Phylogeny and Biogeography of Spinicaudata (Crustacea: Branchiopoda)

© 2020 Academia Sinica, Taiwan Open Access Special Issue: Fossil and Modern Clam Shrimp (Branchiopoda: Spinicaudata, Laevicaudata) Phylogeny and Biogeography of Spinicaudata (Crustacea: Branchiopoda) Martin Schwentner1,2,*, Nicolas Rabet3, Stefan Richter4, Gonzalo Giribet5, Sameer Padhye6, JeanFrançois Cart7, Céline Bonillo3, and D. Christopher Rogers8 1Center of Natural History, Universität Hamburg, Hamburg, Germany. *Correspondence: E-mail: [email protected] (Schwentner) 2Naturhistorisches Museum, Vienna, Austria 3Sorbonne Université, Muséum national d’Histoire naturelle, Biologie des organismes et écosystèmes aquatiques (BOREA), CNRS, IRD, Université de Caen Basse-Normandie, CP26 75231, 43 rue Cuvier Paris Cedex 05, France. E-mail: [email protected] (Rabet), [email protected] (Bonillo) 4Allgemeine und Spezielle Zoologie, Universität Rostock, Rostock, Germany. E-mail: [email protected] (Richter) 5Museum of Comparative Zoology, Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, Massachusetts, USA. E-mail: [email protected] (Giribet) 6Systematics, Ecology & Conservation Lab, Zoo Outreach Organization, Coimbatore, Tamil Nadu, India. E-mail: sameer[email protected] (Padhye) 715 Avenue du Général de Gaulle, 10400 Nogent-sur-Seine, France. Email: [email protected] (Cart) 8Kansas Biological Survey, and The Biodiversity Institute, The University of Kansas, Higuchi Hall, 2101 Constant Avenue, Lawrence, KS 66047-3759, USA. E-mail: [email protected] (Rogers) Received 30 October 2019 / Accepted 9 March 2020 / Published 5 August 2020 Special issue (articles 32-46) communicated by Thomas A. Hegna and D. Christopher Rogers Spinicaudata (spiny clam shrimp) is a taxon of Branchiopoda occurring since the Devonian and today it occurs nearly globally in temporary water bodies. We present the most species-rich phylogenetic analyses of this taxon based on four molecular loci: COI, 16S rRNA, EF1α and 28S rRNA. Our results support previous findings that Cyzicidae sensu lato is paraphyletic. To render Cyzicidae monophyletic we establish a fourth extant spinicaudatan family to accommodate Eocyzicus. Within Cyzicidae, none of the genera Cyzicus, Caenestheria or Caenestheriella are monophyletic, and the morphological characters used to define these genera (condyle length and rostrum shape) are not associated with well-delimited clades within Cyzicidae. There is insufficient resolution to elucidate the relationships within Leptestheriidae. However, there is sufficient evidence to show that the leptestheriid genera Eoleptestheria and Leptestheria are non-monophyletic, and there is no support for the genus Leptestheriella. Molecular clock analyses suggest that the wide geographic distribution of many spinicaudatan taxa across multiple continents is largely based on vicariance associated with the break-up of Pangea and Gondwana. Trans-oceanic dispersal has occurred in some taxa (e.g., Eulimnadia and within Leptestheriidae) but has been relatively rare. Our results highlight the need to revise the taxonomy of Cyzicidae and Leptestheriidae and provide evidence that the global spinicaudatan diversity may be underestimated due to the presence of numerous cryptic species. We establish Eocyzicidae fam. nov. to accommodate the genus Eocyzicus. Consequently, Cyzicidae comprises only two genera – Cyzicus and Ozestheria. Ozestheria occurs also in Africa and Asia and Ozestheria pilosa new comb. is assigned to this genus. Key words: Clam shrimp, Caenestheria, Caenestheriella, Eocyzicidae, Gondwana, Vicariance. Citation: Schwentner M, Rabet N, Richter S, Giribet G, Padhye S, Cart J, Bonillo C, Rogers DC. 2020. Phylogeny and biogeography of Spinicaudata (Crustacea: Branchiopoda). Zool Stud 59:44. doi:10.6620/ZS.2020.59-44. BACKGROUND Spinicaudata (spiny clam shrimp) has the most confused taxonomy of any branchiopod group due to the tremendous morphological intraspecific variability, which often overlaps with what has been considered interspecific variation, a high number of hermaphroditic lineages, and poor and inadequate descriptions and type Zoological Studies 59:44 (2020) doi:10.6620/ZS.2020.59-44 1 © 2020 Academia Sinica, Taiwan material. Daday generated the first monographs for the group and he described and defined most of the primary extant genera and families we recognize today (Daday 1913a b 1914 1915 1923 1925 1926): Limnadiidae Burmeister, 1843, Cyzicidae Stebbing, 1810 (= Caenestheriidae Daday, 1913), and Leptestheriidae Stebbing, 1902. Daday’s (1925 1926) concept of Limnadiidae was based primarily upon the form of the head, containing all genera lacking fornices (Fig. 1). This included three genera: Limnadia Brongniart, 1820, Eulimnadia Packard, 1874, and Limnadopsis Spencer and Hall, 1862. In Cyzicidae, Daday (1914) included all genera with a fornix but lacking the “lamina epipoditalis”, a triangular lobe on the epipodite of the limbs. In his Cyzicidae concept, Daday recognised Caenestheria Daday, 1913, Eocyzicus Daday, 1913, Caenestheriella Daday, 1913, and Cyzicus Audouin, 1837. In Leptestheriidae, Daday (1923) placed all genera bearing the “lamina epipoditalis” and a fornix, which encompassed the genera Eoleptestheria Daday, 1913, Leptestheria Sars, 1898, and Leptestheriella Daday, 1913. Fig. 1. Figure 1. Representative Spinicaudata and their typical head shapes. A) Ozestheria altus Shu et al., 2015, male head left lateral view; B) Cyzicus californicus (Packard, 1874), female head left lateral view; C) Ozestheria pilosa (Rogers et al., 2013), male head left lateral view; D) Leptestheria kunmingensis Shu et al., 2015, male head left lateral view; E) L. kunmingensis Shu, et al., 2015, female head left lateral view; F) Eocyzicus taiwanensis Rogers et al., 2017, male head left lateral view; G) Ozestheria sp. “Mongolia”, DCR collection 729, male head left lateral view; H) Ozestheria sp. “Mongolia”, DCR collection 729, male limb I endopod distal portion, right lateral view; I) Metalimnadia sp. DCR collection 853, male head left lateral view; J) Eoleptestheria cf. ticinensis from Australia, male head left lateral view. Designations: f = fornix; on = occipital notch; oc = occipital condyle; rs = rostral spine; r = rostrum. page 2 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Daday’s (1913a b 1914 1915 1923 1925 1926) monographs were criticized by many workers (e.g., Uéno 1927; Barnard 1929; Brehm 1933; Gauthier 1933; Linder 1945; Botnariuc 1945 1947; Margalef 1953; Straškraba 1965a 1965b 1966), and many authors (Vecchi 1922; Gauthier 1933; Linder 1945; Botnariuc 1945 1947; Straškraba 1965a 1965b 1966; Wiltshire 1973; Marinček and Petrov 1985; Rogers et al. 2012 2017) demonstrated that certain characters used to define spinicaudatan genera and species were age or food quality dependent. The genera Caenestheria and Caenestheriella, created by Daday (1913a), have been particularly contentious. Caenestheria and Caenestheriella were separated by Daday (1914) from Eocyzicus and Cyzicus, respectively, by the shape of the rostrum. In the former genera, the rostrum was subacute or triangular in both sexes, in the latter two genera the male rostrum was broadly spatulate. Furthermore, Cyzicus and Caenestheriella have a long, elongated occipital condyle with a narrow occipital notch, whereas Eocyzicus and Caenestheria have a short, rounded condyle forming a wide occipital notch. However, several authors synonymised Caenestheria into Eocyzicus and Caenestheriella into Cyzicus, based on developmental studies (e.g., Margalef 1953; Straškraba 1965b; Wiltshire 1973; Rogers et al. 2017), demonstrating that the form of the male rostrum changed with the age of the animal. Regardless of the literature demonstrating that the definitions for these genera were problematic, the names Caenestheria and Caenestheriella continued in use (e.g., Tiwari 1962; Smith and Gola 2001; Timms and Richter 2002; Stoicescu 2004; Olesen and Timms 2005; Richter and Timms 2005; Schmidt and Kiviat 2007). Furthermore, Naganawa (2001) treats Eocyzicus as a synonym of Cyzicus, with no explanation or citations, an approach not followed by subsequent authors. In Leptestheriidae, Daday’s genus Leptestheriella was synonymised with Leptestheria by Brtek (1997). based on “… a series of changes between the two groups.” However, Brtek (1997) presented no citations, data, or explanation for this conclusion, and listed no examined material or any experiments in support. Molecular phylogenetic studies of Spinicaudata have consistently rejected the monophyly of Cyzicidae (Schwentner et al. 2009 2018; Sun et al. 2011; Weeks et al. 2009), with Eocyzicus (not including species originally assigned to Caenestheria or Ozestheria) being more closely related to Leptestheriidae and possibly Limnadiidae rather than with the remaining Cyzicidae (Cyzicidae sensu stricto). The species included in Cyzicidae s.s. did not cluster into the respective traditional genera but fell into two large clades. One clade included all Australian representatives that morphologically resembled Caenestheria and Caenestheriella but Schwentner et al. (2015a) showed that the two genera are not monophyletic when including the Australian representatives. The other clade included species that morphologically represented Caenestheriella and Cyzicus from North America, Japan and Europe (Schwentner et al. 2009; Weeks et al. 2009). These results suggest that apart from Eocyzicus none of the traditional cyzicid genera are monophyletic when using molecular data and question the usefulness of the rostral shape and condyle length as genus defining characters, except for the unique combination of a short, rounded condyle and a broadly spatulate rostrum seen in Eocyzicus. However, the sampling of Cyzicidae outside of Australia was sparse in these previous studies. The phylogenetic relationships between Eocyzicus, Leptestheriidae and Limnadiidae were not well resolved in most studies (Schwentner et al. 2009; Weeks et al. 2009); however, phylogenomic analyses with up to 864 loci strongly supported a sister group relationship of Leptestheriidae and Eocyzicus and Limnadiidae as their closest relative (Schwentner et al. 2018). This contrasts with the morphology based hypothesis that suggested a closer relationship of Leptestheriidae and Cyzicidae (Olesen 1998; Astrop and Hegna 2015), but rather suggests that the ancestor of crown-group Spinicaudata was cyzicid-like. Over the last few years, a number of molecular genetic studies have been published that explored the diversity and/or phylogeny within certain spinicaudatan genera, often restricted to a particular geographic region (e.g., Bellec and Rabet 2016; Cesari et al. 2007; Reed et al. 2015; Schwentner et al. 2011 2012a 2014 2015a 2015b; Weeks et al. 2009 2014). Particularly for Australia, these studies revealed a much larger species diversity than expected, highlighting the need for integrative approaches to fully assess spinicaudatan diversity. These studies provided crucial insight into the diversity and local evolution of spinicaudatan taxa, but their geographic and taxonomic limitations did not allow for conclusions regarding the large-scale evolutionary or biogeographic history of Spinicaudata. All spinicaudatan families and several genera have a nearly global distribution, which could either be due to vicariance events associated with the break up and movement of continental plates or of multiple independent more recent colonization events via transcontinental and transoceanic dispersal. Only for Limnadiidae have more comprehensive analyses been published (Bellec and Rabet 2016; Weeks et al. 2009 2014), including a revision of its genera based on molecular phylogenetics (Rogers et al. 2012). To reconstruct the evolutionary and biogeographic history of Spinicaudata, we bring the various studies together page 3 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan into a single comprehensive analysis. MATERIALS AND METHODS Samples, DNA extraction, PCR and sequencing The goal was to obtain an extensive spinicaudatan dataset that brings together published sequences from various studies as well as new sequence data for species from hitherto poorly studied taxa or regions. Available spinicaudatan mitochondrial COI (cytochrome c oxidase subunit I), mitochondrial 16S rRNA, nuclear EF1α (elongation factor 1-alpha) and nuclear 28S rRNA sequences were downloaded from GenBank (Table S1). The published data had been analyzed to different extents in the associated publications. In those cases, where the available sequences had been used to delineate species using clearly stated species delineation methods (e.g., Schwentner et al. 2014 2015a) or that explicitly studied intraspecific variability (e.g., Cesari et al. 2007) only one specimen per putative species was selected. Limnadiidae have been comprehensively studied in detail in previous studies (Bellec and Rabet 2016; Reed et al. 2015; Rogers et al. 2012, Schwentner et al. 2009, Weeks et al. 2009 2014) and especially for many Eulimnadia species only a single marker was available. Therefore, not all available data were utilized for Limnadiidae, but multiple representatives of each limnadiid genus were included (if available) (Table S1). For Cyzicidae and Leptestheriidae, the goal was also to identify potential cryptic species, therefore, as many representatives as possible were included. However, several individuals with obviously erroneous sequences (e.g., GenBank accession KF966550, which probably is an ostracod sequence) or with only a single locus available were excluded as these were problematic in preliminary analyses (e.g., when individuals of the same species did not share any loci). For Australian Eoleptestheria ticinensis and Eulimnadia sp. C loci were retrieved from published transcriptomes (GenBank BioSample SAMN06174119 and SAMN06174118; Schwentner et al. 2018) and for Eulimnadia texana from the published genome (GenBank BioSample SAMN05965515; Baldwin-Brown et al. 2018) using local BLAST searches. For many published records species or genus names other than the currently recognized names had been used. In all our analyses we applied the currently accepted name following Brtek (1997), Rogers et al. (2012), Schwentner et al. (2015a), Timms and Schwentner (2017), and Rogers (2020) (e.g., for most Cyzicidae s.s. this would be Cyzicus rather than Caenestheriella). We provide the name applied herein as well as the name under which the specific DNA sequences were published (Table S1). Fresh material or soil containing viable eggs of species that were previously not studied genetically were collected in a series of field trips and expeditions performed by various authors and sometimes followed by breeding (Table S1). Because not all four markers studied herein had been sequenced in previous studies, voucher specimens or their DNA were obtained if available (e.g., from Schwentner et al. 2009 2014 2015a 2015b) to provide additional loci for these specimens. DNA was extracted using the Qiagen Blood and Tissue or Qiagen QIAamp DNA Micro kits following the manufacturer’s instructions. PCR reactions comprised 0.05 μl DreamTaq DNA Polymerase, 1.5 μl DreamTaq Table 1. List of primers used in this study. For COI and EF1α different combinations of the available primers were used Primer name Marker Sequence 5'–3' Reference LCO1490 COI GGTCAACAAATCATAAAGATATTGG Forward Folmer et al. (1994) LCO2 COI TCNACHAAYCATAAAGAYATTGGAAC Forward Krebes and Bastrop (pers. com.) HCO2198 COI TAAACTTCAGGGTGACCAAAAAATCA Reverse Folmer et al. (1994) HCO-MZ1-rev COI CTTTVATDCCNGTVGGSACWGCRATAATYAT Reverse Krebes et al. (2010) HCOoutout COI GTAAATATATGNTGNGCTC Reverse Giribet and Edgecombe (2006) HCO-709 COI AATNAGAATNTANACTTCNGGGTG Reverse Blank et al. (2008) 16Sar-L 16S CGC CTG TTT ATC AAA AAC AT Forward Palumbi et al. (1991) 16Sb 16S CTC CGG TTT GAA CTC AGA TCA Reverse Xiong and Kocher (1991) HaF2For1 EF1α GGGYAAAGGWTCCTTCAARTATGC Forward Richter et al. (2007) M44-1 EF1α GCTGAGCGYGARCGTGGTATCAC Forward Cho et al. (1995) 2R53ST EF1α CAGGAAACAGCTATGACGCGAACTTGCAAGCAATGTGAGC Reverse Richter et al. (2007) EF1αreverse EF1α GGAAGTCAGAGAAGGACTC Reverse Braband et al. (2002) D1,D2 fw1 28S AGC GGA GGA AAA GAA ACT A Forward Sonnenberg et al. (2007) D1,D2 rev2 28S ACG ATC GAT TTG CAC GTC AG Reverse Sonnenberg et al. (2007) page 4 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Buffer (Thermo Scientific), 0.12 μl dNTPs mix (25 mM each), 1.5 μl of each primer (10 mM each; see Table 1 for list of primers), and 3 μl DNA extract in a total volume of 15 μl. Temperature regimes were: 94°C for 3 min, 37 amplification cycles of 30 s at 94°C, 45 s at 46°C and 1 min at 72°C for COI; 35 amplification cycles of 30 s at 94°C, 45 s at 50°C and 1 min at 72°C for 16S rRNA; 40 amplification cycles of 30 s at 94°C, 30 s at 51°C and 1 min at 72°C for EF1α; 40 amplification cycles of 30 s at 94°C, 30s at 52.5°C and 1.5 min at 72°C for 28S rRNA and a final elongation step of 5 min at 72°C. Success of PCR reactions was assessed on 1.5% TAE gels. PCR products were cleaned-up with FastAP and Exonuclease I (both ThermoFisher Scientific). Sequencing was conducted either on an ABI Prism 3730xl Genetic Analyzer (Applied Biosystems, Carlsbad, USA) at the Bauer Core Facility at Harvard University or with Macrogen. For Leptestheria sp. from Brazil, L. nobilis Sars, 1900 and Ozestheria sp. from Niger, DNA library preparation was performed using roughly 200 ng of extracted DNA as input. After physical shearing using a bioruptor (15 minutes, HI setting with intervalometer adjusted to 30 s ON, 30 s OFF; Diagenode, Liège, Belgium), we used a standard Ion Xpress library preparation protocol (Ion Xpress n°4471269) with enzyme concentration scaled down to 1:2. We performed the final library size selection with a double SPRI protocol using the following bead/DNA ratios: 0.55 first and then 0.25, to select fragments compatible with the 400 bp sequencing kit of the PGM platform. Sequencing libraries were multiplexed with 12 libraries of other organisms prepared in a similar fashion, using Ion Xpress barcoded adapters (Life Technologies, Carlsbad California). The equimolarity of the library pool was adjusted prior to emulsion PCR (emPCR) via a custom real-time PCR assay (SsoAdvanced supermix; Bio-Rad). No amplification of the libraries was necessary prior to emPCR. An equimolar pool of 20 pM was amplified using the One Touch2 400 bp amplification setup (ion-torrent n°4482002). The sequencing was performed on a 316v2 chip with 850 flows (400 bp sequencing kit; ion-torrent n°4479878). For Cyzicus tetracerus from France, Ozestheria from Thailand and Madagascar, Eocyzicus saharicus Gauthier, 1937, Leptestheria cortieri Daday, 1913, L. dahalacensis Rüppel, 1837 from Austria, L. mayeti (Simon, 1885) from Tunisia, Leptestheria spp. from India and Madagascar and the new limnadiid genus from Bolivia, library preps from DNA were performed with a Nextera XT kit (Illumina): fragmentation and Illumina adapter and index ligation. Equimolar pools of each library were established. Qualification and quantification of the final library was established before sequencing on Illumina Miseq with 2*25 Millions reads cartridge of 300 bases each (30 to 45 libraries per run). All raw sequences were assembled and sequencing errors corrected using Geneious® 6.1.8 or 11.1.4. All sequences were submitted to GenBank (accession numbers MN553596–MN553672 and MN584937– MN585093; Table S1). Alignment and phylogenetic analyses Alignments were performed separately for each marker with MAFFT version 7 (Katoh and Standley 2013) using the slower but more thorough LINS-I option. The cyclestherid Cyclestheria hislopi (Baird, 1859) and the laevicaudatan Lynceus biformis (Ishikawa, 1895) were included as outgroup taxa (Table S1). The alignments of both ribosomal genes included poorly aligned regions associated with numerous indels. To avoid problems in the phylogenetic analyses due to putative erroneous aligned regions, the alignments of 16S and 28S rRNAs were masked with Zorro (Wu et al. 2012) and all positions with scores below 5 were removed. For the two protein-coding genes substitution saturation for codon positions 1 & 2 as well as codon position 3 was tested using the test of Xia et al. (2003) implemented in DAMBE6 (Xia 2017). Translated amino acid sequences were evaluated and assessed for putative stop codons. Aligned sequences for all four genes were concatenated into a single matrix. Best-fitting DNA substitution models were assessed for each gene fragment with MEGA 7 (Kumar et al. 2016) following the Bayesian Information criterion (BIC). Prior to phylogenetic analyses, the separate alignments per gene were concatenated into six different matrices. Matrix 1 included all taxa from all four gene fragments. For Matrix 2, individuals with less than two of the four loci present were excluded (to reduce the impact of missing data) and the 3rd codon position of COI removed (to reduce the impact of substitution saturation that was detected for the 3rd codon position of COI). Matrix 3 and Matrix 4 were taxon-specific matrices that included only representatives of Cyzicidae s.s. For these matrices, new taxon-specific alignments were computed for 16S and 28S rRNA. Because the species are more closely related, these alignments had hardly any indels, which greatly decreased alignment uncertainties and eliminated the need for masking. Thereby particularly variable regions were retained, which might be relevant to resolve the relationships among closely related species. Matrix 3 included all Cyzicidae s.s. and Matrix 4 only those representatives with at least two of the four loci present. Phylogenetic analyses of these two matrices were rooted by the wellsupported split between the two main clades within page 5 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Cyzicidae s.s. (see below). Matrix 5 and Matrix 6 were constructed similarly as Matrices 3 and 4, but taxonspecific for Leptestheriidae and Eocyzicus. Matrix 5 included all representatives of these two taxa and Matrix 6 only those with at least two loci present. The phylogenetic analyses were rooted by the split between Eocyzicus and Leptestheriidae. The 3rd codon position of COI was not removed in any of the taxon-specific matrices as substitution saturation within these taxa was lower. No Limnadiidae-specific matrix was constructed and analysed, as no well-supported point for rooting was available. Phylogenetic analyses were conducted with MrBayes 3.2.6 (Ronquist et al. 2012) and RAxML 8.2.12 (Stamatakis 2014). MrBayes analyses consisted of six runs, four chains and 30*106 generations for each of the six matrices. The first 10% of generations were discarded as burn-in, after assessing convergence with Tracer v1.6 (Rambaut et al. 2013), and the majority rule consensus tree calculated for the remaining generations. The matrices were partitioned by gene fragment and the best-fitting evolutionary model (GTR + I + G) was applied to each partition independently. Maximum likelihood analyses with RAxML were run under the GTR + G model (-m GTRGAMMA) as it is suggested to ignore the invariant parameter in RAxML. Tree searching and bootstrapping was performed in a single run (-f a), using 500 bootstrap replicates and partitioned by genes. All MrBayes and RAxML analyses were run on the CIPRES Science Gateway. The resulting phylogenetic trees were visualized with FigTree 1.4.3 Rambaut 2006–2019). Molecular clock analyses Despite the overall wealth of spinicaudatan fossils, few can be safely assigned to extant spinicaudatan families let alone genera (Astrop and Hegna 2015). To overcome these shortcomings, we used a twostep approach for the molecular clock analyses. First, we used the transcriptome-based data set with 606 genomic loci and an average taxon occupancy per locus of 84% of Schwentner et al. (2018) with several fossil calibration points across Branchiopoda. This data set includes eleven Spinicaudata and allows an overall overview of the timing of key cladogenetic events within Spinicaudata but not detailed age estimates within families or with regards to single biogeographic events. As a second step, we analyzed our speciesrich phylogenetic data set using several of the inferred divergence times within Spinicaudata from the former molecular clock analyses of the amino-acid data set as additional calibration points. We used the amino-acid matrix (matrix 7 of Schwentner et al. 2018), which focused on Branchiopoda with one remipede (Xibalbanus tulumensis (Yager, 1987)) and one cephalocarid (Lightiella incisa Gooding, 1963) as outgroups. The topology was fixed to the topology obtained by Schwentner et al. (2018). Node ages were estimated with PhyloBayes 4.1 (Lartillot and Philippe 2004) using the uncorrelated gamma multipliers (-ugam; Drummond et al. 2006) relaxed clock model with free parameters for birth death priors (-bd) on divergence times. The consensus tree was recovered using the bpcomp command. The calibrated trees were visualized with FigTree 1.4.3. Eleven fossil calibration points were applied in this first molecular clock analysis. For all calibration points, the maximum age was set to 636.1 mya (million years ago) based on the youngest Lagerstätten without known Eumetazoa (following Wolfe et al. 2016). Minimum ages for calibration points were: (1) 497 mya for crown-group Allotriocarida based on Rehbachiella kinnekullensis Muller, 1983 (following Wolfe et al. 2016), representing the root of the tree, (2) 495 mya for the split between Branchiopoda and Remipedia based on the oldest stem group branchiopod (following Harvey et al. 2012), (3) 405 mya for crown-group Branchiopoda based on Lepidocaris rhyniensis (following Olesen 2009 and Wolfe et al. 2016), (4) 405 mya for crowngroup Phyllopoda based on Castracollis wilsonae Fayers & Trewin, 2003 (following Olesen 2009), (5) 121.8 mya for crown-group Notostraca based on Chenops yixianensis Hegna & Dong, 2010 (following Wolfe et al. 2016), (6) 386.9 mya for crown-group Diplostraca based on Leaia chinensis, which was placed by Wolfe et al. (2016) as either crown-group Diplostraca or Onychocaudata (setting this constraint for Diplostraca is more conservative), (7) 250 mya for crown-group Cladoceromorpha based on the oldest known fossil Cyclestherioides (following Raymond 1946 and Negrea et al. 1999), (8) 175 mya for the split between Ctenopoda and all other Cladocera based on fossil ctenopod Smirnovidaphnia smirnovi Kotov, 2007 (following van Damme and Kotov 2016 and Wolfe et al. 2016), (9) 145 mya for the split between Daphnia pulex Leydig, 1860 and its closest relative in the data set (Ceriodaphnia quadrangula (Müller, 1875)) based on the oldest Daphnia fossils (following van Damme and Kotov 2016), (10) 145 may for the split between Simocephalus vetulus (Müller, 1776) and its closest relative in the data set (Scapholeberis mucronata (Müller, 1776)) based on the oldest known Simocephalus fossils (following van Damme and Kotov 2016) and (11) 255 mya for the split between Limnadiidae and Eocyzicus + Leptestheriidae based on the oldest known Perilimnadiidae fossils (following page 6 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Astrop and Hegna 2015), as Perilimnadiidae has been suggested to be ancestral to all Limnadiidae except Limnadopsis, though Astrop and Hegna (2015) questioned the exclusion of Limnadopsis. To test the impact of the spinicaudatan fossil calibration point (calibration point (11)), the PhyloBayes analysis was repeated with the split between Limnadiidae and Eocyzicus + Leptestheriidae set to a minimum of 380 mya based on the assumption that Vertexioidea (known since the mid Devonian) is ancestral to Limnadiidae, while Eosestherioidea is ancestral to Leptestheriidae and Cyzicidae (Astrop and Hegna 2015). For the subsequent molecular clock analyses with our species-rich data set, a new matrix was constructed for the four loci that included only individuals with at least three loci present and only one representative per species. This second molecular clock analysis was run in BEAST v2.4.6 (Bouckaert et al. 2014) as here normal distributions for node ages inferred from the previous molecular clock analysis could be more precisely defined than in PhyloBayes. Tree models were linked across all loci and site and clock models were unlinked. The GTR model with invariant sites and a relaxed log normal clock with a Birth Death model was applied to all partitions. Four BEAST molecular clock analyses were run, one run each using the age estimates derived from the two different fossil calibration points for the split between Limnadiidae and Eocyzicus + Leptestheriidae (see above, calibration point (11)). Each of these were run once with an additional age constraint on the split between Eocyzicus and Leptestheriidae (enforcing monophyly for this node) and once without this constraint (the latter accommodates the possibility that Eocyzicus and Leptestheriidae are not sister taxa). The fossil calibration points for Diplostraca (calibration point (6)) as well as the split between Limnadiidae and Eocyzicus + Leptestheriidae (calibration point (11)) were applied as described above with a uniform distribution. For all calibration points that were derived from inferred nodes of the first molecular clock analysis priors with a normal distribution were applied. Here the inferred mean values were coupled with a sigma that best modelled the 95% Highest Probability Density (HPD) distribution of the inferred node ages. Applied node ages were for the molecular clock analyses with the 255 mya calibration point for Limnadiidae + Leptestheriidae + Eocyzicus: (A) 294.6 mya with a sigma of 25 (403 mya with sigma 10 when 380 mya calibration point for Limnadiidae + Leptestheriidae + Eocyzicus was applied) for crown group Spinicaudata (B) 153.5 mya with sigma of 63 (155.6 mya with sigma 70) for the split between the Australian Limnadiidae and Eulimnadia clades, (C) 66.8 mya with a sigma of 40 (61.6 mya with sigma 30) for the split between Limnadopsis and Paralimnadia, (D) 64.8 mya with a sigma of 32 (67 mya with sigma 35) for the split between O. pilosa and the Australian Ozestheria species and where applicable (E) 146 mya with a sigma of 69 (154 mya with sigma 75) for the split between Eocyzicus and Leptestheriidae. RESULTS The alignments including all spinicaudatans for COI, 16S rRNA, EF1α and 28S rRNA were 599 bp, 449 bp, 746 bp and 675 bp, respectively. In EF1α no substitution saturation was detected, though for COI, substitution saturation was high and significant for the 3rd codon position. After removing all 3rd codon positions, the COI alignment had a length of only 400 bp. For each gene fragment, the GTR + G + I evolutionary model was inferred. No stop codons were detected in COI and EF1α. Phylogenetic analyses recovered monophyly of Limnadiidae and Leptestheriidae, while Cyzicidae was paraphyletic. Exclusion of Eocyzicus, renders Cyzicidae monophyletic, with all the members of Eocyzicus forming an independent clade (Figs. 2, 3; Figs. S1–S9). We refer to the former as Cyzicidae sensu stricto. Leptestheriidae was recovered as sister group to Limnadiidae and Eocyzicus as their closest relatives, each with full support (posterior probability [pp] = 1.0) in the Bayesian analyses, and varying poorly supported relationships among families in the RAxML analyses, but never with Eocyzicus as sister to Cyzicidae s.s. Internal relationships within families were not always consistently recovered (Figs. 2–4, Figs. S1–S14), and for these we focus predominantly on the results obtained with the taxon-specific analyses. In the molecular clock analyses of the transcriptome data set with PhyloBayes, we used the topology by Schwentner et al. (2018), which proposed Eocyzicus to be sister taxon to Leptestheriidae. Here the two alternative calibration points for the split between Limnadiidae and Eocyzicus + Leptestheriidae, despite differing by 125 my, had relatively little influence on the inferred node ages within Spinicaudata. Only for this specific node (mean ages 389 mya vs. 271 mya) and the age of crown-group Spinicaudata (mean ages 403 mya vs. 295 mya) was a larger difference in the age estimates observed (Figs. S10, S11). However, in the subsequent molecular clock analyses with BEAST the age of this calibration point had a stronger effect on the inferred node ages, with differences of ~30% between analyses (Fig. 4, Figs. S12–S14). Constraining the BEAST analyses to force a sister group relationships page 7 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Fig. 2. Phylogenetic relationships of Spinicaudata based on COI, 16S rRNA, EF1α and 28S rRNA inferred with MrBayes. Only individuals with at least two of the four loci available were included and 3rd codon positions of COI were excluded (Matrix 2). For each individual, the country of origin and the collection or voucher number are provided (Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Posterior probabilities and bootstrap support values are provided for each branch. Branches are color-coded according the geographic origin of the specimens. # = node in topology with highest likelihood, but bootstrap support < 50%, - = node not recovered in most topology with highest likelihood. page 8 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan of Leptestheriidae and Eocyzicus did not affect inferred node ages (usually < 1% difference between constrained and unconstrained). In general, older node ages were inferred in the more species-rich BEAST analyses than in the preceding loci-rich PhyloBayes analyses (Fig. 4, Figs. S10–S14). In the following, we focus on the BEAST analyses with the more conservative fossil calibration point (276 mya) (Fig. 4). Cyzicidae s.s. Cyzicidae s.s. is divided into two well-supported clades (each pp = 1.0 and bootstrap support [bs] = 100%) that diverged about 239.1 mya (194.9–282.2 mya 95% HPD) (Figs. 2–4, Figs. S4–S6). One clade includes all Australian representatives (Ozestheria), which constitute a monophyletic group itself (pp = 1.0; bs = 93 and 94%), as well as species from Southeast Asia (Thailand), Central Asia (Mongolia) and Africa (Niger and Madagascar). Ozestheria pilosa (Thailand) or O. pilosa together with the Mongolian species constitute the sister group to all Australian Ozestheria species and the whole clade was dated to 151.1 mya (120–184.3 mya 95% HPD). The divergence between O. pilosa and Australian Ozestheria was dated to 119.9 mya (96.3–141.7 mya 95% HPD, Fig. 4) (only 65 mya in the PhyloBayes analyses, Figs. S10, S11), while the oldest divergence in the Australian clade was estimated to about 94.6 mya (76.2–112.6 mya 95% HPD) (36.6 in the PhyloBayes analyses). The two African species clustered together (pp = 1; bs = 100%) and were sister group to all other species of this clade (pp = 0.94–1.0; bs = 33%; Figs. 2, 3, Figs. S4–S6). The two African species, as well as O. pilosa, were originally assigned to Cyzicus. However, subsequent examinations revealed that they meet the morphological characterization of Ozestheria; i.e., the male claspers (= thoracopod I and thoracopod II) having the elongated, spatulate (clawlike) spines or scales in a transverse row at the endopod apex. The second clade includes Palearctic and Nearctic species from Europe (e.g., France, Austria, Romania and Ukraine), East Asia (Japan), northern Africa (Morocco), Middle East (Jordan) and North America (USA). The North American species constitute a monophyletic group (pp = 1.0; bs = 99–100%) that diverged from all other species of this clade about 116.6 mya (86.2–148.4 mya 95% HPD) (Figs. 2–4). The only possible exception being one specimen identified as Cyzicus setosus (Pearse, 1912) from an unknown location (GenBank accessions DQ310628 and DQ310668, deWaard et al. 2006; C. setosus is a North American species), which did not cluster with other representatives of the same species but rather with the northern African, Japanese and European species. We strongly suspect that this specimen was mislabelled or misidentified. Specimens identified as Cyzicus tetracerus (Krynicki, 1830) did not cluster together and may constitute up to four species, respectively collected in France, Ukraine, Austria/Romania and Jordan (Figs. 2, 3B, Figs. S4–S6). Uncorrected p-distances in COI exceeded 13% between each of these putative species. As C. tetracerus was first described from Kharkiv in the Ukraine, the Ukrainian population may constitute the “true” C. tetracerus. Whether the others represent species new to science or species that have been previously synonymized with C. tetracerus is currently unknown. The North American species Cyzicus setosus and Cyzicus californicus (Packard, 1874) clustered together and share identical 16S sequences, suggesting that these could be conspecific, while the other two North American species Cyzicus mexicanus (Claus, 1860) and Cyzicus gynecius (Mattox, 1950) differed by 3.7–5.5% uncorrected p-distances in 16S. The head morphology (i.e., condyle length and rostrum shape) did not correspond to single clades recovered in the phylogenetic analyses (Fig. 3B). Rather all combinations that defined traditional genera were associated with multiple clades that were not closely related. In several instances groups of closely related species shared identical head morphologies (for example in Eocyzicus or several Ozestheria species). However, in other instances individuals that were genetically so similar that they probably constitute a single species differed in their head morphology. In O. pilosa the length of the condyle varied from short to long, whereas the genetically indistinguishable C. setosus and C. californicus differed in the shape of their rostrum. In general, specimens with the condylerostrum combination that is typical for Eocyicus (short condyle, spatulated rostrum) fell into Eocyzicus and not into Cyzicidae s.s., the only exception being Ozestheria sp. from Mongolia. This specimen had the typical head morphology of Eocyzicus but with an additional spine at the tip of the rostrum (Fig. 1). Eocyzicus The phylogenetic relationships within Eocyzicus were not consistently resolved (Figs. 2, 3A, Figs. S7– S9). In particular, with regards to the positions of the North American E. digueti (Richard, 1895) and the South African Eocyzicus species. While the analyses with Matrices 1 and 2 mainly suggested the latter to be sister group to all other Eocyzicus species (potentially together with Asian, North African and Middle Eastern species) (Fig. 2, Figs. S1–S3), analyses based on Matrices 5 and 6 mainly placed E. digueti as sister page 9 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan as the fossil relationships are not well resolved and still debated (Astrop and Hegna 2015). It is worth noting that all four spinicaudatan families have a nearly global distribution (the Americas, Eurasia, Africa and Australia) and even genera such as Leptestheria, Eocyzicus, Cyzicus and Eulimnadia occur on all or most of these landmasses. Within continents, Spinicaudata have revealed remarkable dispersal potential with species being distributed over > 1000 km and relatively low levels of population differentiation (Cesari et al. 2007; Schwentner et al. 2012a 2014 2015a). Such longdistance dispersal can be mediated via birds or other animal vectors or wind (Bilton et al. 2001) and probably follow the same model as described for the related Anostraca (Rogers 2015 and references cited within). Our molecular clock analyses suggest that the four extant spinicaudatan families diverged prior to the break-up of Gondwana and Laurasia and possibly of Pangea. Also the divergence between species and species groups inhabiting different continents mostly predate the separation of the respective continental plates. Historic vicariance appears to be the main factor explaining the transcontinental distribution of extant spinicaudatan taxa, though in a few instances more recent transoceanic dispersal events have occurred as well. For other large Branchiopoda, like Anostraca, Laevicaudata, and Cyclestherida, vicariance apparently also played a major role in shaping today’s distribution of taxa (Schwentner et al. 2013; Rogers 2015; Sigvardt et al. in press), in particular for southern hemisphere taxa. For Notostraca, previous molecular clock analyses obtained different estimates of their divergence times. While some suggested more recent divergence ages, implying trans-oceanic dispersal events (Mather et al. 2013; Vanschoenwinkel et al. 2012), others suggested divergence times that imply vicariance events (Korn et al. 2013). Our own divergence time estimates within Notostraca (Figs. S10, S11) are similar to the former. One should keep in mind that we predominantly discuss our molecular clock results based on the more conservative calibration point of 255 mya for Limnadiidae + Leptestheriidae + Eocyzicus, the divergence ages inferred using the 380 mya calibration point were even older (Figs. S13–S14) and thus even stronger in suggesting vicariance over dispersal. The divergence between northern and southern hemisphere species of Cyzicidae and Limnadiidae (with the exception of Eulimnadia) probably predates the break-up of Pangea in the early to mid-Jurassic, suggesting that geographic distances might have separated these clades already on Pangea. If Limnadia and Imnadia are indeed nested among the two southern continental clades, a southern hemisphere origin can be hypothesized for Limnadiidae. From there the northern hemisphere could have been colonized when Gondwana and Laurasia were still joined in Pangea. Also for Leptestheriidae, a southern hemisphere origin can be assumed, based on the phylogenetic relationships among its taxa. But here multiple independent colonization events of northern hemisphere continents have to be assumed. The strongly diverging relationships within Leptestheriidae among most analyses do not allow establishing detailed biogeographic hypotheses for this taxon. Most inferred clades within Leptestheriidae do not correspond to geographic regions (e.g., African or Indian species do not form monophyletic groups each) and many deep splits within Leptestheriidae were dated to be 90 mya or older, which roughly coincides with the separation between Africa and South America (~100 mya; McLoughlin 2001) and predates the final break-off of Australia (~50 mya; Beaulieu et al. 2013). Ancestral leptestheriid lineages were probably once widely distributed across Gondwana and when Gondwana broke apart, several of these lineages survived on more than one continent. For example, one leptestheriid clade that was consistently recovered comprised Leptestheria nobilis from India, Leptestheria sp. (specimen M124) from Madagascar and L. brevirostris from Botswana with an estimated clade age of 95 mya. A Madagascar-India-Seychelles block was the first landmass that broke-off from Gondwana around 120 mya (Ali and Aitchinson 2008, McLoughlin 2001) and this leptestheriid clade probably evolved on this landmass and dispersed from Madagascar to continental Africa subsequently. Clades with similar African and Indian distributions were also recovered for the notostracan Triops (e.g., clades 18 and 26 in Korn et al. 2013; see also Modak et al. 2018). While some of these distributions might be due to the geological processes (e.g., clade 18 in Korn et al. 2013), others appear to be younger and thus possibly due to more recent dispersal and colonization events (e.g., clade 26 in Korn et al. 2013). The only spinicaudatan example of repeated transoceanic dispersal is Eulimnadia. This genus probably evolved in South America and successfully dispersed to virtually all other continents (there are no extant records from Antarctica, whether Antarctica was historically inhabited is unknown), as well as many oceanic islands (Bellec and Rabet 2016), probably during the last 50 mya. Bellec and Rabet (2016) dated the onset of this global distribution to only 30 mya. Eulimnadia is special among Spinicaudata due to its very fast development and short life-cycle, which enables them to survive in short-lived habitats and by which they might escape competition and predation from slower developing taxa, as well as the presence of hermaphrodites instead of females (Bellec and page 16 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Rabet 2016). The latter enables the colonization of new habitats from single resting eggs, which greatly improves dispersal effectiveness. The Australian fauna is noteworthy not only because of its exceptional diversity (Schwentner et al. 2015b), but it apparently evolved independently from other regions even when Australia was still connected to other Gondwanan landmasses. This seems to have been the case for the Limnadopsis-ParalimnadiaAustralimnadia clade within Limnadiidae and the Australian Ozestheria clade. In both cases, age estimates suggest that the Australian clades (134 and 94 mya, respectively) evolved long before the final separation of Australia from the rest of Gondwana (Beaulieu et al. 2013), which occurred around 50 mya. In this relatively long timespan, apparently no exchange between Australia and other continental masses occurred. Such patterns of ancestral regionalization were also found in many other old Gondwanan lineages (e.g., Murienne et al. 2014). Australia was linked to the rest of Gondwana primarily via Antarctica, the only continent without extant Spinicaudata but with a rich fossil record of these animals (Shen 1994). Antarctica may have restricted exchange between Australia and other regions of Gondwana simply by geographic distance (isolation by distance). The age of the Australian Eocyzicus clade roughly coincides with the final break-off of Australia, implying that Australia was colonized by Eocyzicus around that time. Again, Eulimnadia may be the only taxon that colonized Australia after it separated from Gondwana, potentially as Australia drew closer to Asia. It has been suggested that Eoleptestheria cf. ticinensis invaded Australia recently from China (Timms 2009a); however, our phylogenetic analyses, as well as previous analyses of the genetic diversity within the Australian populations (Schwentner et al. 2015b), suggest a longer presence of this taxon in Australia. Notably, for the large branchiopod taxa Cyclestherida and Notostraca, younger ages have been assumed for the divergence of Australian and non-Australian species (Schwentner et al. 2013; Mathers et al. 2013). In these taxa, dispersal to or from Australia apparently occurred more recently, probably after Australia and Asia moved closer. The case of the notostracan Triops is particularly interesting as the putatively closest relatives to the Australian fauna occur in North America (Mathers et al. 2013). The age estimates presented herein may also help to improve our understanding of the relationships between fossil and extant taxa (Astrop and Hegna 2015). We provide the first molecular clock based age estimated for all extant families and many genera. Our results suggest that within each extant family only one to three extant lineages date back more than 150 million years and the main divergence took place within the last 100 million years. Thus the majority of fossil families probably went extinct without any extant representatives; this may be particularly true for the rich Permian and Carboniferous fauna as crowngroup Spinicaudata may have only originated around that time. The similar carapace shapes of Cyzicidae and Eocyzicidae might go back to comparable carapace shapes already known from different Euestheriidae fossils since the Permian (Astrop and Hegna 2015). Nevertheless, peculiar similarities in carapace shape between fossil and extant taxa may be generally due to convergence rather than evolutionary stasis. For example, the carapace of the fossil Palaeolimnadiopseidae Defretin-Le Franc, 1965 has large similarity to extant species of Limnadopsis and Australimnadia (for example, compare Gallego and Breitkreuz 1994 and Gallego 2005 with Timms 2009b or Timms and Schwentner 2012) and it has been suggested that Palaeolimnadiopseidae are the ancestors of Limnadopsis (Zhang et al. 1976, but questioned by Astrop and Hegna 2015). Palaeolimnadiopseidae date back as far as the Upper Permian (summarized in Gallego 2005) long before the inferred evolution of Limnadiidae, potentially even before the evolution of crown-group Spinicaudata. Of course, it is possible that some younger species that have been assigned to Palaeolimnadiopseidae belong to the stem lineage of extant Limnadiidae and are not related to the older taxa (Astrop and Hegna 2015). Our divergence time estimates may help to improve the current hypotheses of how such fossil and extant taxa may have been related. Species Diversity of Spinicaudata Detailed population-based molecular genetic and integrative taxonomic approaches have revealed much higher species diversities for Spinicaudata (e.g., Schwentner et al. 2011 2014 2015a b; Weeks et al. 2009) and other ‘large Branchiopoda’ like Triops (e.g., Korn et al. 2010; Mathers et al. 2013; Meusel and Schwentner 2017; Vanschoenwinkel et al. 2012). Several species that were assumed to be morphologically variable could be shown to represent an amalgam of multiple, morphologically differentiated, species (e.g., Korn et al. 2010; Schwentner et al. 2012b; Meusel and Schwentner 2017; Tippelt and Schwentner 2018). However, extensive overlaps of intraspecific variability and interspecific variation are prevalent also in these species and their initial delimitation would have been difficult based solely on morphological characters. The majority of these studies have been conducted on the Australian fauna, which now appears to be the continent with the highest extant clam shrimp diversity, harbouring roughly one third of all spinicaudatan page 17 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan species (Schwentner et al. 2015b). However, the spinicaudatan fauna of other continents have not been studied as extensively. In the analyses presented herein, many species were represented by single individuals or were studied from a few populations only. Despite this relatively sparse intraspecific sampling, instances of putatively cryptic species were revealed in Africa, North America and Europe; in the case of Leptestheria rubidgei even within a single population. On the one hand, this suggests that the species diversity of Spinicaudata may be underestimated on local and global scales, on the other hand it shows that the species level taxonomy of many spinicaudata taxa requires revision. Taxonomic revisions that combine detailed morphological and molecular genetic data will become indispensable to assess the true diversity of Spinicaudata and will probably reveal many more currently cryptic species. Acknowledgments: This work (urn:lsid:zoobank. org:pub:665CB83B-7A2A-445F-9154-B18BDCC A573C) and Eocyzidae fam. nov. (urn:lsid:zoobank. org:act:CF3B4318-483E-47B8-A391-DA6A9E77400F) have been registered with ZooBank. We are greatly indebted to Brian V. Timms, who has made this research possible through his relentless efforts in studying Australian Branchiopoda and has taken MS, NR, SR, and DCR on numerous collecting expeditions across Australia. We thank Mikhail Son (Institute of Marine Biology, Odessa, Ukraine) and Federico Marrone (Università degli Studi di Palermo, Palermo, Italy) for providing specimens as well as two anonymous reviewers for their helpful comments and Thomas Hegna for his patience and support. We thank also Daniela Lupi, Eric Gallerne, Eric Quéinnec, Marco Pagni, Michaël Manuel, Pascal Lluch, Sébastien Lacau for the help in the field or soil sampling. Sequencing costs were financed via internal funds from the Museum of Comparative Zoology and the Faculty of Arts and Sciences to G.G. and from the Center of Natural History (Universität Hamburg) to M.S. and via Emergence project of UPMC and ATM gave to N.R. This work benefited from equipment and services from the iGenSeq core facility, at ICM, Sorbonne-Unversité and SSM in MNHN (Paris). Authors’ contributions: M.S. and D.C.R. conceived and planned the study. M.S. performed all Sanger sequencing and phylogenetic analyses. D.C.R. updated the diagnoses of the taxa. M.S., D.C.R. and N.R. drafted the manuscript. All authors contributed to the final version of the manuscript. Competing interests: The authors declare that they have no conflict of interests. Availability of data and materials: All sequences were submitted to GenBank (accession numbers MN553596–MN553672 and MN584937–MN585093). Specimens are stored in museum collections (Table S1). Further data can be obtained form the corresponding author upon request. Consent for publication: All authors consent to the publication of this manuscript. Ethics approval consent to participate: Not applicable. REFERENCES Ali JR, Aitchison JC. 2008. Gondwana to Asia: Plate tectonics, paleogeography and the biological connectivity of the Indian sub-continent from the Middle Jurassic through latest Eocene (166–35 Ma). Earth Sci Rev 88:145–166. doi:10.1016/ j.earscirev.2008.01.007. Astrop TI, Hegna TA. 2015. Phylogenetic relationships between living and fossil spinicaudatan taxa (Branchiopoda Spinicaudata): Reconsidering the evidence. J Crustacean Biol 35:339–354. doi:10.1163/1937240X-00002317. Audouin V. 1837. Seance du 1 fevrier 1837. Ann Soc Entomol France 6:9–11. Baird W. 1862. Description of seven new species of phyllopodous crustaceans, belonging to the genera Estheria and Limnetis. Proc Zool Soc London 1862:147–149. Baldwin-Brown JG, Weeks SC, Long AD. 2018. A new standard for crustacean genomics: The highly contiguous, annotated genome assembly of the clam shrimp Eulimnadia texana reveals HOX gene order and identifies the sex chromosomes. Genome Biol Evol 10:143–156. doi:10.1093/gbe/evx280. Barnard KH. 1924. Contributions to a knowledge of the fauna of southwest Africa, II. Crustacea Entomostraca, Phyllopoda. Ann S Afr Mus 20:213–228. Barnard KH. 1929. Contributions to the crustacean fauna of South Africa, 10. Revision of Branchiopoda. Ann S Afr Mus 29:189– 272. Beaulieu JM, Tank DC, Donoghue MJ. 2013. A Southern Hemisphere origin for campanulid angiosperms, with traces of the break-up of Gondwana. BMC Evol Biol 13:80. doi:10.1186/1471-214813-80. Bellec L, Rabet N. 2016. Dating the Limnadiidae family suggests an American origin of Eulimnadia. Hydrobiol 773:149–161. doi:10.1007/s10750-016-2694-x. Bilton DT, Freeland JR, Okamura B. 2001. Dispersal in freshwater invertebrates. Ann Rev Ecol Syst 32:159–181. doi:10.1146/ annurev.ecolsys.32.081501.114016. Blank M, Laine AO, Jürss K, Bastrop R. 2008. Molecular identification key based on PCR/RFLP for three polychaete sibling species of the genus Marenzelleria, and the species’ current distribution in the Baltic Sea. Helgol Mar Res 62:129– 141. doi:10.1007/s10152-007-0081-8. Bouckaert R, Heled J, Kühnert D, Vaughan T, Wu CH, Xie D, Suchard MA, Rambaut A, Drummond AJ. 2014. BEAST 2: A software platform for Bayesian evolutionary analysis. PLoS Comp Biol 10:e1003537. doi:10.1371/journal.pcbi.1003537. Braband A, Richter S, Hiesel R, Scholtz G. 2002. Phylogenetic page 18 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan relationships within the Phyllopoda (Crustacea, Branchiopoda) based on mitochondrial and nuclear markers. Mol Phylogenet Evol 25:229–244. doi:10.1016/S1055-7903(02)00253-1. Brady GS. 1886. Notes on freshwater Entomostraca from South Australia. Proc Zool Soc London 1886:82–93. Brehm V. 1933. Phyllopoden. Mitteilungen von der WallaceaExpedition Woltereck, 5. Zool Anz 104:31–40. Brtek J. 1997. Checklist of the valid and invalid names of the “large branchiopods” (Anostraca, Notostraca, Spinicaudata and Laevicaudata), with a survey of the taxonomy of all Branchiopoda. Zbor Slov Nar Muz, Prir Vedy 43:3–66. Brtek J, Forró J, Ponyi JE. 1984. Contribution to the knowledge of the Branchiopoda (Crustacea) fauna of Mongolia. Annal Hist-Natur Mus Nat Hun 76:91–99. Botnariuc N. 1945. Révision des Phyllopodes Conchostracés de la collection du Professeur I. Borcea. Bull Sect Sci Acad Roum 28:555–561. Botnariuc N. 1947. Contribution à la connaissance des Phyllopodes Conchostracés de Roumanie. Notationes Biol 5:68–158. Brongniart A. 1820. Mémoires du Muséum d’Histoire Naturelle vol. 6. Paris, France. Burmeister H. 1843. Organisation der Trilobiten aus ihren lebenden Verwandten entwickelt. Berlin, Germany. Cesari M, Luchetti A, Scanabissi F, Mantovani B. 2007. Genetic variability in European Leptestheria dahalacensis (Rüppel, 1837) (Crustacea, Branchiopoda, Spinicaudata). Hydrobiologia 586:249–260. doi:10.1007/s10750-007-0645-2. Cho S, Mitchell A, Regier JC, Mitter C, Poole RW, Friedlander TP, Zhao S. 1995. A highly conserved nuclear gene for low-level phylogenetics: Elongation factor-1α recovers morphology-based tree for heliothine moths. Mol Biol Evol 12:650–656. Daday E. 1913a. Az eddig ismert kagyios levellabu rakok attekintese. Mat Term Ertes (Budapest) 31:559–601. Daday E. 1913b. Deux aberrations intérressantes dans le sous-ordre Phyllopoda Conchostraca. Ann Sci Natur Zool 17:195–206. Daday E. 1914. Monographie systématique des Phyllopodes Conchostracés. I. Caenestheriidae. Ann Sci Natur Zool 9e séries 20:39–330. Daday E. 1915. Monographie Systématique des Phyllopodes Conchostracés. Ann Sci Natur Zool 9e séries 20:39–330. Daday E. 1923. Monographie systématique des Phyllopodes Conchostracés. II. Leptestheriidae. Ann Sci Natur Zool 10e séries 6:255–390 (= 331–446). Daday E. 1925. Monographie systématique des Phyllopodes Conchostracés. III. Limnadiidae. Ann Sci Natur Zool 10e séries 8:143–184 (= 463–504). Daday E. 1926. Monographie systématique des Phyllopodous Conchostracés. III. Limnadiidae (suite). Ann Sci Natur Zool 10e séries 9:1–81 (= 505–586). Drummond AJ, Ho SYW, Philipps MJ, Rambaut A. 2006. Relaxed phylogenetics and dating with confidence. PLoS Biol 4:699–710. doi:10.1371/journal.pbio.0040088. deWaard JR, Sacherova V, Cristescu MEA, Remigio EA, Crease TJ, Hebert PDN. 2006. Probing the relationships of the branchiopod crustaceans. Mol Phylogenet Evol 39:491–502. doi:10.1016/ j.ympev.2005.11.003. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3:294–299. Gauthier MH. 1933. Note sur certains Conchostracés de l'Algérie et de la Tunisie. Bull Soc Hist nat Afr Nord 25:117–126. Gauthier MH. 1937. Euphyllopodes et cladocérans continentaux récoltés par M. Monod au Sahara Occidental, et en Mauritanie. Bull Soc Sci Nat Maroc 17:1–24. Gallego OF, Breitkreuz C. 1994. Conchostracos (CrustaceaeConchostraca) paleozoicos de la Región de Antofagasta, norte de Chile. Rev Geol Chile 21:31–53. Gallego OF. 2005. First record of the family Palaeolimnadiopseidae Defretin-Le Franc, 1965 (Crustacea-Conchostraca) in the Triassic of Argentina. J S Am Earth Sci 18:223–231. doi:10.1016/ j.jsames.2004.10.002. Giribet G, Edgecombe G. 2006. Conflict between datasets and phylogeny of centipedes: an analysis based ons even genes and morphology. Proc R Soc B 273:531–538. doi:10.1098/ rspb.2005.3365. Harvey THP, Valez MI, Butterfield NJ. 2012. Exceptionally preserved crustaceans from western Canada reveal a cryptic Cambrian radiation. Proc Natl Acad Sci USA 109:1589–1594. doi:10.1073/ pnas.1115244109. Hertzog L. 1935. Crustacés. Notes faunistiques de la Camargue, 1. Bull Soc Zool France 60:265–281. Hoeh WR, Smallwood ND, Senyo DM, Chapman EG, Weeks SC. 2006. Evaluating the monophyly of Eulimnadia and the Limnadiidae (Branchiopoda: Spinicaudata) using DNA sequences. J Crustacean Biol 26:182–192. doi:10.1651/C-2623.1. Ishikawa C. 1895. Phyllopod Crustacea of Japan. Zool Mag 7:1–5, plus 2 plates. Katoh K, Standley DM. 2013. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol 30:772–780. doi:10.1093/molbev/mst010. Korn M, Green AJ, Machado M, García-de-Lomas J, Cristo M, Cancela da Fonseca L, Frisch D, Pérez-Bote JL, Hundsdoerfer AK. 2010. Phylogeny, molecular ecology and taxonomy of southern Iberian lineages of Triops mauritanicus (Crustacea: Notostraca). Org Divers Evol 10:409–440. doi:10.1007/s13127010-0026-y. Korn M, Rabet N, Ghate HV, Marrone F, Hundsdoerfer AK. 2013. Molecular phylogeny of the Notostraca. Mol Phylogenet Evol 69:1159–1171. doi:10.1016/j.ympev.2013.08.006. Krebes L, Blank M, Jurss K, Zettler ML, Bastrop R. 2010. Glacialdriven vicariance in the amphipod Gammarus duebeni. Mol Phylogenet Evol 54:372–385. doi:10.1016/j.ympev.2009.07.034. Kumar S, Stecher G, Tamura K. 2016. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol Biol Evol 33:1870–1874. doi:10.1093/molbev/msw054. Lartillot N, Philippe H. 2004. A Bayesian mixture model for acrosssite heterogeneities in the amino acid replacement process. Mol Biol Evol 21:1095–1109. doi:10.1093/molbev/msh112. Lartillot N, Rodrigue N, Stubbs D, Richer J. 2013. PhyloBayes MPI: phylogenetic reconstruction with infinite mixtures of profiles in a parallel environment. Syst Biol 62:611–615. doi:10.1093/sysbio/ syt022. Linder F. 1945. Affinities within the Branchiopoda with notes on some dubious fossils. Arkiv Zool 37A:1–28. Linnaeus C. 1761. Fauna Svecica, Sistens Animalia Sveciae Regni: Mammalia, Aves, Amphibia, Pisces, Insecta, Vermes. Distributa per Classes et Ordines, Genera and Species. Laurentii Salviaw, Stockholm, Sweden. Marinček M, Petrov B. 1985. Taxonomical study of the genus Leptestheria (Conchostraca, Crustacea) I. Glas Prirod Muz Beogradu B40:97–111. Margalef R. 1953. Los crustáceos de las aguas continentales ibéricas. Biol Aguas Continent 10:1–243. Mathers TC, Hammond RL, Jenner RA, Hänfling B, Gómez A. 2013. Multiple global radiations in tadpole shrimps challenge the concept of ‘living fossils’. PeerJ 1:e62. doi:10.7717/peerj.62. Mattox NT. 1950. Notes on the life history and description of a new species of conchostracan phyllopod, Caenestheriella gynecia. Trans Am Mic Soc 69:50–53. page 19 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan McLoughlin S. 2001. The breakup history of Gondwana and its impact on pre-Cenozoic floristic provincialism. Aust J Bot 49:271–300. doi:10.1071/BT00023. Meusel F, Schwentner M. 2017. Molecular and morphological delimitation of Australian Triops species (Crustacea: Branchiopoda: Notostraca) - large diversity and little morphological differentiation. Org Divers Evol 17:137–156. doi:10.1007/s13127-016-0306-2. Modak N, Korn M, Padhye SM. 2018. Molecular phylogenetic investigations of Triops granarius (Lucas, 1864) (Branchiopoda: Notostraca) from the type locality of the former Apus orientalis Tiwari, 1952 and three other localities in the Western Ghats of India. Zootaxa 4531:541–553. doi:10.11646/zootaxa.4531.4.5. Murienne J, Daniels SR, Buckley TR, Mayer G, Giribet G. 2014. A living fossil tale of Pangaean biogeography. Proc R Soc B 281:20132648. doi:10.1098/rspb.2013.2648. Naganawa H. 2001. Current prospect of the recent large branchiopodan fauna of East Asia: 3. revised classification of the recent Spinicaudata. Aquabiology 23:291–299. Negrea S, Botnariuc N, Dumont HJ. 1999. Phylogeny, evolution and classification of the Branchiopoda (Crustacea). Hydrobiologia 412:191–212. Olesen J. 1998. A phylogenetic analysis of the Conchostraca and Cladocera (Crustacea, Branchiopoda, Diplostraca. Biol J Linn Soc 122:491–536. Olesen J. 2007. Monophyly and phylogeny of Branchiopoda, with focus on morphology and homologies of branchiopod phyllopodous limbs. J Crustacean Biol 27:165–183. doi:10.1651/ S-2727.1. Olesen J. 2009. Phylogeny of Branchiopoda (Crustacea) - Character evolution and contribution of uniquely preserved fossils. Arthr Syst Phylog 67:3–39. Olesen J, Grygier M. 2004. Larval development of Japanese ‘conchostracans’: part 2, larval development of Caenestheriella gifuensis (Crustacea, Branchiopoda, Spinicaudata, Cyzicidae), with notes on homologies and evolution of certain naupliar appendages within the Branchiopoda. Arthr Struc Dev 33:453– 469. doi:10.1016/j.asd.2004.07.001. Olesen J, Timms BV. 2005. Caenestheriella mariae sp. nov. (Crustacea: Branchiopoda: Spinicaudata: Cyzicidae): a new clam shrimp from Western Australia. Zootaxa 824:1–8. doi:10.11646/ zootaxa.824.1.1. Packard AS. 1871. Preliminary notice of new North American Phyllopoda. Am J Sci (1820–1879) 2, 8:108–114. Packard AS. 1874. Synopsis of the freshwater Phyllopod Crustacea of North America. Rept Geol Surv Terr 7:613–622. Packard AS. 1877. Descriptions of new phyllopod Crustacea from the west. Bull US Geol Geog Surv Terr 3:171–179. Palumbi SR, Martin A, Romano S, Mcmillan WO, Stice L, Grabowski G. 1991. The simple fool’s guide to PCR. Acollection of PCR protocols, version 2. Honolulu: University of Hawaii. Pearse AS. 1912. Notes on Phyllopod Crustacea. Report Michigan Acad Sci 14:191–197. Rambaut A. 2006–2019. FigTree: Tree Figure Drawing Tool version 1.4.4. Source code available from: http://github.com/rambaut/ figtree/. Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, Scotland, UK. Rambaut A, Suchard MA, Xie D, Drummond AJ. 2013. Tracer v1.6, Available from http://beast.bio.ed.ac.uk/Tracer. Accessed 2017. Raymond PE. 1946. The genera of fossil Conchostraca – an order of bivalved Crustacea. Bull Mus Comp Zool 96:217–307. Reed SK, Duff RJ, Weeks SC. 2015. A systematic study of the genus Eulimnadia. J Crustacean Biol 35:379–391. doi:10.1163/1937240X-00002345. Richter S, Timms BV. 2005. A list of the recent clam shrimps (Crustacea: Laevicaudata, Spinicaudata, Cyclestherida) of Australia, including a description of a new species of Eocyzicus. Rec Aust Mus 57:341–354. doi:10.3853/j.00671975.57.2005.1454. Richter S, Olesen J, Wheeler WC. 2007. Phylogeny of Branchiopoda (Crustacea) based on a combined analysis of morphological data and six molecular loci. Cladistics 23:301–336. doi:10.1111/ j.1096-0031.2007.00148.x. Rogers DC. 2015. A conceptual model for anostracan biogeography. J Crustacean Biol 35: 686–699. doi:10.1163/1937240X-00002369. Rogers DC. 2020. Spinicaudata Catalogus (Crustacea: Branchiopoda). Zool Stud 59:45. doi:10.6620/ZS.2020.59-45. Rogers DC, Rabet N, Weeks SC. 2012. Revision of the extant genera of Limnadiidae (Branchiopoda: Spinicaudata). J Crustacean Biol 32:827–842. doi:10.1163/193724012X637212. Rogers DC, Chang TC, Wang Y-C. 2017. A new Eocyzicus (Branchiopoda: Spinicaudata) from Taiwan, with a review of the genus. Zootaxa 4318:254–270. doi:10.11646/zootaxa.4318.2.2. Rogers DC, Thaimuangphol W, Saengphan N, Sanoamuang L. 2013. Current knowledge of the South East Asian large branchiopod Crustacea (Anostraca, Notostraca, Laevicaudata, Spinicaudata, Cyclestherida). J Limnol 72:69–80. doi:10.4081/jlimnol.2013. s2.e5. Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2: Efficient Bayesian phylogenetic inference and model selection across a large model space. Syst Biol 61:539– 542. doi:10.1093/sysbio/sys029. Sars OG. 1898. Description of two additional South African Phyllopoda. Arch Math Naturvidenskab 20:1–23. Schmidt RE, Kiviat E. 2007. State records and habitat of clam shrimp, Caenestheriella gynecia (Crustacea: Conchostraca), in New York and New Jersey. Can Field Nat 121:128–132. doi:10.22621/cfn. v121i2.435. Schwentner M, Clavier S, Fritsch M, Olesen J, Padhye S, Timms BV, Richter S. 2013. Cyclestheria hislopi (Crustacea: Branchiopoda): a group of morphologically cryptic species with origins in the Cretaceous. Mol Phylogenet Evol 66:800–810. doi:10.1016/ j.ympev.2012.11.005. Schwentner M, Just F, Richter S. 2015a. Evolutionary systematics of the Australian Cyzicidae (Crustacea, Branchiopoda, Spinicaudata) with the description of a new genus. Zool J Linn Soc 173:271–295. doi:10.1111/zoj.12209. Schwentner M, Richter S, Rogers DC, Giribet G. 2018. Tetraconatan phylogeny with special focus on Malacos t raca and Branchiopoda: highlighting the strength of taxon-specific matrices in phylogenomics. Proc R Soc B 285:20181524. doi:10.1098/rspb.2018.1524. Schwentner M, Timms BV, Bastrop R, Richter S. 2009. Phylogeny of Spinicaudata (Branchiopoda, Crustacea) based on three molecular markers – an Australian origin for Limnadopsis. Mol Phylogenet Evol 53:716–725. doi:10.1016/j.ympev.2009.07.021. Schwentner M, Timms BV, Richter S. 2011. An integrative approach to species delineation incorporating different species concepts: a case study of Limnadopsis (Branchiopoda: Spinicaudata). Biol J Linn Soc 104:575–599. doi:10.1111/j.1095-8312.2011.01746.x. Schwentner M, Timms BV, Richter S. 2012a. Flying with the birds? Recent large-area dispersal of four Australian Limnadopsis species (Crustacea: Branchiopoda: Spinicaudata). Ecol Evol 2:1605–1626. doi:10.1002/ece3.265. Schwentner M, Timms BV, Richter S. 2012b. Description of four new species of Limnadopsis from Australia (Crustacea: Branchiopoda: Spinicaudata). Zootaxa 3315:42–64. page 20 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan doi:10.11646/zootaxa.3315.1.2. Schwentner M, Timms BV, Richter S. 2014. Evolutionary systematics of the Australian Eocyzicus fauna (Crustacea: Branchiopoda: Spinicaudata) reveals hidden diversity and phylogeographic structure. J Zool Syst Evol Res 52:15–31. doi:10.1111/jzs.12038. Schwentner M, Timms BV, Richter S. 2015b. Spinicaudata (Branchiopoda: Diplostraca) in Australia’s arid zone: unparalleled diversity at regional scales and within water bodies. J Crustacean Biol 35:366–378. doi:10.1163/1937240X-00002339. Shen YB. 1994. Jurassic conchostracans from Carapace Nunatak, southern Victoria Land. Antarctica. Antarc Sci 6:105–113. Shu S, Rogers DC, Chen X, Yang J. 2015. Two new species of clam shrimp (Branchiopoda: Spinicaudata) from Yunnan Province, China. J Crustacean Biol 35:454–460. doi:10.1163/1937240X-00002338. Sigvardt ZMS, Olesen J, Rogers DC, Timms BV, Palero F. in press. The Gondwana connection: molecular phylogenetics suggests vicariant speciation in smooth clam shrimps (Branchiopoda, Laevicaudata). Zoologica Scripta. Smith DG, Gola AA. 2001. The discovery of Caenestheriella gynecia Mattox, 1950 (Branchiopoda, Cyzicidae) in New England, with ecological and systematic notes. Northeastern Naturalist 8:443– 454. doi:10.2307/3858448. Sonnenberg R, Nolte AW, Tautz D. 2007. An evaluation of LSU rDNA D1-D2 sequences for their use in species identification. Front Zool 4:6. doi:10.1186/1742-9994-4-6. Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30:1312–1313. doi:10.1093/bioinformatics/btu033. Stoicescu A. 2004. Caenestheriella variabilis (Daday) (Conchostraca: Crustacea) a species new to the faune of Romania and its validity. Rev Roumanie Biol 49:11–18. Straškraba M. 1965a. Taxonomic studies on Czechoslovak Conchostraca, 1. Family Limnadiidae. Crustaceana 9:263–273. Straškraba M. 1965b. Taxonomical studies on Czechoslovak Conchostraca II. Families Lynceidae and Cyzicidae. Vestnik Cesk Spol Zool 29:205–214. Straškraba M. 1966. Taxonomical studies on Czechoslovak Conchostraca III. Family Leptestheriidae, with some remarks on the variability and distribution of Conchostraca and a key to the Middle-European species. Hydrobiologia 27:571–589. doi:10.1007/BF00042714. Sun X, Xia X, Yang Q. 2011. Phylogeny of Conchostraca based on 28S rDNA D1-D2 and partial 16S rDNA sequences. Acta Micropalaeontol Sin 28:370–380. Timms BV, Richter S. 2002. A preliminary analysis of the conchostracans (Crustacea: Spinicaudata and Laevicaudata) of the middle Paroo catchment of the Australian arid-zone. Hydrobiologia 486:239–247. doi:10.1023/A:1021315221708. Timms BV. 2009a. First records of a leptestherid clam shrimp in Australia (Crustacea, Spinicaudata, Leptestheriidae, Eoleptestheria). ZooKeys 18:1–16. doi:10.3897/zookeys.18.92. Timms BV. 2009b. A revision of the Australian endemic clam shrimp Limnadopsis Spencer and Hall (Crustacea: Branchiopoda: Spinicaudata: Limnadiidae). Rec Austr Mus 61:49–72. doi:10.3853/j.0067-1975.61.2009.1498. Timms BV, Schwentner M. 2012. A new genus and species of large limnadiid clam shrimp from Australia (Spinicaudata: Limnadiidae). J Crustacean Biol 32:981–990. doi:10.1163/ 1937240X-00002098. Tippelt L, Schwentner M. 2018. Taxonomic assessment of Australian Eocyzicus species (Crustacea: Branchiopoda: Spinicaudata). Zootaxa 4410:401–452. doi:10.11646/zootaxa.4410.3.1. Tiwari KK. 1962. New species of Conchostraca (Crustacea: Phyllopoda) from Rajastan. Proc All-India Cong Zool 1:180–190. Uéno M. 1927. On some freshwater branchiopods of China. Ann Zool Jap 11:157–165. van Damme K, Kotov AA. 2016. The fossil record of the Cladocera (Crustacea: Branchiopoda): evidence and hypotheses. Earth-Sci Rev 163:162–189. doi:10.1016/j.earscirev.2016.10.009. Vanschoenwinkel B, Pinceel T, Vanhoe MPM, Denis C, Jocque M, Timms BV, Brendonck L. 2012. Toward a global phylogeny of the “living fossil” crustacean order of the Notostraca. PLoS ONE 7:e34998. doi:10.1371/journal.pone.0034998. Vecchi A. 1922. Nuova specie di Concostraco di Cirenaica. Soc Ital Sci Natur Mus 61:58–67. Weeks SC, Brantner JS, Astrop TI, Ott DW, Rabet N. 2014. The evolution of hermaphroditism from dioecy in crustaceans: selfing hermaphroditism described in a fourth spinicaudatan genus. Evol Biol 41:251–261. doi:10.1007/s11692-013-9265-0. Weeks SC, Chapman EG, Rogers DC, Senyo DM, Hoeh WR. 2009. Evolutionary transitions among dioecy, androdioecy and hermaphroditism in limnadiid clam shrimp (Branchiopoda: Spinicaudata). J Evol Biol 22:1781–1799. doi:10.1111/j.14209101.2009.01813.x. Wiltshire CT. 1973. The developmental morphology of Cyzicus morsei (Packard) (Crustacea: Conchostraca) from hatching through adulthood with comments on taxonomy within the family Cyzicidae. PhD dissertation, University of Missouri, Columbia, Missouri, USA. Wolfe JM, Daley AC, Leff DA, Edgecombe GD. 2016. Fossil calibrations for the arthropod Tree of Life. Earth-Sci Rev 160:43–110. doi:10.1016/j.earscirev.2016.06.008. Wu M, Chatterji S, Eisen JA. 2012. Accounting for alignment uncertainty in phylogenomics. PLoS ONE 7:e30288. doi:10.1371/journal.pone.0030288. Xia X. 2017. DAMBE6: New tools for microbial genomics, phylogenetics and molecular evolution. J Hered 108:431–437. doi:10.1093/jhered/esx033. Xia X, Xie Z, Salemi M, Chen L, Wang L. 2003. An index of substitution saturation and its application. Mol Phylogenet Evol 26:1–7. doi:10.1016/S1055-7903(02)00326-3. Xiong B, Kocher TD. 1991. Comparison of mitochondrial DNA sequences of seven morphospecies of balck flies (Diptera: Simuliidae). Genome 34:306–311. doi:10.1139/g91-050. Zhang WT, Chen PJ, Shen YB. 1976. Fossil Conchostraca of China. Science Press, Beijing, China. page 21 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan Supplementary Materials Fig. S1. Phylogenetic relationships of Spinicaudata based on COI, 16S rRNA, EF1α and 28S rRNA inferred with RAxML. All individuals and codon positions were included (Matrix 1). For each individual, the country of origin and the collection or voucher number are provided (see also Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Bootstrap support values are provided for each node. (download) Fig. S2. Phylogenetic relationships of Spinicaudata based on COI, 16S rRNA, EF1α and 28S rRNA inferred with MrBayes. All individuals and codon positions were included (Matrix 1). For each individual, the country of origin and the collection or voucher number are provided (see also Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Posterior probabilities are provided for each node. (download) Fig. S3. Phylogenetic relationships of Spinicaudata based on COI, 16S rRNA, EF1α and 28S rRNA inferred with RAxML. Only individuals with at least two of the four loci available were included and 3rd codon positions of COI were excluded (Matrix 2). For each individual, the country of origin and the collection or voucher number are provided (Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Bootstrap support values are provided for each node. (download) Fig. S4. Phylogenetic relationships of Cyzicidae s.s. based on COI, 16S rRNA, EF1α and 28S rRNA with RAxML. All individuals and codon positions were included (Matrix 3). For each individual, the country of origin and the collection or voucher number are provided (Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Bootstrap support values are provided for each node. (download) Fig. S5. Phylogenetic relationships of Cyzicidae s.s. based on COI, 16S rRNA, EF1α and 28S rRNA inferred with RAxML. Only individuals with at least two of the four loci available were included and 3rd codon positions of COI were excluded (Matrix 4). For each individual, the country of origin and the collection or voucher number are provided (Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Bootstrap support values are provided for each node. (download) Fig. S6. Phylogenetic relationships of Cyzicidae s.s. based on COI, 16S rRNA, EF1α and 28S rRNA with MrBayes. Only individuals with at least two of the four loci available were included and 3rd codon positions of COI were excluded (Matrix 4). For each individual, the country of origin and the collection or voucher number are provided (Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Posterior probabilities are provided for each node. (download) Fig. S7. Phylogenetic relationships of Eocyzicus and Leptestehriidae based on COI, 16S rRNA, EF1α and 28S rRNA inferred with RAxML. All individuals and codon positions were included (Matrix 5). For each individual, the country of origin and the collection or voucher number are provided (Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Bootstrap support values are provided for each node. (download) Fig. S8. Phylogenetic relationships of Eocyzicus and Leptestehriidae based on COI, 16S rRNA, EF1α and 28S rRNA inferred with RAxML. Only individuals with at least two of the four loci available were included and 3rd codon positions of COI were excluded (Matrix 6). For each individual, the country of origin and the collection or voucher number are provided (see also Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Bootstrap support values are provided for each node. (download) Fig. S9. Phylogenetic relationships of Eocyzicus and Leptestehriidae based on COI, 16S rRNA, EF1α and 28S rRNA inferred with MrBayes. Only individuals with at least two of the four loci available were included and 3rd codon positions of COI were excluded (Matrix 6). For each individual, the country of origin and the collection or voucher number are provided (see also Table S1), if no collection or voucher number was available one of the GenBank numbers is provided. Posterior probabilities are provided for each node. (download) Fig. S10. Molecular clock dated phylogeny based on the amino acid data set of Schwentner et al. (2018) employing minimum age 255 mya for Limnadiidae + Eocyzicus + Leptestheriidae. The topology was fixed to the one obtained by Schwentner et al. (2018). This analysis was performed to obtain node age estimates within Spinicaudata for the subsequent molecular clock analyses using the four gene data set. (download) Fig. S11. Molecular clock dated phylogeny based on the amino acid data set of Schwentner et al. (2018) employing minimum age 380 mya for Limnadiidae + Eocyzicus + Leptestheriidae. The topology was fixed page 22 of 23Zoological Studies 59:44 (2020) © 2020 Academia Sinica, Taiwan to the one obtained by Schwentner et al. (2018). This analysis was performed to obtain node age estimates within Spinicaudata for the subsequent molecular clock analyses using the four gene data set. (download) Fig. S12. Constrained molecular clock dated phylogeny employing minimum age 255 mya for Limnadiidae + Eocyzicus + Leptestheriidae. The divergence time estimates are based on the BEAST analyses of COI, 16S rRNA, EF1α and 28S rRNA that included only individuals with at least three of the loci present. The topology was constrained to enforce a sister group relationship of Leptestheriidae and Eocyzicus (see Fig. 4 for the unconstrained analysis). Calibration points (6) and (11) are based on fossils for Diplostraca (minimum age 386.9 mya based on Leaia chinensis following Wolfe et al. (2016) and Limnadiidae + Eocyzicus + Leptestheriidae (minimum age 255 mya based on oldest known Perilimnadiidae fossils following Astrop and Hegna (2015), respectively (Figs. S13 and S14 for an alternative age constraint for Limnadiidae + Eocyzicus + Leptestheriidae). Calibrations points (A)–(D) were inferred from the preceding molecular clock analyses of the amino acid data set of Schwentner et al. (2018) (Figs. S10 and S11) and coded as normal distributed priors: (A) 294.6 mya with sigma of 25, (B) 153.5 mya with sigma of 63, (C) 66.8 mya with a sigma of 40, (D) 64.8 mya with a sigma of 32 and (E) 146 mya with a sigma of 69. Branches are color-coded according the geographic origin of the specimens. Posterior probabilities are provided for each branch. (download) Fig. S13. Unconstrained molecular clock dated phylogeny employing minimum age 380 mya for Limnadiidae + Eocyzicus + Leptestheriidae. The divergence time estimates are based on the BEAST analyses of COI, 16S rRNA, EF1α and 28S rRNA that included only individuals with at least three of the loci present. The topology was not constrained to enforce a sister group relationship of Leptestheriidae and Eocyzicus, thus also no prior was defined for their divergence (see Fig. S14 for the constrained analysis). Calibration points (6) and (11) are based on fossils for Diplostraca (minimum age 386.9 mya based on Leaia chinensis following Wolfe et al. (2016)) and Limnadiidae + Eocyzicus + Leptestheriidae (minimum age 380mya based on the assumption that Vertexioidea (known since the mid Devonian) is ancestral to Limnadiidae, while Eosestherioidea is ancestral to Leptestheriidae and Cyzicidae following Astrop and Hegna (2015), respectively (Fig. 4 and Fig. S12 alternative age constraint for Limnadiidae + Eocyzicus + Leptestheriidae). Calibrations points (A)–(D) were inferred from the preceding molecular clock analyses of the amino acid data set of Schwentner et al. (2018) (see Figs. S10 and S11) and coded as normal distributed priors: (A) 403 mya with sigma 10, (B) 155.6 mya with sigma 70, (C) 61.6 mya with sigma 30 and (D) 67 mya with sigma 35. Branches are color-coded according the geographic origin of the specimens. Posterior probabilities are provided for each branch. (download) Fig. S14. Constrained molecular clock dated phylogeny employing minimum age 380 mya for Limnadiidae + Eocyzicus + Leptestheriidae. The divergence time estimates are based on the BEAST analyses of COI, 16S rRNA, EF1α and 28S rRNA that included only individuals with at least three of the loci present. The topology was constrained to enforce a sister group relationship of Leptestheriidae and Eocyzicus, thus also no prior was defined for their divergence (Fig. S13 for the constrained analysis). Calibration points (6) and (11) are based on fossils for Diplostraca (minimum age 386.9 mya based on Leaia chinensis following Wolfe et al. (2016)) and Limnadiidae + Eocyzicus + Leptestheriidae (minimum age 380 mya based on the assumption that Vertexioidea (known since the mid Devonian) is ancestral to Limnadiidae, while Eosestherioidea is ancestral to Leptestheriidae and Cyzicidae following Astrop and Hegna (2015), respectively (Fig. 4 and Fig. S12 alternative age constraint for Limnadiidae + Eocyzicus + Leptestheriidae). Calibrations points (A)–(D) were inferred from the preceding molecular clock analyses of the amino acid data set of Schwentner et al. (2018) (Figs. S10 and S11) and coded as normal distributed priors: (A) 403 mya with sigma 10, (B) 155.6 mya with sigma 70, (C) 61.6 mya with sigma 30, (D) 67 mya with sigma 35 and (E) 154 mya with sigma 75. Branches are colorcoded according the geographic origin of the specimens. Posterior probabilities are provided for each branch. (download) Table S1. List of all individuals studied. For each individual, we provide species names, the species name under which the respective genetic resources have been deposited in GenBank (only applicable for sequences obtained from earlier studies), details on the collection locality and collection event (for sequences obtained from GenBank only the country is provided), collection or specimen numbers used to identify the specimens, the respective citations for published sequences and the GenBank accession numbers for the four genes. Collection numbers refer to the Australian Museum Sydney (AM P.), Museum of Comparative Zoology, Harvard University (MCZ IZ) and the Center of Natural History in Hamburg (ZMH K). (download) page 23 of 23Zoological Studies 59:44 (2020)