strategy, predation, and have colonized habitats with a broad range of salinities. In order to evaluate the origins and systematics of this group, we derived a molecular phylogeny for its three component families including nine of 10 recognized genera based on three mitochondrial (mt) gene sequences: cytochrome c oxidase subunit I (COI), the ribosomal small and large subunits (12S and 16S) and one nuclear gene sequence: the small ribosomal subunit (18S). Maximum-parsimony, maximum-likelihood and neighbour-joining phylogenetic analyses were largely congruent, supporting the monophyly of the suborder and each of its families. Comparative analyses of data based on total evidence and the conditional combination of the ribosomal genes produced relatively congruent patterns of phylogenetic affinity. By contrast, analyses of single gene results were inconsistent in recovering the monophyletic groups identified by the multigene analyses. Based on the reconstructed phylogeny, we discriminate among the existing hypotheses regarding the evolutionary history of the onychopods. We identify a prolonged episode of speciation from the Miocene to the Pleistocene with two pulses of diversification. We discuss our results with reference to the geological history of the Ponto-Caspian basin, the region which fostered the onychopod radiation.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Members of the order Cladocera show remarkable morphological diversity (Fryer, 1987) and species comprising the suborder Onychopoda form one of the most morphologically distinctive groups of cladocerans. Their morphological radiation is linked to their predatory mode of feeding and their success in colonizing habitats with a wide range of salinities. This group includes 10 genera and about 33 species (Mordukhai-Boltovskoi, 1968; Rivier, 1998) which share a grasping mode of feeding in contrast to the filter feeding strategy employed by other cladocerans. They also have a novel reproductive system – their embryos are protected by a closed brood pouch, which secretes the nutrients necessary for their development (Egloff et al., 1997; Rivier, 1998). It has been suggested that these characteristics facilitated the transition of an ancestral freshwater cladoceran to the ocean, a ‘hostile’ habitat with scarce and intermittent food resources (Aladin & Potts, 1995; Rivier, 1998). One argument supporting this hypothesis is the fact that the only other marine cladoceran aside from the onychopods, the filter feeding Penilia avirostris (Sididae), has independently acquired a similar brood pouch.

Members of the Onychopoda have traditionally been assigned to three families: Cercopagidae, Podonidae, and Polyphemidae. The Polyphemidae is least diverse, including a single freshwater genus with two recognized species, one of which is restricted to the Caspian Sea. The family Cercopagidae is slightly more diverse as it includes two genera (Cercopagis and Bythotrephes) and about 14 described species (Rivier, 1998). All but one of these species were, until recently, restricted to the vicinity of the Black and Caspian Seas. The final family, the Podonidae, includes seven genera and 17 species. Three of these genera (Caspievadne, Cornigerius and Podonevadne) are restricted to the Ponto-Caspian basin, but the other four (Evadne, Pleopis, Podon, Pseudevadne) occur in the world's oceans (Onbe, 1999). The fact that most (72%) onychopod species are endemic to the Caspian, Azov and Aral Seas, or to the estuaries and lagoons of the Black Sea (Fig. 1) suggests that the Ponto-Caspian basin fostered much of the radiation within this group (Zenkevitch, 1963; Mordukhai-Boltovskoi, 1965; Dumont, 1998b). The onychopods are not the only group which diversified in this area. Similar radiations have occurred in many other Ponto-Caspian lineages, such as cumaceans, mysids, copepods, amphipods, decapods, mollusks and fishes (Mordukhai-Boltovskoi, 1979; Dumont, 1998a). However, the timing of these radiations and the extent of their diversification outside the Ponto-Caspian area remain uncertain. Unlike other ancient lakes, where adaptive radiations were restricted to benthic lineages, the Ponto-Caspian area fostered several episodes of radiation in planktonic lineages such as Mysis (Väinölä, 1995), the cyclopoid copepods (Monchenko, 1998) and the onychopods (Mordukhai-Boltovskoi, 1965). Most of these groups are euryhaline, which suggests that fluctuations in salinities, coupled with the fragmentation ofwaterbodies within the Ponto-Caspian basin, were important in triggering their diversification (Zenkevitch, 1963; Dumont, 1998b).

image

Figure 1. Collection sites for predatory cladocerans examined in this study and the natural distribution (black areas) of the endemic Ponto-Caspian onychopods.

Download figure to PowerPoint

Hypotheses of onychopod phylogeny

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Most past efforts to understand the phylogenetic relationships of cladocerans have focused on higher taxonomic levels (Fryer, 1987), examining the monophyly of the order and its component suborders. Martin & Cash-Clark (1995) proposed the first morphologically based phylogeny for the onychopod families. They suggested that the family Polyphemidae was basal, with the Cercopagidae and Podonidae as sister groups derived from it (Fig. 2a). By contrast, Rivier (1998) provided evidence for the monophyly of the Polyphemide and Cercopagidae based on shared structures in their eye, head shield and valve morphology (Fig. 2b). The uncertainty in morphologically based phylogenies for this group derives from the fact that their reconstruction cannot be made on the basis of unique synapomorphies. In fact it seems likely that the assemblage of primitive features and shared derived characters has been altered by convergent adaptations linked to a predatory lifestyle and the occupancy of shared environments. Moreover, most taxa show phenotypic plasticity which affects diagnosable characters and makes taxon delimitation difficult. A possible solution to the uncertain phylogeny of this group lies in the application of molecular or combined molecular–morphological approaches.

image

Figure 2. Morphologically based hypotheses of onychopod phylogeny. (a) Onychopod families ( Martin & Cash-Clark, 1995 ). (b) Cladogram inferred from the evolutionary tree proposed by Rivier (1998) .

Download figure to PowerPoint

Ages and rates of diversification in the onychopods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Given the apparent importance of the Ponto-Caspian basin as the site of onychopod evolution, it is worth noting the major geological and climatic events which shaped this basin and influenced its ecology. During the late Miocene, some 8–10 million years (myr) ago, the Paratethys Sea, a remnant of the marine Tethys Sea, evolved into the brackish Sarmatian Lake, and later into the slightly brackish Pontic Lake (Zenkevitch, 1963; Jones & Simmons, 1997). By the end of the Pontic period (5–4 myr ago), uplifts fragmented this lake into the Black and Caspian Seas. Since then, these two basins have largely been separated, barring periods of ephemeral contact. One opportunity for faunal exchange between the basins occurred during the late Pliocene (2–3 myr ago) when the vast Akchagyl Lake had connections with both the Kuyalnits (Black Sea) and the Persian Gulf. During the Pleistocene, climatic and hydrologic changes produced dramatic transformations of the Black Sea basin. The global rise in sea level following each glaciation led to repeated incursions of saline Mediterranean water into the brackish Black Sea. This inflow increased the salinity of the Black Sea, creating a pycnocline in which a surface layer derived from river inflows was separated from the more saline, denser water of Mediterranean origin. This stable stratification of water masses created a transition from oxic to anoxic conditions at the sediment–water interface. Palaeontological studies on mollusks and pelagic microfossils provide evidence that much extinction of the local, brackish water biota occurred about 7000–3000 years ago as a result of the latest flood of saline Mediterranean waters (Wall & Dale, 1974). The survivors of this extinction event are currently confined to low salinity estuaries, lagoons and rivers along the margins of the Black and Azov Seas (Banarescu, 1991). By contrast, the Caspian basin maintained its endemic fauna throughout the Pleistocene. Despite drastic climatic fluctuations and ephemeral periods of contact with the Black Sea (Chepalyga, 1985), the salinity of this basin has remained relatively steady over this interval maintaining its characteristic north – south gradient.

A major challenge in reconstructing the history of life lies in establishing the relative age of lineages lacking fossil records. Despite the absence of palaeontological data for most of the groups which diversified in the Ponto-Caspian, there has been a tendency to link local radiations to Miocene environments. The term ‘Sarmatian relicts’ has been used by many authors when referring to the Ponto-Caspian endemics, suggesting that most of them originated in the Sarmatian Lake some 8–10 myr ago (Motas, 1977). However, other authors suggest that most of the Ponto-Caspian taxa radiated 5–6 myr ago, during the Pontian (Mordukhai-Boltovskoi, 1979; Banarescu, 1991). The clearest evidence for such diversification relates to the species flock of limnocardiids, a molluscan family with an excellent fossil record which radiated during the Pontian (Banarescu, 1991). Interestingly, the Pontian Lake had ecological conditions similar to those of the modern Caspian Sea, whereas the Sarmatian Sea was more saline, resembling conditions in the Black Sea today. Focusing specifically on the onychopods, Rivier (1998) proposed that much of the speciation in this group (including the radiation of the marine podonids) occurred very recently, during the last 10 000–20 000 years of the Pleistocene. Dumont (2000) similarly suggested that the marine podonids radiated within the Ponto-Caspian area and were only released into the world's oceans during the Holocene, when the Black Sea established contact with the Mediterranean. By contrast, based on their analysis of 12S sequence diversity in onychopods, Richter et al. (2001) suggested that marine podonids first radiated in the world's oceans, entering the Ponto-Caspian basin, just 2–3 myr ago, during the Akchagylian transgression. Consequently, these authors place the major speciation events of the Ponto-Caspian podonids during the upper Pliocene and the radiation of the Cercopagidae and Polyphemidae during the Pleistocene.

Our work examines the tempo of morphological and genetic changes in the predatory cladocerans based on a molecular phylogenetic approach in an attempt to reconcile the competing hypotheses regarding their evolution. In contrast to the earlier molecular study on this group, which relied on the analysis of a single mitochondrial gene (Richter et al., 2001), the present results are based on a study of both a nuclear gene (18S) and three mitochondrial genes (COI, 12S, 16S).

Taxon sampling

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Fourteen onychopod species, including representatives from nine of the 10 recognized genera, were examined (Table 1). The habitats sampled included marine waters, brackish estuaries, brackish inland seas, and freshwater ponds and lakes (Fig. 1). Samples were collected using a 100 or 200 μm mesh net and were subsequently sorted and preserved in 90% ethanol. Taxa were identified using Rivier's (1998) and Negrea's (1983) keys. As multiple outgroups are preferable for polarizing characters, we chose two anomopod cladocerans, Daphnia pulex and Bosmina coregoni, as outgroup species in our phylogenetic analyses. Our outgroup choice was based on previous morphological and recent molecular analysis of cladoceran phylogeny (Martin & Cash-Clark, 1995; Negrea et al., 1999) which suggest that the anomopods are the sister group to the onychopods.

Table 1.  Species included in the molecular analyses, their habitat preferences (F – freshwater; M – neritic or oceanic; B – inland brackish water), and GenBank accession numbers.
  GenBank accession no.
Taxon*HabitatCOI12S16S18S
  1. * Classification after Rivier (1998 ). † GenBank accession no. AY075047AY075093 represent taxa sequenced during this project. GenBank accession no. AF117817 ; AF435130 ; AF320013AF320014 ; AY009493AY009499 ; correspond to specimens used in Crease (1999), Therriault et al. (2002) ; Cristescu et al. (2001) ; Richter et al. (2001) , respectively. ‡ Taxa used as outgroup.

Family Polyphemidae
 Polyphemus pediculusFAY075048AY009495AY075066AY075080
Family Cercopagidae
 Cercopagis pengoiF–BAF320014AY009494AY075067AY075081
 Bythotrephes longimanusFAF435130AY009493AY075069AY075082
Family Podonidae
 Cornigerius maeoticusF–BAY075047AY075058AY075068AY075083
 Evadne anonyxB AY009499  
 E. nordmanniMAY075049AY009498AY075070AY075084
 E. spiniferaM AY075059AY075071AY075085
 Pleopis polyphemoidesMAY075050AY075060AY075072AY075086
 Podon leuckartiMAY075051AY009496AY075073AY075087
 P. intermediusMAY075052AY009497AY075074AY075088
 Pseudevadne tergestinaMAY075053AY075061AY075075AY075089
 Podonevadne trigonaF–BAY075054AY075062AY075076AY075090
 P. angustaBAY075055AY075063AY075077AY075091
 P. camptonyxBAY075056AY075064AY075078AY075092
Family Daphniidae
 Daphnia pulexFAF117817AF117817AF117817AF117817
Family Bosminidae
 Bosmina coregoniFAY075057AY075065AY075079AY075093

DNA extraction, amplification and sequencing

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Phylogenies were constructed using three mitochondrial genes – cytochrome c oxidase subunit I (COI), 12S, 16S ribosomal DNA (rDNA), and one nuclear gene – 18S rDNA. The primer pairs LCOI490 and HCO2198 (Folmer et al., 1994), 12CR3 and 12s-5c (Richter et al., 2001), 16Sar and 16Sbr (Palumbi, 1996), were used to amplify a 658 base pair (bp) fragment of the COI gene, a 565-bp fragment of the 12S gene, and a 570-bp fragment of the 16S gene, respectively. For the nuclear 18S gene fragment, the primers 9F and 2004R (Crease & Colbourne, 1998) were used to amplify a 2000-bp fragment which was subsequently partially sequenced using the amplification primer 2004R. The sequenced 18S fragment was approximately 800 bp in length. Total DNA was extracted from three to five individuals from each habitat for 13 of the onychopod species using proteinase K methods. Amplification and sequencing were performed as described in Cristescu et al. (2001), with the only difference being a change in the temperature profiles for the 16S and 18S polymerase chain reaction (PCR) amplification. The PCR for 16S consisted of two cycles of 94 °C (30 s), 60 °C (45 s), 72 °C (45 s); five cycles of 93 °C (30 s), 55 °C (45 s), 72 °C (45 s); followed by 29 cycles of 93° C (30 s) 50° C (1 min) and 72° C (1 min). The temperature profiles for 18S consisted of 35 cycles of 93 °C (30 s), 50 °C (30 s) and 72 °C (3 min). We performed sequencing in both directions only when ambiguous sites were encountered. Two onychopod taxa were incompletely represented in the four partitions. For Evadne spinifera we obtained an aberrant COI sequence (possibly a pseudogene), whereas for E. anonyx we only used the 12S sequence from GenBank (Richter et al., 2001). All sequences obtained during this study have been submitted to GenBank under accession numbers: AY075047AY075093 (Table 1).

Phylogenetic reconstruction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Sequences for the four genes were initially aligned using Sequencher (Gene Codes Corporation, Ann Arbor, MI, USA). Secondary sequence structure models for 12S, 16S (Taylor et al., 1998) and 18S (Crease & Colbourne, 1998) were used as a guide to align hypervariable regions of the ribosomal genes. Sites that contained gaps and the ambiguous sections of the alignment (almost 430 sites from the ribosomal genes) were excluded from subsequent analyses. Molecular phylogenies were inferred using three analytical approaches. Maximum-parsimony (MP), maximum-likelihood (ML) and neighbour-joining (NJ) (Saitou & Nei, 1987) methods were employed using paup* 4.0b.8 (Swofford, 2001). MP trees were estimated using the branch-and-bound search with all characters having equal weight and gaps treated as ‘missing’, whereas ML trees were constructed using the HKY85 model and the heuristic search option with sequences added at random and tree bisection-reconnection branch swapping. All estimates of sequence divergence were corrected using Tamura and Nei's method (Tamura & Nei, 1993). We assessed the stability of phylogenetic hypotheses with both bootstrap analyses (Felsenstein, 1985) (1000 replicates for both NJ and MP and 100 for ML) and decay indices (Bremer, 1994) calculated using the AutoDecay program (version 4.0.2; Eriksson, 1999, Bergius foundation, Stockholm). The homogeneity of base composition across taxa was assessed using the goodness-of-fit (χ2) test implemented in paup* (Swofford, 2001). The appropriateness of performing a total evidence analysis, a conditional data combination or separate analyses of all data partitions was assessed using the incongruence length difference (ILD) test (Farris et al., 1995) performed with the partition homogeneity test in paup* with 1000 random bipartitions analysed by TBR branch swapping on 10 random sequence-addition replicates. To test for constancy of rates of molecular evolution, we used a log-likelihood ratio test, performed in paup* to statistically estimate the validity of the molecular clock hypothesis (Huelsenbeck & Crandall, 1997). The linearized NJ tree was employed to examine the relative timing of the onychopod evolution under the assumption of a molecular clock (Takezaki et al., 1995). As two taxa, E. spinifera and E. anonyx, were coded as missing in several data partitions, we performed the ILD measures and the likelihood ratio tests after removing these taxa from the data set.

Molecular clock, ages and rates of diversification in Onychopoda

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

The snapping shrimp mitochondrial COI clock (Knowlton & Weigt, 1998) of 1.4% sequence divergence per million years , the porcelain crab mitochondrial 16S clock (Stillman & Reeb 2001) of 0.53% sequence divergence per million years and the arthropod mitochondrial clock (Brower, 1994) of about 2.3% sequence divergence were used for dating major evolutionary events within the onychopod group.

Sequence diversity

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

The magnitude of interspecific sequence divergence as well as that between the four genes was highly variable. A maximum pairwise nucleotide divergence of 16% for all genes combined was observed between the onychopod families. Within the Cercopagidae and the Podonidae, the mean level of genetic divergence among species was 14 and 8%, respectively. The least genetically diverged group was the Ponto-Caspian podonids which showed less than 1% sequence divergence. As expected, less divergence was found for the rDNA genes than for the faster evolving protein coding gene COI. Consequently, the number of parsimony informative sites varied among the genes from 46 informative sites in the 18S data set to 230 informative sites in the COI data set (Table 2). A total of 2162 characters, including 498 cladistically informative characters, were available after combining the data sets for 12S, 16S, 18S rDNA and COI. Each of the four genes displayed unequal base frequencies with 18S, the slowest evolving of the four genes, displaying the least bias. All mitochondrial genes (COI, 12S and 16S) had ahigher A-T content (59.1, 62.5, 60.5%) than thenuclear 18S rDNA gene (48.7%). The estimated Transition/Transversion (Ti/Tv) ratio among all taxa for the complete data set was 1.95. Among the individual data partitions, the COI data set had the highest Ti/Tv (5.42). Base composition for each gene was homogenous across the 14 taxa (0.99 < P < 1.00) (Table 2). Partition homogeneity tests indicated the presence of conflicting phylogenetic signal when data for all four genes were included in the comparison (P = 0.04). Although there is not a generally accepted P-value for a significant result, most authors argue for combining data when P-values are greater than 0.05. No significant heterogeneity among genes remained after the removal of COI from the data set (P=0.41). This result suggests that only the COI gene provides a different phylogenetic signal, probably because of its saturation at the third codon position. There is no generally accepted protocol to handle data showing incongruence (Huelsenbeck et al., 1996). In the present study we employed two common strategies: a total evidence approach and a conditional combination of all ribosomal genes (12S, 16S and 18S) vs. the protein coding gene, COI, followed by an independent analysis of each gene. Likelihood ratio tests for the molecular clock hypothesis suggested that the COI and 16S data partitions were consistent with a constant rate of evolution among the taxa included in the study. A molecular clock hypothesis was, however, rejected for the 12S and 18S data partitions (Table 3).

Table 2.  The total number of sites available (TS), the number of variable sites (VS), the number of cladistically informative sites (IS), base frequencies, transition/transversion ratios (Ti/Tv) and χ 2 test of homogeneity of base frequencies across ingroup taxa for each data partition.
    Base composition (%)
Data setTSVSISACGTTi/Tvχ2
CO163027123025.6520.2820.5933.485.4216.93; P=0.99
12S39018412133.1820.0817.4629.271.75 9.32; P=0.99
16S44614610129.9916.2123.2530.551.80 4.75; P=1.00
18S696116 4624.6724.1627.1324.041.29 1.45; P=1.00
12S, 16S, 18S153244626826.0421.3023.7228.931.49 9.87; P=0.99
Total evidence216271749827.6420.6122.6329.131.9510.48; P=0.99
Table 3.  Likelihood ratio test for the molecular clock hypothesis 2Δ= log  Lnoclock − log  Lclock ( χ(14,0.05)2 =23.68; χ(13,0.05)2 =22.36; χ(12,0.05)2 = 21.03).
   Likelihood ratio test
Data sets–log Lno clock–log Lclockd.f. = n−2Null hypothesis
  1. Maximum-likelihood trees were reconstructed using the HKY85 model.

CO1 4112.8 4121.818.012Not rejected
12S 2208.2 2230.444.514Rejected
16S 1982.8 1987.1 8.613Not rejected
18S 1878.4 1892.628.413Rejected
Total evidence10618.210633.530.612Rejected

Molecular phylogenetic analyses

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Maximum-parsimony analysis of total evidence using the branch-and-bound algorithm found a single most parsimonious tree of 1748 steps long (CI=0.53, RI=0.53) (Fig. 3a). This tree had a similar topology with the strict consensus of the two shortest MP trees (tree length 863, CI=0.7, RI=0.28) obtained for the conditional data combination of the three rDNA genes (Fig. 3b). Maximum-likelihood analysis of the total evidence, and the rDNA gene data produced trees with similar topologies to both the MP trees and NJ trees (Figs 4 and 5). The most likely parsimony trees of the total evidence (Fig. 4a) and the conditional combination (Fig. 4b) had log-likelihood scores of −10618.3 and −6259.2, respectively. All analyses of the total evidence and conditional combination unambiguously supported the monophyly of the taxa belonging to the family Cercopagidae and to the family Podonidae. Polyphemus pediculus occurred as a monophyletic entity.

image

Figure 3. (a) The most parsimonious tree obtained using the maximum parsimony (MP) criterion of the total evidence (COI, 12S, 16S and 18S)and (b) Strict consensus of two shortest trees obtained using the same (MP) criterion on sequence divergence from the rDNA genes (12S, 16S and 18S). The numbers above the branches indicate bootstrap support greater than 50% (1000 replications). The numbers below the branches show the decay indices. MP trees were estimated using the branch-and-bound search with all characters having equal weight and gaps treated as ‘missing’.

Download figure to PowerPoint

image

Figure 4. Maximum-likelihood (ML) trees for (a) the total evidence (COI, 12S, 16S and 18S) and (b) the conditional data combination of rDNAgenes (12S, 16S and 18S). The numbers above the branches indicate bootstrap support greater than 50% (100 replications). ML treeswere constructed using the HKY85 model and the heuristic search option (with sequences added at random and tree bisection-reconnection branch swapping).

Download figure to PowerPoint

image

Figure 5. Neighbour-Joining (NJ) trees (a) of the total evidence (COI, 12S, 16S and 18S genes) (b) of the rDNA genes (conditional data combination of 12S, 16S and 18S). The trees are rooted with the anomopod genera Daphnia and Bosmina . The number above the branches indicates the bootstrap support after 1000 replications, whereas the number below each branch shows the corrected distances based on Tamura-Nei corrected distance matrix.

Download figure to PowerPoint

Ages of Onychopods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Application of calibrations for the mitochondrial COI and 16S genes (Knowlton & Weigt, 1998; Stillman & Reeb, 2001), and the arthropod mitochondrial clock (Brower, 1994) to the linearized NJ trees of the corresponding gene fragments suggested that diversification of the onychopods has occurred over the past 10–20 myr. The diversification of the cercopagid genera and of the marine podonids occurred during Middle or Late Miocene. However, the radiation of the Ponto-Caspian podonids took place much more recently, during the Late Pleistocene (Fig. 6).

image

Figure 6. Phylogenetic relationships among 13 onychopod species based on a NJ linearized tree of the mtDNA genes (COI, 12S, 16S). The numbers above the branches indicate the bootstrap support (1000 replicates). Genetic distances were corrected using the Tamura-Nei method. The arthropod mitochondrial clock ( Brower, 1994 ) of 2.3% sequence divergence per million years was used to estimate divergence times. The grey scale represents the salinity gradient from 35‰ (black) to 2–4‰ (light grey) of the corresponding geological basins (after Jones & Simmons, 1997 ).

Download figure to PowerPoint

Phylogenetic implications

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

The present study has confirmed the monophyly of the three onychopod families (Polyphemidae, Cercopagidae and Podonidae) by all tree reconstruction methods and all data sets except COI (which did not support the monophyly of the Cercopagidae). However, the homoplasious phylogenetic signals of the COI gene were not apparent in the total evidence trees. In fact, the support for the monophyly of the cercopagids in the total evidence trees was robust (100% bootstrap support for all algorithms). The results further indicate that the families Cercopagidae and Podonidae are monophyletic, forming a sister group to the Polyphemidae. This result supports Martin & Cash-Clark's (1995) phylogeny based on morphology and Richter et al.'s (2001) molecular hypothesis based on the ribosomal 12S gene. This relationship among the families was recovered by ML and MP approaches in both the total evidence data set and the conditional data combination. NJ produced the same result when the total evidence was employed, but failed to recover it when only the rDNA genes were analysed. Within the podonids, we recovered two well supported clusters: one group includes the two marine genera Podon and Pleopis, whereas the other included the Ponto-Caspian podonids together with the marine genera Evadne and Pseudevadne. Within the latter, heterogeneous group, we found strong support (100% bootstrap andnine Bremer support) for the monophyly of thePonto-Caspian podonids belonging to the genera Podonevadne and Cornigerius. In fact, the data reveal a remarkably close evolutionary relationship among the members of this flock, but their marked divergence from the genera Evadne and Pseudevadne. It was also apparent that the EvadnePseudevadne group is paraphyletic incorporating two Ponto-Caspian species flocks: the brackish – freshwater PodonevadneCornigerius group and the more salt tolerant Caspian Evadne group which wasrepresented in our phylogeny by a single species, E. anonyx. Although, the polyphyly of the Ponto-Caspian podonids was evident, the relationship between the two flocks with the marine Evadne was not clear. ML and NJanalyses suggest an affinity between Pseudevadne tergestina and Podonevadne – Cornigerius group, but the support for this group was weak. Although all methods ofphylogenetic reconstruction employed suggested that E. normanni is the sister taxa of the Caspian E. anonyx, this conclusion was based only on the 12S gene and its support was weak.

We regard the total evidence hypotheses as the most robust as evidenced by the highest support for the major clades. The evidence from the conditional combination of the rDNA genes, which had a congruent pattern of phylogenetic affinity with the total evidence analysis, high bootstrap and Bremer support provides lower phylogenetic resolution within the shallow Ponto-Caspian groups. By contrast, analysis of the individual gene partitions yielded phylogenies which were inconsistent with the monophyletic groups and with the topology supported by the total evidence trees. Moreover, these conflicting taxonomic assemblages were only weakly supported. Our results strengthen arguments for the value of a total evidence approach (Hillis, 1987; Kluge, 1989; Remsen & DeSalle, 1998). It is apparent that combining data sets is a good option for intensifying phylogenetic signal, even when there is evidence of conflicts in the nature of this information.

Ages and rates of diversification in Onychopoda

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

Based upon the linearized NJ phylogeny for three mitochondrial genes and the arthropod mitochondrial clock (Fig. 6), we identified two relatively brief periods of onychopod diversification. The initial radiation of the marine podonids probably occurred during the late Miocene or early Pliocene. However, if the calibrations of Knowlton & Weigt (1998) and Stillman & Reeb (2001) are applied to the corrected nucleotide divergences of the COI and 16S fragments, respectively, this radiation is placed earlier in time, during the early and middle Miocene. Regardless of errors introduced by the calibration used or by the assumption of a constant rate of evolution among crustaceans, it is likely that the Paratethys and Sarmatian basins fostered this first diversification of the podonids. The two cercopagid genera, Cercopagis and Bythotrephes, last shared a common ancestor sometimes during upper or middle Miocene. The marine cladocerans most probably gained access to the oceans via the Mediterranean long before the Holocene, possibly during the late Miocene or the Pliocene when the Ponto-Caspian basin was connected with the world's oceans. This hypothesis is supported by the fact that intraspecific levels of genetic divergence inthe COI gene of marine cladocerans are 100–250 timeshigher than the level expected if these taxa first colonized the oceans in the Holocene (M.E.A. Cristescu and P.D.N. Hebert unpublished work). Simultaneous with the diversification of the Evadne and Pseudevadne lineages, the lineage leading to the endemic Ponto-Caspian genera Cornigerius and Podonevadne diverged. This ancestral lineage survived the less saline periods (Pontic or Balakhan basins of the late Miocene, early Pliocene, respectively) and made a gradual transition to environments with lower salinity. The second major radiation, the diversification of the Ponto-Caspian endemic species belonging to the genera Cornigerius and Podonevadne, took place much more recently, during the Pleistocene. Members of this group are genetically almost indistinguishable. Only the genera Cornigerius and Podonevadne have a long enough history of isolation (1–1.5 myr) to enable their reliable discrimination. It appears that the radiation of the Podonevadne flock: P.trigona, P. angusta and P. camptonyx occurred very recently, possibly during the Holocene. Morover, analyses of COI diversity within these species revealed shared haplotypes between sympatric sister species suggesting that introgression occurs between these closely related taxa (M.E.A. Cristescu et al., unpublished work). We do not have molecular evidence regarding relatedness within the other endemic flock, the Evadne group (E. anonyx, E. angusta and Caspievadne maximovitschi), which is confined to the middle and south Caspian Sea. Despite the morphological peculiarity of the genus Caspievadne, its similar limb structure to Caspian Evadne suggests their close evolutionary relationship (Fig. 7). Future genetic work is necessary to further explore relationships between these taxa with a view towards clarifying the factors involved in radiation. Nevertheless, the convergent evolution of body shapes in the two Caspian flocks (e.g. C. maximovitschi and C. maeoticus; E.anonyx and P. camptonyx; E.prolongata and P. angusta) suggest that these two groups evolved almost simultaneously, shaped by the same intralacustrine, evolutionary forces (Fig. 7). As the radiation in each group did not involve significant habitat or food specialization, we propose that disruptive selection by predators was the driving force in shaping the body plane diversity of onychopods.

image

Figure 7. Convergent evolution of body shape in two Ponto-Caspian podonid groups. The monophyly of the genera Cornigerius and Pseudevadne issupported by both limb morphology and genetic data, the monophyly of the genera Caspievadne and the Caspian Evadne is suggested only bytheir similar limb structure. Drawings after Rivier (1998) .

Download figure to PowerPoint

In summary, our phylogenetic data reveal an onychopod radiation extending from the Miocene to the late Pleistocene in a chain of events intimately linked with the geological history of the Ponto-Caspian basin. This example resembles previously studied vertebrate radiations in which the Quaternary events were important in extending a radiation inaugurated earlier (Klicka & Zink, 1997; Avise, 2000). In this context, the identification of the evolutionary factors which directed the radiation in Ponto-Caspian lineages, and the reasons for these pulses of diversification appear more important than assigning a precise age to individual lineages.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References

We thank E. Remigio, J.D.S. Witt, V. Sacherova and T.J. Crease for helpful suggestions during the project. We are grateful to all colleagues who provided us with samples: M.G. Mazzocchi, I. Siokou, M. Pfrender, J.K. Colbourne and M. Tedergren. We thank the research team of the Caspian Institute in Astrakhan, especially A. Sokolsky, A.S. Mikouiza, V. Ivanov, V.I. Beliaeva and E. Tikonova for their support. We are also grateful to H. MacIsaac, I. Grigorovich, L. Galatchi, and V.R. Alekseev for aiding in collections. A. Holliss assisted us with DNA sequencing. Two anonymous referees provided valuable comments on a previous version of this manuscript. This work was supported by grants from the Natural Sciences and Engineering Research Council of Canada and the Canada Research Chairs Program to P.D.N.H.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Hypotheses of onychopod phylogeny
  5. Ages and rates of diversification in the onychopods
  6. Materials and methods
  7. Taxon sampling
  8. DNA extraction, amplification and sequencing
  9. Phylogenetic reconstruction
  10. Molecular clock, ages and rates of diversification in Onychopoda
  11. Results
  12. Sequence diversity
  13. Molecular phylogenetic analyses
  14. Ages of Onychopods
  15. Discussion
  16. Phylogenetic implications
  17. Ages and rates of diversification in Onychopoda
  18. Acknowledgments
  19. References
  • Aladin, N.V. & Potts, W.T.W. 1995. Osmoregulatory capacity of the Cladocera. J. Comp. Physiol. B. 164: 671683.
  • Avise, J.C. 2000. Phylogeography: The History and Formation of Species. Harvard University Press, London.
  • Banarescu, P. 1991. Zoogeography of Fresh Waters. Distribution and Dispersal of Freshwater Animals in North America and Eurasia, Vol. 2. AULA-Verlag, Weisbaden.
  • Bremer, K. 1994. Branch support and tree stability. Cladistics 10: 295304.
  • Brower, A.V.Z. 1994. Rapid morphological radiation and convergence among races of the butterfly Heliconius erato inferred from patterns of mitochondrial DNA evolution. Proc. Natl. Acad. Sci. USA 91: 64916495.
  • Chepalyga, L.A. 1985. Inland sea basins. In: Late Quaternary Environments of the Soviet Union. (A. A.Velichko, ed.), pp. 229247. University of Minnesota Press, Minneapolis.
  • Crease, T.J. 1999. The complete sequence of the mitochondrial genome of Daphnia pulex (Cladocera: Crustacea). Gene 233: 8999.
  • Crease, T.J. & Colbourne, J.K. 1998. The unusually long small-subunit ribosomal RNA of the crustacean Daphnia pulex: Sequence and predicted secondary structure. J. Mol. Evol. 46: 307313.
  • Cristescu, M.E.A., Hebert, P.D.N., Witt, J.D.S., MacIsaac, H.J. & Grigorovich, I.A. 2001. An invasion history for Cercopagis pengoi based on mitochondrial gene sequences. Limnol. Oceanogr. 46: 224229.
  • Dumont, H.J. 1998a. Caspian Lake ecosystem. Limnol. Oceanogr. 43: 4452.
  • Dumont, H.J. 1998b. The Caspian cradle. In: The Predatory Cladocera (Onychopoda: Podonidae, Polyphemidae, Cercopagidae) and Leptodoridae of the World. Guides to the Identification of the Microinvertebrates of the Continental Waters of the World (H. J. F.Dumont, ed.), pp. 915. Backhuys, Leiden.
  • Dumont, H.J. 2000. Endemism in the Ponto-Caspian fauna, with special emphasis on the Onychopoda (Crustacea). In: Advances in Ecological Research. Ancient Lakes: Biodiversity, Ecology and Evolution (A.Rossiter & H. Kawanabe, eds), pp. 181196. Academic Press, London.
  • Egloff, D.A., Fofonoff, P.W. & Onbe, T. 1997. Reproductive biology of marine cladocerans. In: Advances in Marine Biology (J. H. S.Blaxter & A. J. Southward, eds), pp. 79167. Academic Press, London.
  • Farris, J.S., Kallersjo, M., Kluge, A.G. & Bult, C. 1995. Testing significance of incongruence. Cladistics 10: 315319.
  • Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 785791.
  • Folmer, O., Black, M., Hoeh, W., Lutz, R. & Vrijenhoek, R. 1994. DNA primers for amplification of mitochondrial cytochrome coxidase subunit I from diverse metazoan invertebrates. Mol. Mar. Biol. Biotechnol. 3: 294299.
  • Fryer, G. 1987. Morphology and the classification of the so-called Cladocera. Hydrobiol. 145: 517.
  • Hillis, D.M. 1987. Molecular versus morphological approaches to systematics. Annu. Rev. Ecol. Syst. 18: 2342.
  • Huelsenbeck, J.P., Bull., J.J. & Cunningham, C.W. 1996. Combining data in phylogenetic analysis. Trends Ecol. Evol. 11: 152158.
  • Huelsenbeck, J.P. & Crandall, K.A. 1997. Phylogeny estimation and hypothesis testing using maximum likelihood. Annu. Rev. Ecol. Syst. 28: 437466.
  • Jones, R.W. & Simmons, M.D. 1997. A review of the stratigraphy of eastern Paratethys (Oligocene – Holocene), with particular emphasis on the Black Sea. In: Regional and Petroleum Geology of the Black Sea and Surrounding Region (A. G.Robinson, ed.), pp. 3952. AAPG Memoir 68, Tulsa, OK, USA.
  • Klicka, J. & Zink, R.M. 1997. The importance of recent Ice Ages in speciation: a failed paradigm. Science 277: 16661669.
  • Kluge, A.G. 1989. A concern for evidence and a phylogenetic hypothesis of relationships among Epicrates (Boidae, Serpentes). Syst. Zool. 38: 725.
  • Knowlton, N. & Weigt, L.A. 1998. New dates and new rates for divergence across the Isthmus of Panama. Proc. R. Soc. Lond., Series B: Biol. Sci. 265: 22572263.
  • Martin, J.W. & Cash-Clark, C.E. 1995. The external morphology of the onychopod ‘cladoceran’ genus Bythotrephes (Crustacea, Branchiopoda, Onychopoda, Cercopagidae), with notes on the morphology and phylogeny of the order Onychopoda. Zool. Scripta 24: 6190.
  • Monchenko, V.J. 1998. The Ponto-Caspian zoogeographic complex of Cyclopida in the Caspian, Azov and Black Sea. J. Mar. Syst. 15: 421424.
  • Mordukhai-Boltovskoi 1965. Polyphemidae of the Ponto-Caspian Basin. Hydrobiol. 25: 212220.
  • Mordukhai-Boltovskoi 1968. On the taxonomy of the Polyphemidae. Crustaceana 14: 197209.
  • Mordukhai-Boltovskoi 1979. Contribution and distribution of Caspian fauna in the light of modern data. Int. Revue ges. Hydrobiol. 64: 138.
  • Motas C. 1977. L'origine de la fauna actuelle de la mer Noire. In: Biologie Des Eaux Saumatres de la Mer Noire (E. A.Pora & M.C.Bãcescu, eds), pp. 5659. Institut Roumain de Recherches Marines, Constanta.
  • Negrea, S. 1983. Crustacea. Cladocera. In: Fauna Republicii Socialiste Romania, Vol. 4 (N.Botnariuc, P.Bãnãrescu, R.Codreanu & M. A. Ionescu, eds), pp. 395. Editura RSR, Bucharest (in Romanian).
  • Negrea, S., Botnariuc, N. & Dumont, H.J. 1999. Phylogeny, evolution and classification of the Branchiopoda (Crustacea). Hydrobiol. 412: 191212.
  • Onbe, T. 1999. Ctenopoda and Onychopoda (=Cladocera). In: South Atlantic Zooplankton. (D.Boltovskoy, ed.), pp. 797813. Backhuys, Leiden, The Netherlands.
  • Palumbi, S. 1996. Nucleic acids II: The polymerase chain reaction. In: Molecular Systematics (D.Hillis, B.Mable & C. Moritz, eds), pp. 205247. Sinauer, Sunderland, MA.
  • Remsen, J. & DeSalle, R. 1998. Character congruence of multiple data partitions and the origin of the Hawaiian Drosophilidae. Mol. Phylog. Evol. 9: 225235.DOI: 10.1006/mpev.1997.0484
  • Richter, S., Braband, A., Aladin, N. & Scholtz, G. 2001. The phylogenetic relationships of ‘predatory water-fleas’ (Cladocera: Onychopoda, Haplopoda) inferred from 12S rDNA. Mol. Phylog. Evol. 19: 105113.DOI: 10.1006/mpev.2000.0901
  • Rivier, I.K. 1998. The Predatory Cladocera (Onychopoda: Podonidae, Polyphemidae, Cercopagidae) and Leptodoridae of the world. In: Guides to the Identification of the Microinvertebrates of the Continental Waters of the World, Vol. 13 (H. J. F.Dumont, ed.), pp. 213. Backhuys, Leiden.
  • Saitou, N. & Nei, M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406425.
  • Stillman, J.H. & Reeb, C.A. 2001. Molecular phylogeny of eastern Pacific porcelain crabs, genera Petrolisthes and Pachycheles, based on the mtDNA 16S rDNA sequence: phylogeographic and systematic implications. Mol. Phylog. Evol. 19: 236245.DOI: 10.1006/mpev.2001.0924
  • Swofford, D.L. 2001. Paup: Phylogenetic Analysis Using Parsimony, Version 4.Ob8. Sinauer Associates, Sunderland, MA.
  • Takezaki, N., Rzhetsky, A. & Nei, M. 1995. Phylogenetic test of the molecular clock and linearized trees. Mol. Biol. Evol. 12: 823833.
  • Tamura, K. & Nei, M. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Mol. Biol. Evol. 10: 512526.
  • Taylor, D.J., Finston, T.L. & Hebert, P.D.N. 1998. Biogeography of a widespread freshwater crustacean: pseudocongruence and cryptic endemism in the North American Daphnia laevis complex. Evolution 52: 16481670.
  • Therriault, T.W., Grigorovich, I.A., Cristescu, M.E., Ketelaars, H.A.M., Viljanen, M., Heath, D.D. & MacIsaac, H.J. 2002. Taxonomic resolution of the genus Bythotrephes Leydig using molecular markers and re-evaluation of its global distribution, with notes on factors affecting dispersal, establishment and abundance. Diversity Distribution (in press).
  • Väinölä, R. 1995. Origin and recent endemic divergence of a Caspian Mysis species flock with affinities to the ‘glacial relict’ crustaceans in boreal lakes. Evolution 49: 12151223.
  • Wall, D. & Dale, B. 1974. Dinoflagellates in late Quaternary deep-water sediments of Black Sea. In: The Black Sea – Geology, Chemistry, and Biology (E. T.Degens & D. A. Ross, eds), pp. 364380. AAPG Tulsa, Oklahoma.
  • Zenkevitch, L. 1963. Biology of the Seas of the USSR. George Allen & Unwin Ltd, London.