Embed Size (px)
Citation preview
Patterns of habitat affinity and AustralHolarcticparallelism in dictynoid spiders (Araneae Entelegynae)
Joseph C SpagnaAD Sarah C CrewsBC and Rosemary G GillespieB
ADepartment of Biology William Paterson University 300 Pompton Road Wayne NJ 07470 USABDivision of Organisms and Environment University of California Berkeley 137 Mulford Hall BerkeleyCA 94720-3114 USA
CPresent address Berkeley City College Department of Sciences 2050 Center Street Berkeley CA 94709 USADCorresponding author Email SpagnaJwpunjedu
Abstract The ability to survive in a terrestrial environment was amajor evolutionary hurdle for animals that once passedallowed the diversification of most arthropod and vertebrate lineages Return to a truly aquatic lifestyle has occurred onlyrarely among terrestrial lineages and is generally associatedwithmodifications of the respiratory system to conserve oxygenand allow extended periods of apnea Among chelicerates in particular spiders where the circulatory system also serves as ahydrostatic skeleton very few taxa have exploited aquatic environments though these environments are abundant and rangefrom freshwater ponds to the marine intertidal and relictual (salt) lakes The traditional systematic positions of the taxainhabiting these environments are controversial Partitioned Bayesian analysis using a doublet model for stems in the nearlycomplete 18S rRNA gene (~1800 nt) and in the D2 andD3 regions of the 28S rRNA gene (~690 nt) and standard models forloops and full protein-coding histoneH3 (349 nt) partitions (totalling 3133 bpwhen aligned) of dictynoid spiders and relatedlineages revealed that the only truly aquatic spider species Argyroneta aquatica (Clerck 1767) (Cybaeidae Banks 1892)belongs in a clade containing other taxa with unusual habitat affinities related to an aquatic existence including occupationof semi-aquatic (intertidal) areas (Desidae Pocock 1985 Paratheuma spp) and highly alkaline salt-crusts (DictynidaeO Pickard-Cambridge 1871 Saltonia incerta (Banks 1898)) In a contrasting pattern other spiders that also occupyintertidal zones including some other members of the family Desidae (Desis spp Badumna longinqua (L Koch 1867))are an independently derived clade found primarily in the southern hemisphere Use of the doublet model reduced somebranch-support values in the single-gene trees for rRNA data but resulted in a robust combined-data phylogeny from 18SrRNA 28S rRNA and histoneH3 This combination of results ndash reduction in support in single-gene trees and gain in supportin combined-data trees ndashis consistent with use of the doublet model reducing problematic signal from non-independent basepairs in individual data partitions resulting in improved resolution in the combined-data analyses
Additional keywords Argyroneta Cybaeidae Desidae Dictynoidea doublet modelling intertidal habitatsParatheumapartitioned Bayesian analysis Saltonia secondary structure
Introduction
Theability to exploit dry terrestrial environments has representeda large obstacle in the evolutionary history of complexmulticellular organisms and has occurred only once in each ofvascular plants vertebrates molluscs oligochaetes andnematodes since the Palaeozoic (Thomas 1972 Jeram et al1990 Selden 1990 MacNaughton et al 2002 Pisani et al2004 Poinar et al 2007) Among arthropods the mostsuccessful terrestrial lineages have been the insects myriapodsand arachnids (Coddington et al 2004) However the return to atruly aquatic lifestyle requires a suite of adaptations related toobtaining oxygen and to osmoregulation and has occurred onlyrarely with extensive diversification in a single group ndash thehydrachnids or water mites ndash and is associated with a parasiticlifestyle (Thorp and Kovich 1991 Coddington et al 2004) Thespider taxa that are now found in aquatic environments tend to belimited in diversity although they occur in highly disparate
environments including marine intertidal salt marshes and inonecase submerged in freshwater (Foelix 1996)Theyalsooccurin areas that were wet in the past but are now dry salt lakesCurrent taxonomy scatters these unusually distributed taxa acrossdifferent families (Platnick 2010) leading to the supposition thatthe ability to exploit such habitats has evolved independentlymultiple times
In this study we focus on taxa in the superfamilyDictynoideawithin which the greatest range of aquatic lifestyles is foundin three different families Desidae Cybaeidae and DictynidaeWithin the Desidae two genera characteristically inhabit theintertidal zone (1) the genus Desis (Walckenaer 1837) whichis distributed across the southern hemisphere and contains speciesthat hide in kelp holdfasts during daylight and high tide(McQueen and McLay 1983 Wise 1993 Forster and Forster1999) and (2) the genus Paratheuma (Bryant 1940) whichoccupies intertidal zones from the South Pacific to Mexico
CSIRO 30 August 2010 101071IS10001 1445-522610030238
CSIRO PUBLISHING
wwwpublishcsiroaujournalsis Invertebrate Systematics 2010 24 238ndash257
Florida and the Caribbean and hides in barnacle shells rock orcoral rubble during daylight and high tide (Roth 1967 Roth andBrown 1975a Beatty and Berry 1988a 1988b) Within theCybaeidae one species is well known for its freshwateraquatic lifestyle Argyroneta aquatica The sole exemplar ofits genus it lives under water one of the rarest environmentsfor spiders It is the only spider to live and hunt fully immersedusing a silken bubble attached to an aquatic plant to hold abreathable air supply (Foelix 1996 Selden 2002) It is foundin slow-moving freshwater streams throughout Europe and AsiaFinally a most unusual species Saltonia incerta (Dictynidae)(Banks 1898) inhabits an environment that is the historicalremnant of an aquatic habitat the dry alkaline salt-flats in thesouth-western United States of America and Northern Mexico(Roth and Brown 1975b)
Each of these lineages of spiders (representatives ofParatheuma and Desis and the single species S incerta andA aquatica) has been difficult to assign accurately to thetaxonomic rank of family based on morphology because eachshares some special morphological features that may or maynot be associated with its unusual lifestyle For exampleA aquatica has modifications to the respiratory systemincluding fewer lung lamellae but larger tracheal tubes thanmost other spiders (Selden 2002) which may provide someadvantage when breathing underwater Saltonia incerta hassimilar book-lungs but not the obvious association with aquatichabitats (Roth and Brown 1975b) Likewise S incerta showssome morphological similarity to desids such as havingan oversized colulus (a non-functional vestige of a pair ofspinnerets) (Roth and Brown 1975b) The uncertainty aboutwhether these similarities are homologies or homoplasies led usto use molecular evidence to explore these relationships
Inferring the evolutionary history of organisms in extremeor unusual environments based solely on morphologicalcharacters can be difficult as adaptation is often accompaniedby substantial modifications that may be convergent due tosimilar environmental pressures (Wake 1991 Givnish andSystma 1997 Lee 1998) The advent of molecularphylogenetics has greatly improved understanding of the roleof convergent adaptation to a wide variety of environments(Martins 2000) Here we build on previous work on theaquatic semi-aquatic and salt-flat living spider lineages thathas shown that these taxa belong to a small number of relatedfamilies nested within an established spider group the lsquoRTArsquoclade (Coddington and Levi 1991) Themonophyly of the lsquoRTArsquoclade so-called because of the characteristic presence of aretrolateral tibial apophysis (RTA) on the male intromittentorgan is supported by recent morphological and molecularanalyses (Griswold et al 1999 2005 Spagna and Gillespie2008) which provide a framework for rational sampling
To address the questions of evolutionary history we generateDNA sequence data and make use of improved methods ofanalysis of rRNA genes which form the basis of many of themolecular phylogenetic analyses of both deep and recentdivergences in arthropods including crustaceans insects andspiders (Dixon and Hillis 1993 Hedin and Maddison 2001Wheeler et al 2001 Mallatt et al 2004 Spagna and Gillespie2006 Crews et al 2010) It has long been known that thesecondary structure of these genes influences their evolution
and that constraints differ between the paired bases in self-complementary helices (stems) and unpaired structures (loops)(Schoumlniger and von Haeseler 1994) Recently it has becomepossible to account for the non-independence of stem bases usingadoubletmodelwhich has nowbeen applied in several studies onarthropods (Kim et al 2003 Kjer 2004 Angelini and Jockusch2008 Crews et al 2010) Such partitioned modelling hasbeen shown to improve upon traditional modelling regimes(Erpenbeck et al 2007) and is facilitated here by spider-specific secondary structure models which exist for thecommonly used nuclear ribosomal-RNA genes for the smallsubunit (18S) and large subunit (28S) rRNAs (Hendriks et al1988 Hedin and Maddison 2001 Spagna and Gillespie 2006)and can be evaluated and improved upon for use in phylogeneticanalyses
Here we sample the RTA clade broadly focussing ondictynoid lineages including the fully-aquatic semi-aquaticand salt-flat endemic species to improve understanding of theevolutionary history of taxa that occur in habitats not typicallyinhabited by spiders To this end we develop a secondarystructure model for 28S rRNA for RTA-clade spiders usingthermodynamic algorithms to modify and extend existingmodels Secondary structure models (the new RTA-clade-specific 28S rRNA models and the pre-existing spider 18SrRNA model) are used to identify the self-complementarystem positions for doublet modelling in a partitioned Bayesiananalysis framework (Huelsenbeck and Ronquist 2001) Withthis we estimate a phylogenetic tree that can serve as atemplate for further comparative ecological physiological andgene-expression studies
Methods
Taxon sampling
We chose taxa from throughout the RTA clade (Coddington andLevi 1991) sampling multiple exemplars of the hypothesisedrelatives of the focal taxa from within the Dictynoidea inparticular Cybaeidae for Argyroneta Dictynidae for Saltoniaand both Austral and Holarctic genera of Desidae and theirputative relatives (Spagna and Gillespie 2008) We alsosampled from families that have been closely allied to theDictynoidea in particular Agelenidae C L Koch 1837(including representatives of the agelenine tegenariine andtextricine subfamilies) and lsquoCoelotinesrsquo (AmaurobiidaeThorell 1870) that are associated with riparian habitats(Glesne 1998 Ayoub and Riechert 2004)
Though representative taxon sampling for a highly diversegroup such as the RTA clade (gt18 000 species Platnick 2010) isdifficult and for practical reasons imperfect our choice of taxais based on both traditional sources including revisionary andtaxonomic literature (Lehtinen 1967 Coddington and Levi 1991Platnick 2010) and more recent molecular work inclusive ofmany of the same taxa (Spagna and Gillespie 2008 Miller et al2010) We follow the latter two papers in our general regimeof including exemplars from as many of Lehtinenrsquos (1967)lsquosubfamiliesrsquo as possible Though Lycosoidea and Dionychaare relatively undersampled here the purpose of theirinclusion is to allow broad sampling across the RTA-cladeingroup rather than use them to test hypotheses about specific
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 239
relationships within the Dictynidae or other 3-clawed taxa forwhich deep sampling of lycosoids and dionychawould have littleutilityTaxondeposition andGenBank accession information canbe found in Appendix 1
Three non-RTA clade taxa were included as potentialoutgroups with exemplars of Nicodamidae Simon 1897(Megadictyna thilenii Dahl 1906) Eresidae C L Koch 1950(Stegodyphus sp) andAraneidaeClerck 1757 (Zygiella x-notataClerck 1757) used to represent the groups thought to be thenearest relatives including the sister-group to the RTA clade(Griswold et al 1999 2005 Spagna and Gillespie 2008)
Molecular markers and amplification conditions
Markers chosen for this study were nuclear sequences from theD-loop sectionof the 28S rRNAgene two large sections coveringgt90 of the 18S rRNA gene plus the full sequence for theprotein-codinghistoneH3geneThesemarkerswhich are amongthebest characterised in spiderswere specifically chosen for slowaccumulation of mutations making them useful for decipheringphylogenetic relationships in arachnids at relatively deep levels(Giribet et al 1996 Wheeler and Hayashi 1998 Maddison andHedin 2003 Murphy et al 2006) Faster-evolving markerssuch as mitochondrial cytochrome c oxidase subunit I (COI)and 16S rRNA were not used as they have been shown tocontribute little unsaturated signal to analysis of divergences atthis depth (Spagna and Gillespie 2008)
Theprimer pairs 28SOand28SC(Hedin andMaddison2001)were used to amplify a section of 28S rRNA sequence (~690 bp)in length Because some RTA-clade spiders appear to have28S rRNA paralogues (Murphy et al 2006 J C Spagna andR G Gillespie unpubl data) care was taken to make sure thatno double or multiple banding was seen in PCR amplificationproducts resulting in clean interpretable chromatogramswithoutdouble peaks for all sequences used in this project Two primerpairs 18S 1F and 18S 5R and 18S 4F and 18S 9R (Giribet et al1996) were used to amplify the first ~850 and second ~950 basepairs of the 18S rRNAgene respectively A 349 base pair sectionof histone H3 was amplified using the primer pairs histone H3aFand histone H3aR (Colgan et al 1998) For 28S rRNA FailsafePCR Optimization Kit (Epicentre Technologies Madison WI)buffers were used to improve PCR reaction yields
The PCR mixtures used were 12mL total final volume with abasic mix of 25mL dNTPs 10mL buffer 02mLTaq 25mL perprimer 15mL of template DNA and 18mL H2O The bufferused for each rRNA locus was determined using a FailSafeoptimization kit while standard buffer was used for thehistone H3 amplifications For PCR cycles a basic protocol of95C melting temperature for 30 s 55C annealing temperaturefor 40 s and72Cextension temperature for 45 swas repeated for35 cycles Minor modifications were made for individual taxa interms of temperatures and cycle segment lengths (PCR buffersused and cycle optimisations are available upon request from theauthors)
Alignment secondary structure assessmentand incorporation in partitioned models
Alignments of the 18S and 28S rRNA datasets were done usingMUSCLE MUSCLE was chosen for alignment for its relative
accuracy speed and ease of use (Edgar 2004) Alignment of28S rRNAwas also conducted using secondary structure modelsto produce a second set of aligned datasets Alignment usingsecondary structure has been demonstrated to be moreappropriate than traditional alignment methods (Kjer 1995Angelini and Jockusch 2008) Also finding the preciselocation of stems and loops allows for them to be modelledseparately in phylogenetic analysis and also allows for theimplementation of a doublet model (Kjer 2004) To accountfor secondary structure in 18S rRNA we used the pre-existingmodel of Hendriks et al (1988) based on sequences from atheraphosid spider which has been shown to be generallyapplicable across a broad range of arachnids (Spagna andGillespie 2006) There was no conflict between the MUSCLE-aligned 18S rRNA data and the secondary structure model usedthat required post hoc realignment so bases could simply beassigned to lsquostemsrsquo or lsquoloopsrsquo based on the Hendriks et al (1988)model
Modelling of the secondary structure of 28S rRNA was morecomplex because the interspecific variability in this marker madefolding and self-complementary pairing more ambiguousExisting models for arthropods (Hedin and Maddison 2001)and the mfold web server (Zuker 2003) were used to findappropriate structures for the amplified sections (~690 bp) ofthe D2 and D3 segments of 28S rRNA Mfold uses a free energyalgorithm following the thermodynamic rules of SantaLucia(1998) and the default parameters were used
Assessment of behaviour of themfold algorithmdemonstratedthat lower folding temperatures resulted in consistently greaterlevels of pairingwithin each arm than at the default temperature of37C (mean increase in the number of bases forming pairs was48at 21C49at 4C)Though this default temperature settingfor the mfold algorithm might appear high for an ectothermicanimal the spiders sampled live at a wide range of temperaturesoften exceeding 37C in the case of Saltonia so it falls withinrealistic physiological limits of the animals Because we lackcorroborating crystallographic data for our sequencesrsquo secondarystructures we used 37C as a conservative approach to base-pairmodelling as any inappropriate assignment of bases to stempartitions would result in underweighting these effectivelyreducing the size of the matrix
Each helix of the D2 region for each taxon was input singlyinto the server thus producing a possible structure or structuresfor eachhelix of each specimenThe entire lsquocore +D3rsquo regionwasinput for each specimen producing a possible structure Gapswere inserted and adjusted by eye to produce the final alignmentin such a way to minimise discontinuities in the stem structuresshared across taxa assuming that unpaired loop regions are morelikely to have undergone the indel events represented by the gapsIn the final alignment it was ambiguous whether certain baseswere stems or loops and in these cases the ambiguities wereassigned to loops
The protein coding gene histone H3 (349 base pairs) wasaligned by eye after the sequences were translated to codonswhich also allowed us to determine that therewere no stop codonsin any of the sequences
With the data matrix including lineages with known longbranches inconsistent molecular clocks (Spagna and Gillespie2008) anddeep divergences (gt100million years) (Penney 2003)
240 Invertebrate Systematics J C Spagna et al
we used a Bayesian likelihood approach to analyse the data inorder to avoid possible long-branch attraction artefacts(Felsenstein 1978 Bergsten 2005) Additionally secondarystructure information for rRNA genes using both pre-existingmodels and new models developed herein was included in theBayesian analyses in order to compensate for the non-independence of base-pairing in the self-complementaryhelices of the lsquostem regionsrsquo of rRNA
Although generally considered superior to parsimonymethods for dealing with long branches Bayesian methodshave also been shown to be potentially susceptible to suchartefacts (Kolaczkowski and Thornton 2009) For this reasonwe performed accelerated maximum likelihood analysis usingRAxML-HPC 704 (Stamatakis 2006a) on the CIPRES Portal v21 (Miller et al 2009) for comparison with Bayesian outcomesThe matrix was partitioned into stems and loops for the rRNAgenes and by codon position for histone H3 using the GTRCATmodel (Stamatakis 2006b) for each partition with the number ofrate categories set to 25 Bootstrapping was conducted usingthe rapid bootstrap algorithm for webservers (Stamatakis et al2008) and these values were calculated for nodes on the best-scoring tree
Bayesian analysis was conducted on each gene separatelyand in one concatenated dataset for both the MUSCLE-aligneddatasets and the structurally aligned datasets using mixed-modelpartitioned analyses (Nylander 2004 Brandley et al 2005) inMrBayes 31 (Huelsenbeck and Ronquist 2001 Ronquist andHuelsenbeck 2003) The rRNA genes were partitioned by stemshalf-stems and loops using a doublet model of nucleotidesubstitution (Kjer 2004 Angelini and Jockusch 2008) for fullstems to account for the non-independence of bases for which wehad explicit information about complementary base positionswhile histone H3 was partitioned by codon position The doubletmodel as implemented in MrBayes 31 is highly parameterised(16 parameters total) but previous studies have shown it to besuperior to models that assume independence (Tsagkogeorgaet al 2009) The primary risk from over-parameterisation isan increase in variance of parameters (eg branch lengths)rather than decreased accuracy in consensus tree topology(Huelsenbeck and Rannala 2004 Kelchner and Thomas 2007)For all non-stem partitions models were chosen using the AIC inMrModeltest 22 (Nylander 2004 Posada and Buckley 2004)Chosen models are shown in Table 1 Gaps were treated aslsquomissing datarsquo in all analyses The 28S rRNA and H3 datasetswere run for 10million generations the 18S rRNA dataset wasrun for 9million generations and the concatenated dataset wasrun for 12million generations The cumulative and comparecommands were used in AWTY (Wilgenbusch et al 2004) todetermine burn-in and compare the two results of each MrBayesrun
Results
Modelling of secondary structure
The 18S rRNA model for spiders of Hendriks et al (1988) wasfound to be sufficient for assignment of strings of bases to looppartitions and pairs of bases to complementary stem partitionsfor model and parameter partitioning before analysis The 28SrRNAdata were considerablymore variable in terms of both base
and length polymorphisms and the secondary structure modelsdeveloped for the region sequenced are seen in Figs 1ndash5
Arm 1 is the most variable of the 28S rRNA structuresmodelled and a representation of the variation can be seen inFig 2 Variation occurs both between and within clades (seephylogenetic results) in terms of total transcript length andnumber and size of loops and stems Within-taxon variationoccurs even in species of the same genus (such as Paratheumainsulana (Banks 1902) and P armata (Marples 1964) Fig 2Group1) Of groups represented in Fig 1 the Cybaeidae sl anddictynoids+Desis appear themost structurally uniform with theAgelenidae (including the subfamily Coelotinae) appearing quitevariable
Structures for Arm 2 aremuchmore uniform thanArm 1 withmost variation occurring in the form of presence or absence of asingle loop (represented by an asterisk in Fig 3) in the terminalhelix of the arm (Fig 3 panels 1ndash4)
While longer than both Arm 1 and Arm 2 across the spidertaxa modelled the Arm 3s (Fig 4) are similar in uniformity toArm 2 with a set of 7ndash8 positionally homologisable loopstructures Though not all Arm 3s have identical numbers ofloops the task of homologising individual loops was madesimpler by the amount of asymmetry possibly resulting fromslip-strand mispairing events (Gillespie et al 2004) between the50 and 30 sides of the intervening loops as these were oftenremarkably unequal and this asymmetry was fairly consistentacross taxa
The primary sequences for the D3 + core structures were inmost cases highly similar and thus alignment was largely trivialsimplifying the modelling of stems and loops across multipletaxa with some exceptions The structure for Zygiella x-notata(Fig 5) represents the structure for conforming taxa while thethree taxa with notable variations from the primary sequencelength (Paratheuma Saltonia and Dictyna) were modelledindividually Structurally Paratheuma and Dictyna appear toshare a positionally homologous self-complementary four-base-pair motif which makes the centre of the D-loop core appearlsquopinchedrsquo into two smaller loops (top 2 panels of (F 5) which isnot seen in the other study taxa
Combined data phylogeny
The majority-rule consensus tree for 6000 post-burn-incombined-data trees (representing 6million generations after4million generations were discarded as burn-in) is shown inFig 6 The monophyly of the ingroup representing all sampledmembers of the RTA clade was strongly supported (posteriorprobability (pp) = 10) The focal taxa of this study (those with
Table 1 Partitions and models used in partitioned Bayesian analyses
Partition Selected model
18S rRNA stems HKY+I+G (doublet)18S rRNA loops GTR+I+GHistone H3 position 1 GTR+I+GHistone H3 position 2 HKYHistone H3 position 3 GTR+I+G28S rRNA stems GTR+G(doublet)28S rRNA loops GTR+G28S rRNA half-stems GTR+G
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 241
GC
G
GC
UU
A
U
C
AG
G
G
G
G
G
G
U
GC
C
A
GC
A
C
A
A
A
AU
U
G
C
UU
A
CU
C
C
CU
C
CC
C A
C
U
C
U
G
G
G
C
C
U
G
U
C
C
G
GGG
U
U
U
U
U
U
G
U
U
G
G
U
G
G
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
5
G C C
G G A
U
GG
C
C
G
U
G
C
G
C
C
C
G
C
A
C
U
U
G
U
G
C
GC
CG
G
G
G
G
G
C
G
A
C
G
C
A
GA
G
CC
G
C
G
G
C
C
C
G
G
C
G
C
C
U
U
C
A
A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
C
C
G
U
C
G
C
A
UU
A
A
C
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
UCU
CCC
GU
CCC
G A GG
ACG C U G C A
CA
CGGUC
UG
GU A U G G U AAA C G
AC
UC
UC
GG
A U GG A G A
AGUCGACA
GA C C
GUGGUGAC
G C C GG
CCC
GA A
CG
GGG A G U C G
C
UGCAGC
U
U
G GC
G
C
G
C
G
A
C
C
GUGG
U
G
A
CG
CC
GG
CC
C G A A C GG
G Y A G U CG
C
3
ARM 1
ARM 2
ARM 3
ldquoCORErdquoldquoD3rdquo
Fig 1 Hypothesis of secondary structure for the D2 and D3 regions and the core region connecting them for Saltonia incerta showing relative positions ofthe Core D3 and Arms 1ndash3 in the 50 (lower left) to 30 (upper right) Arm and core orientations are maintained in Figs 2ndash5 for comparison
242 Invertebrate Systematics J C Spagna et al
G U C
C A GU
A
A
G
G G GG
G
GGG
GG
G
G
GG
G
A
A
A
A
A
A
A
AA
AU
U
U
UU U
C
C
C
CC
C
C
C CCC CCC
C
CC
U
G G
G
G
G G
C
C C CCC C
G CG G G
C
CC
G
G
GU U UU
GG
G
U
U
U
A
A A
U
A
AA
G
G G
A
A
Arg
yron
eta
aqua
tica
G U C
C A GU
A
A
GG
GGGGG
GG
GG
GGG
G
GGG
GG
A A
A
A
AA
A
AAAA
AAA
AA
AU
U
U
U
U
U
U UU
UU
UU
U
U U
C
CC
C
C C C C
C
C C C CC
C
C CC
CC
C
C
C
C
U
G
G
G
G
G
G GG
G
G
G
C
C C
CC
C
CC C
Dic
tyna
sp
G U C
C A GU
A
A
G
G G
G
G
GGG
G
G
G
G
G
GGG
A A
A
A
AA
A
A
A
AA
A
AU U
U
U
UU
U U
CCC
C
C
C
C
C
C C
CC
C C
CCC
C
C
U
G G
G
G G G
C
C C CC
C C
G
C G G G
C
C
C
G
G
GU
U
U
U
GGG
U
UA A
UU
Par
athe
uma
arm
ata
G U C
C A GU
A
A
G
G
G
GGG
G
GG
G
GG
GG
G
GCUG
A A
A
A
AAA
A
A
A
A
AA
AU U
U
U
U
U
U U
C
C
C
C
C C
C
CC
C
C
C C
CCC
CC
C
C
U
G G
G
G
G GG
GG
G
C
C C
CC
C
C
C
G
C GG
G C CC
G
Para
theu
ma
insu
lana
G C G
G C
UU
A
U
C AG
G
G
G
G
G
GU
GCC
A GC
A
C
A
AAAU UG
C
UU
ACU
C
C
C U CCCC
A
C
U
C
U
G
G
G
C
C U GU
C
C
G G G G U
U
U
U
U
U
G
U
U
G
GU
G
G
Sal
toni
a in
certa
Aqu
atic
Cla
de
G U C
C A G
A
G G
U
UG C
U
A
GG
AU
G
UU
A
G
CCCC
U
A
C
U
G
G
G G
C
A
CG C
G
C
G
C
G
G
C
A
U
A
A
G
G
CC
A
U
U
G
G
C G GC CG C G
GU C
U
G
C
CG
G
C
U
A
A
C
G
U
A
GG
G
U
C
CC
UC
A
C
AGUUA
Am
aruo
bius
sim
ilis
Am
auro
biid
ae a
nd O
utgr
oups
G U U
C
A
C
GU
G
GC
U
GCCC
GU
U
A
G
G
G
A
G
A
A
A
U
A
A
G
A
C
AU
CC
U U
C
GU
C
G
A
UC
CC
GC
C
U
G
C
C
G
C
U
U
G
C C
CC
GG
C
CG
G
G
U
G
AA
GG
U
G
G
AG A
AC
C
G
GC
AU
AC
UU
U
Ste
gody
phus
sp
G U C
C A GG
C
A
G
G
A
U
GC
G
UC CAA
A
A
G
A
A
U
AU
U
U
C
U U
AU
U
A
CCCC C
G
G
G
C
U
G
G
C
G G
C
U
U
C
GU
A
G G G
U
U
C
A
A
U
G
G
G
G
GA
U
U
U
C
G
G
U
A
U
C
C C
C
C
G
C
G
G
AC
UG
UG
C
Meg
adic
tyna
thile
nii
G U C
C A GG
C
A
AA
G
G
GG
GG
GGC
G
GGGG
G
A
A
G
AG
A
AUAA
A
AC
AA
AU
U
UU
G
U
U
UUU
UA
C
G
U U
CC
CG
C U C C
G
U CC
CC
U
GC
U
C
C
U
G C
G
G
G CU
C
G
A
C G
C
C
C
G C
GCG U
Zygi
ella
x-n
otat
a5
3
G UC G C
U
A
G G
U
G CG
A
GGC
A
AU
U U
A
C
CC
G
G
C
U
G
G
C
G
GC
U
G
U
G
A
G
G
U
G
C
G
C
G
C
G
C
A
C
C
C
G
G
G
C
A
GGC
C
G
GA
G
A
C
C
U
CG
CA
A
U UGAA
A
CG
G
A
AAC
C
CC
CC
GG
G G
G
CC
Cry
phoe
ca s
p
G UC A G
C
A
G C
U
G
GC CG
A
CG
C
A
AU
G
U U
A
G
C
GG
G
G
C
U
G
A
C
C CG
G
C
G
U
A
A
G
GA
U
C
U
G
C
G
C
G
U
C
G
C
A
C
C
G
G
G
C
C
G
G
C G
C
UGC
G CA
G
G
AG
C
C
C
C
UC
GC
C
GA
U GCAG
A
CG
Bla
bom
ma
sp
G UC A G
C
A
U C
C
G
UG GG
A
CG
G
A
AU
G
U U
A
G
A
GA
C
G
C
U
G A
C
C CG
A
G
G
C
U
U
G
UA
U
G
G
U
A
U
A
C
G
U
C
G
C
A
C
C
G
C
GC C
G
G
C G
C
CGU
G CA
G
G
AG
C
G U
C
C
C
UC
GC
CGG
G GCCG
G
Cal
ymm
aria
sp
G UC G C
C
A
G G
A
G C
A
G
GU
U U
A
C
CG
G
C
C
U
GC G
C
G
C
G
C
G
G
G
U
G
C
U
G
G G
C
A
C
C
C
G
G
G
C
A
GCC
C
G
GA
G
A
C U
CG
C
U
C
U
AG
A
CG
G
A
A
AC CC
CG
G
G G
G
AA
AA
Rot
haei
na s
p
G UC A G
C
A
G C
A
G
G CG
A
CG
C
AAU
C
U U
A
G
C
G
G
G
C
C
U
G
A
C
C CG
C
C
G
C
G
A
G
GG
U
U
G
C
A
G
G
U
C
G
C
A
C
C
G
G
G
C C
G
G
C G
C
GCG
U CA
G
G
GG
C
C
C
C
UCG
CC
GA
U GCAG
A
CG
G
Cyb
aeus
sp
2
Cyb
aeid
ae s
l C
lade
G U C
C A GU
C
A
G GG
A
U
GC
G
G
GG GGG
A
A
G
G A
UAU
U U
C
U U
U
C
C
C
C
C
C
CC
G
CC
C
C
C
U
G
G
G
G
G
C
C
C
CG
C
G
G
G G
G
C
C
C
A
A
A
U
A
U
U
G
G
C
U
U
U
A
C
A
G
A
A
A
G
G
C
A
G
G
U
C
GG
GC
U
C
53
Text
rix d
entic
ulat
a
Age
leni
nes
+ C
oelo
tines
G U C
C A GUA
G
G
GG
G
G
GGG
A
A
G
C
A
AU
U
G
U U
A
C
C
C CCCC
C
U
C
G
C
U
G G
G
G G
C
C
C
CG
C
C
G
CC
C
G
G
U
G
U
G
G
C
C
U
G
C
A
G
G
A
A
G
C
A
A
G
G
GG
C
U
GC C
G
G G
CG
UCA
CUCG
Tege
naria
dom
estic
a
G U C
C A
A
C G
U
G
UG CC
A
G
A
AU
G
U U
G
CA
CCC
A
CU
G
G
C
U
G
G
G G
C
C
G
CG G
C
A
C
C
G
T
G
GU
GC
U
G
A
C
A
U
C
CGU
C
U
C
U
GC G G
C
G
UC
CGCA
G
C
GC
C
GU
Wad
otes
dix
iens
is
G U C
C A GUA
G
C
AG
GG GGG
A
UG
C
A
AU
U
G
U U
A
G
C CCCC
CU
C
U
C
U
G
G G
GG
C
C
C
CG
C
C
G
UC
C
G
A
U
A
U
GC
C
C
U
G
G
G
G
G
A
U
G
C
A
A
G
C
GGC
UC
A CG
G G
CG
UC
A
GCCA U
Coe
lote
s te
rres
tris
G U C
C A GU
A
A
G GG
GG
GG
G
G
GG GGG
A
A
A
A
A
A
A
AU
U
U
U
U U
C
CC
CC
C
C
CC
C
CC
C
C
C
U
G
G
G
G
G
C
C
C
CC
C
G
C
G G
G
C
C
C
G
G
G
U
U
U
UG
G
G
UU
U
A
C
A
U
A
A
A
G
G
A
A
UU
U
C
Alla
gele
na g
raci
lens
G U C
C A GU
C
A
G
G
AG
GG GGG
A
UG
C
A
AU
U
U
G
U U
A
C
C
C CCCG
CU
C
U
C
U
G
G
G
G G
C
C
C
GG
G
C
G
UC
C
G
A
U
A
U
G
G
C
U
U
G
C
A
G
G
A
C
G
C A
A
G
C
GG
C
UUC C
C
G G
CG
UC
Age
leno
psis
ape
rta
Lyco
soid
s +
Dio
nych
ans
G U C
C A GUA
C
CU
A
A
CC
G
G
G
A
U
C
A
A
AU
C
UG
U U
G
G
G
G
CCC C
A CG
C
U
G U
U G
C
C
C
U
GG
A
G
C G
G
C
A
A
GG
G
G
C
A
G
U
U
C
C
C
UA
A
G C
G
G
CG
G
U
G
C
U
G
G
G CG G
G A
CC
A AA
UU
U
Lath
ys a
lber
ta
Dic
tyno
ids
+ D
esis
G U C
C A G
A
G G
UUG CU
A
GG
AU
G
U U
A
G
CCCC
G
G
C
U
G
G
G C
C
A
C
CG U
G
C
G
C
G
A
C
A
U
A
A
G
G
A
C
C
A
A
T
G
G
C G GC
G
CG C
G
GC
C
C
G
A
C
UG G
C
C U
A
UC
G
A
A
GG
G
CU
U
GG
Met
alte
lla s
imon
i
G U C
G
A
C
G
G
G
GGC
A
GGGC
C
U
C
G
G
G
G
A
C
A
A
A
A
A
A
G
C
A
AAU G
UG
U U
C
CC
C
C
C C
G
C
CC
GCG
C
U
G
G
G
G
C
C U CCC C
CG CG
U
C
G
G
G
A
A
AG
U
U
A
U
G
C
AA
A
GC
GC
G
GC C
GUA
AA
C
Neo
ram
ia ja
nus
G U C
C A G
A
G G
AUG
C
G
A
G
C
AU
U U
G
GGG
A
A
GG
G
C
U
G
C
C C
C
C CG U
G
A
C
A
G
A
C
UC
U
G
A
U
C
U
C
A
U
U
C
A
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G
G
C
C
A
C
G
A
A
GG
G
U
G GC
C
C
C
UC
Des
is m
arin
a
G U C
C A G
A
G G
UUG CU
A
G
G
AU
A
U U
A
G
GGG
A
A
UG
G
C
U
G
C
C C
C
A CG C
G
C
G
C
G
A
C
UC
U
A
A
G
G
A
U
A
U
C
C
G
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G G
C
C
A
U
C
G
U
A
G GG
U
C G
G
C
Bad
umna
long
iqua
G U C
C A G
CC
A
G
G U
GG
G
G
G
G
C
GG
GC
G
G
GGG
A
A
A
A
A
A
A
A
A AAAAU
U
U
A
UUU U
CC U
C
C
C
U
C
UC
A G
C C
C C
C
U
C
U
G
G
G
G G
C
G
C
C
C
U
C
G
G
C UC G
C
U
U
U U UU
U
UU
A
U
Teng
ella
radi
ata
G U C
C A GU
A
A
A
CU
AA
UG G
G
G
A
U
C
A
A
AU UC
U
U U
G
A
A
C
GGG G
C
C
C
C
U
G G
UG C
C
C
U
GG U
G
U G
GC
A
A
G
GG
C
C
G
U
U
U
C
G
C
U A
C
G C
G
C
C
G
G
UG
CU
C
G
GC
C C GAGA
Alo
peco
sa k
ochi
G U C
C A GU
A
A
G
G
U
GCG
GG GGG
A
G
G
G
G
C
AU
C
AG
U U
A
C
C
C GGGC
GA
U
U
C
U
G
G
GC
C
C
UUG
U
G C
GU
C
C
G
U
U
G
U
C
G
C
U
A
C
A
C
A
G
G
C G
U
G
U
CC
G
U
C
CU
U
GC
UC
UGC
Any
phae
na c
alifo
rnic
a
G U C
G
UA
GG
GG
G
G
GG
G
GG
G
G
G
G
G
A
A
A
A
A
A
A
A
A
AA
A
AU
U
UU
U U
C
C
C
C
C
C C
C
C
CC
C
C
C
C
U
G
G
GG
G G
C
C C CCC C
G
CG GG
C
CC G
G
G
U
U
G
GGU
U
A
U
U
C
AA A
A A
G
G C
G
Stip
hidi
on fa
cetu
m
G U C
C A
A
C
G
U
G
C
U
A
C
G
AU
U
UU
A
G
UC
U
U
C
U
G
U
C
CG G
G
C
G
C
G
C
A
A
U
A
U
G
G
A
G
A
G
C
C
G G G
GCCC
C
GG
G
A
A
C
C
A
A
U
U
GC
C
C G
G
AU
C
C
AGU
C
G
G
G
U C CCU
G
Zelo
tes
sp
ARM
1
95 (8
0-10
7)10
15
(101
-102
)
Am
auro
biid
ae
103
5 (9
9-11
8)
Out
grou
ps
100
2 (9
0-10
6)10
04
(93-
104)
997
(97-
102)
101
1 (1
00-1
02)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig2
Groups1
and2thesecondarystructuresofthelsquoA
rm1rsquooftheD2region
of28
SrRNAfrom
mem
bersofthelsquoaqu
aticcladersquothe
amaurobiidcladeandthreeou
tgroup
taxa
from
thecombined-
datatreeA
rm1ishigh
lyvariableevenwith
ingeneraasdemon
stratedby
thetwospeciesof
Paratheum
aGroups3nad4Arm
1sfrom
mem
bersof
theAgelenidaeandCoelotin
ae(G
roup
3)and
thecybaeidsensulatoclade(G
roup
4)Ingeneralthelatterappearmorestructurally
conservedthan
theform
erG
roups5and6
Arm
1sfrom
mem
bersof
theLathys(D
esidae)andAustralclade
spiders(row5)andforL
ycosoids
andDiony
chans(Group
6)O
fthe
form
erthe
genu
sDesishasa
particularlyconservedsecondarystructureforthisfeatureThe
armsshowninthesecondarystructure
figureswerechosen
torepresentthe
rangeof
variationBoxes
totherighto
feach
grouprepresentm
eanlength
andminim
um-m
axim
umrange(inparentheses)to
capturefullvariationin
figu
red
andunfigu
redarms
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 243
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Florida and the Caribbean and hides in barnacle shells rock orcoral rubble during daylight and high tide (Roth 1967 Roth andBrown 1975a Beatty and Berry 1988a 1988b) Within theCybaeidae one species is well known for its freshwateraquatic lifestyle Argyroneta aquatica The sole exemplar ofits genus it lives under water one of the rarest environmentsfor spiders It is the only spider to live and hunt fully immersedusing a silken bubble attached to an aquatic plant to hold abreathable air supply (Foelix 1996 Selden 2002) It is foundin slow-moving freshwater streams throughout Europe and AsiaFinally a most unusual species Saltonia incerta (Dictynidae)(Banks 1898) inhabits an environment that is the historicalremnant of an aquatic habitat the dry alkaline salt-flats in thesouth-western United States of America and Northern Mexico(Roth and Brown 1975b)
Each of these lineages of spiders (representatives ofParatheuma and Desis and the single species S incerta andA aquatica) has been difficult to assign accurately to thetaxonomic rank of family based on morphology because eachshares some special morphological features that may or maynot be associated with its unusual lifestyle For exampleA aquatica has modifications to the respiratory systemincluding fewer lung lamellae but larger tracheal tubes thanmost other spiders (Selden 2002) which may provide someadvantage when breathing underwater Saltonia incerta hassimilar book-lungs but not the obvious association with aquatichabitats (Roth and Brown 1975b) Likewise S incerta showssome morphological similarity to desids such as havingan oversized colulus (a non-functional vestige of a pair ofspinnerets) (Roth and Brown 1975b) The uncertainty aboutwhether these similarities are homologies or homoplasies led usto use molecular evidence to explore these relationships
Inferring the evolutionary history of organisms in extremeor unusual environments based solely on morphologicalcharacters can be difficult as adaptation is often accompaniedby substantial modifications that may be convergent due tosimilar environmental pressures (Wake 1991 Givnish andSystma 1997 Lee 1998) The advent of molecularphylogenetics has greatly improved understanding of the roleof convergent adaptation to a wide variety of environments(Martins 2000) Here we build on previous work on theaquatic semi-aquatic and salt-flat living spider lineages thathas shown that these taxa belong to a small number of relatedfamilies nested within an established spider group the lsquoRTArsquoclade (Coddington and Levi 1991) Themonophyly of the lsquoRTArsquoclade so-called because of the characteristic presence of aretrolateral tibial apophysis (RTA) on the male intromittentorgan is supported by recent morphological and molecularanalyses (Griswold et al 1999 2005 Spagna and Gillespie2008) which provide a framework for rational sampling
To address the questions of evolutionary history we generateDNA sequence data and make use of improved methods ofanalysis of rRNA genes which form the basis of many of themolecular phylogenetic analyses of both deep and recentdivergences in arthropods including crustaceans insects andspiders (Dixon and Hillis 1993 Hedin and Maddison 2001Wheeler et al 2001 Mallatt et al 2004 Spagna and Gillespie2006 Crews et al 2010) It has long been known that thesecondary structure of these genes influences their evolution
and that constraints differ between the paired bases in self-complementary helices (stems) and unpaired structures (loops)(Schoumlniger and von Haeseler 1994) Recently it has becomepossible to account for the non-independence of stem bases usingadoubletmodelwhich has nowbeen applied in several studies onarthropods (Kim et al 2003 Kjer 2004 Angelini and Jockusch2008 Crews et al 2010) Such partitioned modelling hasbeen shown to improve upon traditional modelling regimes(Erpenbeck et al 2007) and is facilitated here by spider-specific secondary structure models which exist for thecommonly used nuclear ribosomal-RNA genes for the smallsubunit (18S) and large subunit (28S) rRNAs (Hendriks et al1988 Hedin and Maddison 2001 Spagna and Gillespie 2006)and can be evaluated and improved upon for use in phylogeneticanalyses
Here we sample the RTA clade broadly focussing ondictynoid lineages including the fully-aquatic semi-aquaticand salt-flat endemic species to improve understanding of theevolutionary history of taxa that occur in habitats not typicallyinhabited by spiders To this end we develop a secondarystructure model for 28S rRNA for RTA-clade spiders usingthermodynamic algorithms to modify and extend existingmodels Secondary structure models (the new RTA-clade-specific 28S rRNA models and the pre-existing spider 18SrRNA model) are used to identify the self-complementarystem positions for doublet modelling in a partitioned Bayesiananalysis framework (Huelsenbeck and Ronquist 2001) Withthis we estimate a phylogenetic tree that can serve as atemplate for further comparative ecological physiological andgene-expression studies
Methods
Taxon sampling
We chose taxa from throughout the RTA clade (Coddington andLevi 1991) sampling multiple exemplars of the hypothesisedrelatives of the focal taxa from within the Dictynoidea inparticular Cybaeidae for Argyroneta Dictynidae for Saltoniaand both Austral and Holarctic genera of Desidae and theirputative relatives (Spagna and Gillespie 2008) We alsosampled from families that have been closely allied to theDictynoidea in particular Agelenidae C L Koch 1837(including representatives of the agelenine tegenariine andtextricine subfamilies) and lsquoCoelotinesrsquo (AmaurobiidaeThorell 1870) that are associated with riparian habitats(Glesne 1998 Ayoub and Riechert 2004)
Though representative taxon sampling for a highly diversegroup such as the RTA clade (gt18 000 species Platnick 2010) isdifficult and for practical reasons imperfect our choice of taxais based on both traditional sources including revisionary andtaxonomic literature (Lehtinen 1967 Coddington and Levi 1991Platnick 2010) and more recent molecular work inclusive ofmany of the same taxa (Spagna and Gillespie 2008 Miller et al2010) We follow the latter two papers in our general regimeof including exemplars from as many of Lehtinenrsquos (1967)lsquosubfamiliesrsquo as possible Though Lycosoidea and Dionychaare relatively undersampled here the purpose of theirinclusion is to allow broad sampling across the RTA-cladeingroup rather than use them to test hypotheses about specific
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 239
relationships within the Dictynidae or other 3-clawed taxa forwhich deep sampling of lycosoids and dionychawould have littleutilityTaxondeposition andGenBank accession information canbe found in Appendix 1
Three non-RTA clade taxa were included as potentialoutgroups with exemplars of Nicodamidae Simon 1897(Megadictyna thilenii Dahl 1906) Eresidae C L Koch 1950(Stegodyphus sp) andAraneidaeClerck 1757 (Zygiella x-notataClerck 1757) used to represent the groups thought to be thenearest relatives including the sister-group to the RTA clade(Griswold et al 1999 2005 Spagna and Gillespie 2008)
Molecular markers and amplification conditions
Markers chosen for this study were nuclear sequences from theD-loop sectionof the 28S rRNAgene two large sections coveringgt90 of the 18S rRNA gene plus the full sequence for theprotein-codinghistoneH3geneThesemarkerswhich are amongthebest characterised in spiderswere specifically chosen for slowaccumulation of mutations making them useful for decipheringphylogenetic relationships in arachnids at relatively deep levels(Giribet et al 1996 Wheeler and Hayashi 1998 Maddison andHedin 2003 Murphy et al 2006) Faster-evolving markerssuch as mitochondrial cytochrome c oxidase subunit I (COI)and 16S rRNA were not used as they have been shown tocontribute little unsaturated signal to analysis of divergences atthis depth (Spagna and Gillespie 2008)
Theprimer pairs 28SOand28SC(Hedin andMaddison2001)were used to amplify a section of 28S rRNA sequence (~690 bp)in length Because some RTA-clade spiders appear to have28S rRNA paralogues (Murphy et al 2006 J C Spagna andR G Gillespie unpubl data) care was taken to make sure thatno double or multiple banding was seen in PCR amplificationproducts resulting in clean interpretable chromatogramswithoutdouble peaks for all sequences used in this project Two primerpairs 18S 1F and 18S 5R and 18S 4F and 18S 9R (Giribet et al1996) were used to amplify the first ~850 and second ~950 basepairs of the 18S rRNAgene respectively A 349 base pair sectionof histone H3 was amplified using the primer pairs histone H3aFand histone H3aR (Colgan et al 1998) For 28S rRNA FailsafePCR Optimization Kit (Epicentre Technologies Madison WI)buffers were used to improve PCR reaction yields
The PCR mixtures used were 12mL total final volume with abasic mix of 25mL dNTPs 10mL buffer 02mLTaq 25mL perprimer 15mL of template DNA and 18mL H2O The bufferused for each rRNA locus was determined using a FailSafeoptimization kit while standard buffer was used for thehistone H3 amplifications For PCR cycles a basic protocol of95C melting temperature for 30 s 55C annealing temperaturefor 40 s and72Cextension temperature for 45 swas repeated for35 cycles Minor modifications were made for individual taxa interms of temperatures and cycle segment lengths (PCR buffersused and cycle optimisations are available upon request from theauthors)
Alignment secondary structure assessmentand incorporation in partitioned models
Alignments of the 18S and 28S rRNA datasets were done usingMUSCLE MUSCLE was chosen for alignment for its relative
accuracy speed and ease of use (Edgar 2004) Alignment of28S rRNAwas also conducted using secondary structure modelsto produce a second set of aligned datasets Alignment usingsecondary structure has been demonstrated to be moreappropriate than traditional alignment methods (Kjer 1995Angelini and Jockusch 2008) Also finding the preciselocation of stems and loops allows for them to be modelledseparately in phylogenetic analysis and also allows for theimplementation of a doublet model (Kjer 2004) To accountfor secondary structure in 18S rRNA we used the pre-existingmodel of Hendriks et al (1988) based on sequences from atheraphosid spider which has been shown to be generallyapplicable across a broad range of arachnids (Spagna andGillespie 2006) There was no conflict between the MUSCLE-aligned 18S rRNA data and the secondary structure model usedthat required post hoc realignment so bases could simply beassigned to lsquostemsrsquo or lsquoloopsrsquo based on the Hendriks et al (1988)model
Modelling of the secondary structure of 28S rRNA was morecomplex because the interspecific variability in this marker madefolding and self-complementary pairing more ambiguousExisting models for arthropods (Hedin and Maddison 2001)and the mfold web server (Zuker 2003) were used to findappropriate structures for the amplified sections (~690 bp) ofthe D2 and D3 segments of 28S rRNA Mfold uses a free energyalgorithm following the thermodynamic rules of SantaLucia(1998) and the default parameters were used
Assessment of behaviour of themfold algorithmdemonstratedthat lower folding temperatures resulted in consistently greaterlevels of pairingwithin each arm than at the default temperature of37C (mean increase in the number of bases forming pairs was48at 21C49at 4C)Though this default temperature settingfor the mfold algorithm might appear high for an ectothermicanimal the spiders sampled live at a wide range of temperaturesoften exceeding 37C in the case of Saltonia so it falls withinrealistic physiological limits of the animals Because we lackcorroborating crystallographic data for our sequencesrsquo secondarystructures we used 37C as a conservative approach to base-pairmodelling as any inappropriate assignment of bases to stempartitions would result in underweighting these effectivelyreducing the size of the matrix
Each helix of the D2 region for each taxon was input singlyinto the server thus producing a possible structure or structuresfor eachhelix of each specimenThe entire lsquocore +D3rsquo regionwasinput for each specimen producing a possible structure Gapswere inserted and adjusted by eye to produce the final alignmentin such a way to minimise discontinuities in the stem structuresshared across taxa assuming that unpaired loop regions are morelikely to have undergone the indel events represented by the gapsIn the final alignment it was ambiguous whether certain baseswere stems or loops and in these cases the ambiguities wereassigned to loops
The protein coding gene histone H3 (349 base pairs) wasaligned by eye after the sequences were translated to codonswhich also allowed us to determine that therewere no stop codonsin any of the sequences
With the data matrix including lineages with known longbranches inconsistent molecular clocks (Spagna and Gillespie2008) anddeep divergences (gt100million years) (Penney 2003)
240 Invertebrate Systematics J C Spagna et al
we used a Bayesian likelihood approach to analyse the data inorder to avoid possible long-branch attraction artefacts(Felsenstein 1978 Bergsten 2005) Additionally secondarystructure information for rRNA genes using both pre-existingmodels and new models developed herein was included in theBayesian analyses in order to compensate for the non-independence of base-pairing in the self-complementaryhelices of the lsquostem regionsrsquo of rRNA
Although generally considered superior to parsimonymethods for dealing with long branches Bayesian methodshave also been shown to be potentially susceptible to suchartefacts (Kolaczkowski and Thornton 2009) For this reasonwe performed accelerated maximum likelihood analysis usingRAxML-HPC 704 (Stamatakis 2006a) on the CIPRES Portal v21 (Miller et al 2009) for comparison with Bayesian outcomesThe matrix was partitioned into stems and loops for the rRNAgenes and by codon position for histone H3 using the GTRCATmodel (Stamatakis 2006b) for each partition with the number ofrate categories set to 25 Bootstrapping was conducted usingthe rapid bootstrap algorithm for webservers (Stamatakis et al2008) and these values were calculated for nodes on the best-scoring tree
Bayesian analysis was conducted on each gene separatelyand in one concatenated dataset for both the MUSCLE-aligneddatasets and the structurally aligned datasets using mixed-modelpartitioned analyses (Nylander 2004 Brandley et al 2005) inMrBayes 31 (Huelsenbeck and Ronquist 2001 Ronquist andHuelsenbeck 2003) The rRNA genes were partitioned by stemshalf-stems and loops using a doublet model of nucleotidesubstitution (Kjer 2004 Angelini and Jockusch 2008) for fullstems to account for the non-independence of bases for which wehad explicit information about complementary base positionswhile histone H3 was partitioned by codon position The doubletmodel as implemented in MrBayes 31 is highly parameterised(16 parameters total) but previous studies have shown it to besuperior to models that assume independence (Tsagkogeorgaet al 2009) The primary risk from over-parameterisation isan increase in variance of parameters (eg branch lengths)rather than decreased accuracy in consensus tree topology(Huelsenbeck and Rannala 2004 Kelchner and Thomas 2007)For all non-stem partitions models were chosen using the AIC inMrModeltest 22 (Nylander 2004 Posada and Buckley 2004)Chosen models are shown in Table 1 Gaps were treated aslsquomissing datarsquo in all analyses The 28S rRNA and H3 datasetswere run for 10million generations the 18S rRNA dataset wasrun for 9million generations and the concatenated dataset wasrun for 12million generations The cumulative and comparecommands were used in AWTY (Wilgenbusch et al 2004) todetermine burn-in and compare the two results of each MrBayesrun
Results
Modelling of secondary structure
The 18S rRNA model for spiders of Hendriks et al (1988) wasfound to be sufficient for assignment of strings of bases to looppartitions and pairs of bases to complementary stem partitionsfor model and parameter partitioning before analysis The 28SrRNAdata were considerablymore variable in terms of both base
and length polymorphisms and the secondary structure modelsdeveloped for the region sequenced are seen in Figs 1ndash5
Arm 1 is the most variable of the 28S rRNA structuresmodelled and a representation of the variation can be seen inFig 2 Variation occurs both between and within clades (seephylogenetic results) in terms of total transcript length andnumber and size of loops and stems Within-taxon variationoccurs even in species of the same genus (such as Paratheumainsulana (Banks 1902) and P armata (Marples 1964) Fig 2Group1) Of groups represented in Fig 1 the Cybaeidae sl anddictynoids+Desis appear themost structurally uniform with theAgelenidae (including the subfamily Coelotinae) appearing quitevariable
Structures for Arm 2 aremuchmore uniform thanArm 1 withmost variation occurring in the form of presence or absence of asingle loop (represented by an asterisk in Fig 3) in the terminalhelix of the arm (Fig 3 panels 1ndash4)
While longer than both Arm 1 and Arm 2 across the spidertaxa modelled the Arm 3s (Fig 4) are similar in uniformity toArm 2 with a set of 7ndash8 positionally homologisable loopstructures Though not all Arm 3s have identical numbers ofloops the task of homologising individual loops was madesimpler by the amount of asymmetry possibly resulting fromslip-strand mispairing events (Gillespie et al 2004) between the50 and 30 sides of the intervening loops as these were oftenremarkably unequal and this asymmetry was fairly consistentacross taxa
The primary sequences for the D3 + core structures were inmost cases highly similar and thus alignment was largely trivialsimplifying the modelling of stems and loops across multipletaxa with some exceptions The structure for Zygiella x-notata(Fig 5) represents the structure for conforming taxa while thethree taxa with notable variations from the primary sequencelength (Paratheuma Saltonia and Dictyna) were modelledindividually Structurally Paratheuma and Dictyna appear toshare a positionally homologous self-complementary four-base-pair motif which makes the centre of the D-loop core appearlsquopinchedrsquo into two smaller loops (top 2 panels of (F 5) which isnot seen in the other study taxa
Combined data phylogeny
The majority-rule consensus tree for 6000 post-burn-incombined-data trees (representing 6million generations after4million generations were discarded as burn-in) is shown inFig 6 The monophyly of the ingroup representing all sampledmembers of the RTA clade was strongly supported (posteriorprobability (pp) = 10) The focal taxa of this study (those with
Table 1 Partitions and models used in partitioned Bayesian analyses
Partition Selected model
18S rRNA stems HKY+I+G (doublet)18S rRNA loops GTR+I+GHistone H3 position 1 GTR+I+GHistone H3 position 2 HKYHistone H3 position 3 GTR+I+G28S rRNA stems GTR+G(doublet)28S rRNA loops GTR+G28S rRNA half-stems GTR+G
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 241
GC
G
GC
UU
A
U
C
AG
G
G
G
G
G
G
U
GC
C
A
GC
A
C
A
A
A
AU
U
G
C
UU
A
CU
C
C
CU
C
CC
C A
C
U
C
U
G
G
G
C
C
U
G
U
C
C
G
GGG
U
U
U
U
U
U
G
U
U
G
G
U
G
G
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
5
G C C
G G A
U
GG
C
C
G
U
G
C
G
C
C
C
G
C
A
C
U
U
G
U
G
C
GC
CG
G
G
G
G
G
C
G
A
C
G
C
A
GA
G
CC
G
C
G
G
C
C
C
G
G
C
G
C
C
U
U
C
A
A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
C
C
G
U
C
G
C
A
UU
A
A
C
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
UCU
CCC
GU
CCC
G A GG
ACG C U G C A
CA
CGGUC
UG
GU A U G G U AAA C G
AC
UC
UC
GG
A U GG A G A
AGUCGACA
GA C C
GUGGUGAC
G C C GG
CCC
GA A
CG
GGG A G U C G
C
UGCAGC
U
U
G GC
G
C
G
C
G
A
C
C
GUGG
U
G
A
CG
CC
GG
CC
C G A A C GG
G Y A G U CG
C
3
ARM 1
ARM 2
ARM 3
ldquoCORErdquoldquoD3rdquo
Fig 1 Hypothesis of secondary structure for the D2 and D3 regions and the core region connecting them for Saltonia incerta showing relative positions ofthe Core D3 and Arms 1ndash3 in the 50 (lower left) to 30 (upper right) Arm and core orientations are maintained in Figs 2ndash5 for comparison
242 Invertebrate Systematics J C Spagna et al
G U C
C A GU
A
A
G
G G GG
G
GGG
GG
G
G
GG
G
A
A
A
A
A
A
A
AA
AU
U
U
UU U
C
C
C
CC
C
C
C CCC CCC
C
CC
U
G G
G
G
G G
C
C C CCC C
G CG G G
C
CC
G
G
GU U UU
GG
G
U
U
U
A
A A
U
A
AA
G
G G
A
A
Arg
yron
eta
aqua
tica
G U C
C A GU
A
A
GG
GGGGG
GG
GG
GGG
G
GGG
GG
A A
A
A
AA
A
AAAA
AAA
AA
AU
U
U
U
U
U
U UU
UU
UU
U
U U
C
CC
C
C C C C
C
C C C CC
C
C CC
CC
C
C
C
C
U
G
G
G
G
G
G GG
G
G
G
C
C C
CC
C
CC C
Dic
tyna
sp
G U C
C A GU
A
A
G
G G
G
G
GGG
G
G
G
G
G
GGG
A A
A
A
AA
A
A
A
AA
A
AU U
U
U
UU
U U
CCC
C
C
C
C
C
C C
CC
C C
CCC
C
C
U
G G
G
G G G
C
C C CC
C C
G
C G G G
C
C
C
G
G
GU
U
U
U
GGG
U
UA A
UU
Par
athe
uma
arm
ata
G U C
C A GU
A
A
G
G
G
GGG
G
GG
G
GG
GG
G
GCUG
A A
A
A
AAA
A
A
A
A
AA
AU U
U
U
U
U
U U
C
C
C
C
C C
C
CC
C
C
C C
CCC
CC
C
C
U
G G
G
G
G GG
GG
G
C
C C
CC
C
C
C
G
C GG
G C CC
G
Para
theu
ma
insu
lana
G C G
G C
UU
A
U
C AG
G
G
G
G
G
GU
GCC
A GC
A
C
A
AAAU UG
C
UU
ACU
C
C
C U CCCC
A
C
U
C
U
G
G
G
C
C U GU
C
C
G G G G U
U
U
U
U
U
G
U
U
G
GU
G
G
Sal
toni
a in
certa
Aqu
atic
Cla
de
G U C
C A G
A
G G
U
UG C
U
A
GG
AU
G
UU
A
G
CCCC
U
A
C
U
G
G
G G
C
A
CG C
G
C
G
C
G
G
C
A
U
A
A
G
G
CC
A
U
U
G
G
C G GC CG C G
GU C
U
G
C
CG
G
C
U
A
A
C
G
U
A
GG
G
U
C
CC
UC
A
C
AGUUA
Am
aruo
bius
sim
ilis
Am
auro
biid
ae a
nd O
utgr
oups
G U U
C
A
C
GU
G
GC
U
GCCC
GU
U
A
G
G
G
A
G
A
A
A
U
A
A
G
A
C
AU
CC
U U
C
GU
C
G
A
UC
CC
GC
C
U
G
C
C
G
C
U
U
G
C C
CC
GG
C
CG
G
G
U
G
AA
GG
U
G
G
AG A
AC
C
G
GC
AU
AC
UU
U
Ste
gody
phus
sp
G U C
C A GG
C
A
G
G
A
U
GC
G
UC CAA
A
A
G
A
A
U
AU
U
U
C
U U
AU
U
A
CCCC C
G
G
G
C
U
G
G
C
G G
C
U
U
C
GU
A
G G G
U
U
C
A
A
U
G
G
G
G
GA
U
U
U
C
G
G
U
A
U
C
C C
C
C
G
C
G
G
AC
UG
UG
C
Meg
adic
tyna
thile
nii
G U C
C A GG
C
A
AA
G
G
GG
GG
GGC
G
GGGG
G
A
A
G
AG
A
AUAA
A
AC
AA
AU
U
UU
G
U
U
UUU
UA
C
G
U U
CC
CG
C U C C
G
U CC
CC
U
GC
U
C
C
U
G C
G
G
G CU
C
G
A
C G
C
C
C
G C
GCG U
Zygi
ella
x-n
otat
a5
3
G UC G C
U
A
G G
U
G CG
A
GGC
A
AU
U U
A
C
CC
G
G
C
U
G
G
C
G
GC
U
G
U
G
A
G
G
U
G
C
G
C
G
C
G
C
A
C
C
C
G
G
G
C
A
GGC
C
G
GA
G
A
C
C
U
CG
CA
A
U UGAA
A
CG
G
A
AAC
C
CC
CC
GG
G G
G
CC
Cry
phoe
ca s
p
G UC A G
C
A
G C
U
G
GC CG
A
CG
C
A
AU
G
U U
A
G
C
GG
G
G
C
U
G
A
C
C CG
G
C
G
U
A
A
G
GA
U
C
U
G
C
G
C
G
U
C
G
C
A
C
C
G
G
G
C
C
G
G
C G
C
UGC
G CA
G
G
AG
C
C
C
C
UC
GC
C
GA
U GCAG
A
CG
Bla
bom
ma
sp
G UC A G
C
A
U C
C
G
UG GG
A
CG
G
A
AU
G
U U
A
G
A
GA
C
G
C
U
G A
C
C CG
A
G
G
C
U
U
G
UA
U
G
G
U
A
U
A
C
G
U
C
G
C
A
C
C
G
C
GC C
G
G
C G
C
CGU
G CA
G
G
AG
C
G U
C
C
C
UC
GC
CGG
G GCCG
G
Cal
ymm
aria
sp
G UC G C
C
A
G G
A
G C
A
G
GU
U U
A
C
CG
G
C
C
U
GC G
C
G
C
G
C
G
G
G
U
G
C
U
G
G G
C
A
C
C
C
G
G
G
C
A
GCC
C
G
GA
G
A
C U
CG
C
U
C
U
AG
A
CG
G
A
A
AC CC
CG
G
G G
G
AA
AA
Rot
haei
na s
p
G UC A G
C
A
G C
A
G
G CG
A
CG
C
AAU
C
U U
A
G
C
G
G
G
C
C
U
G
A
C
C CG
C
C
G
C
G
A
G
GG
U
U
G
C
A
G
G
U
C
G
C
A
C
C
G
G
G
C C
G
G
C G
C
GCG
U CA
G
G
GG
C
C
C
C
UCG
CC
GA
U GCAG
A
CG
G
Cyb
aeus
sp
2
Cyb
aeid
ae s
l C
lade
G U C
C A GU
C
A
G GG
A
U
GC
G
G
GG GGG
A
A
G
G A
UAU
U U
C
U U
U
C
C
C
C
C
C
CC
G
CC
C
C
C
U
G
G
G
G
G
C
C
C
CG
C
G
G
G G
G
C
C
C
A
A
A
U
A
U
U
G
G
C
U
U
U
A
C
A
G
A
A
A
G
G
C
A
G
G
U
C
GG
GC
U
C
53
Text
rix d
entic
ulat
a
Age
leni
nes
+ C
oelo
tines
G U C
C A GUA
G
G
GG
G
G
GGG
A
A
G
C
A
AU
U
G
U U
A
C
C
C CCCC
C
U
C
G
C
U
G G
G
G G
C
C
C
CG
C
C
G
CC
C
G
G
U
G
U
G
G
C
C
U
G
C
A
G
G
A
A
G
C
A
A
G
G
GG
C
U
GC C
G
G G
CG
UCA
CUCG
Tege
naria
dom
estic
a
G U C
C A
A
C G
U
G
UG CC
A
G
A
AU
G
U U
G
CA
CCC
A
CU
G
G
C
U
G
G
G G
C
C
G
CG G
C
A
C
C
G
T
G
GU
GC
U
G
A
C
A
U
C
CGU
C
U
C
U
GC G G
C
G
UC
CGCA
G
C
GC
C
GU
Wad
otes
dix
iens
is
G U C
C A GUA
G
C
AG
GG GGG
A
UG
C
A
AU
U
G
U U
A
G
C CCCC
CU
C
U
C
U
G
G G
GG
C
C
C
CG
C
C
G
UC
C
G
A
U
A
U
GC
C
C
U
G
G
G
G
G
A
U
G
C
A
A
G
C
GGC
UC
A CG
G G
CG
UC
A
GCCA U
Coe
lote
s te
rres
tris
G U C
C A GU
A
A
G GG
GG
GG
G
G
GG GGG
A
A
A
A
A
A
A
AU
U
U
U
U U
C
CC
CC
C
C
CC
C
CC
C
C
C
U
G
G
G
G
G
C
C
C
CC
C
G
C
G G
G
C
C
C
G
G
G
U
U
U
UG
G
G
UU
U
A
C
A
U
A
A
A
G
G
A
A
UU
U
C
Alla
gele
na g
raci
lens
G U C
C A GU
C
A
G
G
AG
GG GGG
A
UG
C
A
AU
U
U
G
U U
A
C
C
C CCCG
CU
C
U
C
U
G
G
G
G G
C
C
C
GG
G
C
G
UC
C
G
A
U
A
U
G
G
C
U
U
G
C
A
G
G
A
C
G
C A
A
G
C
GG
C
UUC C
C
G G
CG
UC
Age
leno
psis
ape
rta
Lyco
soid
s +
Dio
nych
ans
G U C
C A GUA
C
CU
A
A
CC
G
G
G
A
U
C
A
A
AU
C
UG
U U
G
G
G
G
CCC C
A CG
C
U
G U
U G
C
C
C
U
GG
A
G
C G
G
C
A
A
GG
G
G
C
A
G
U
U
C
C
C
UA
A
G C
G
G
CG
G
U
G
C
U
G
G
G CG G
G A
CC
A AA
UU
U
Lath
ys a
lber
ta
Dic
tyno
ids
+ D
esis
G U C
C A G
A
G G
UUG CU
A
GG
AU
G
U U
A
G
CCCC
G
G
C
U
G
G
G C
C
A
C
CG U
G
C
G
C
G
A
C
A
U
A
A
G
G
A
C
C
A
A
T
G
G
C G GC
G
CG C
G
GC
C
C
G
A
C
UG G
C
C U
A
UC
G
A
A
GG
G
CU
U
GG
Met
alte
lla s
imon
i
G U C
G
A
C
G
G
G
GGC
A
GGGC
C
U
C
G
G
G
G
A
C
A
A
A
A
A
A
G
C
A
AAU G
UG
U U
C
CC
C
C
C C
G
C
CC
GCG
C
U
G
G
G
G
C
C U CCC C
CG CG
U
C
G
G
G
A
A
AG
U
U
A
U
G
C
AA
A
GC
GC
G
GC C
GUA
AA
C
Neo
ram
ia ja
nus
G U C
C A G
A
G G
AUG
C
G
A
G
C
AU
U U
G
GGG
A
A
GG
G
C
U
G
C
C C
C
C CG U
G
A
C
A
G
A
C
UC
U
G
A
U
C
U
C
A
U
U
C
A
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G
G
C
C
A
C
G
A
A
GG
G
U
G GC
C
C
C
UC
Des
is m
arin
a
G U C
C A G
A
G G
UUG CU
A
G
G
AU
A
U U
A
G
GGG
A
A
UG
G
C
U
G
C
C C
C
A CG C
G
C
G
C
G
A
C
UC
U
A
A
G
G
A
U
A
U
C
C
G
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G G
C
C
A
U
C
G
U
A
G GG
U
C G
G
C
Bad
umna
long
iqua
G U C
C A G
CC
A
G
G U
GG
G
G
G
G
C
GG
GC
G
G
GGG
A
A
A
A
A
A
A
A
A AAAAU
U
U
A
UUU U
CC U
C
C
C
U
C
UC
A G
C C
C C
C
U
C
U
G
G
G
G G
C
G
C
C
C
U
C
G
G
C UC G
C
U
U
U U UU
U
UU
A
U
Teng
ella
radi
ata
G U C
C A GU
A
A
A
CU
AA
UG G
G
G
A
U
C
A
A
AU UC
U
U U
G
A
A
C
GGG G
C
C
C
C
U
G G
UG C
C
C
U
GG U
G
U G
GC
A
A
G
GG
C
C
G
U
U
U
C
G
C
U A
C
G C
G
C
C
G
G
UG
CU
C
G
GC
C C GAGA
Alo
peco
sa k
ochi
G U C
C A GU
A
A
G
G
U
GCG
GG GGG
A
G
G
G
G
C
AU
C
AG
U U
A
C
C
C GGGC
GA
U
U
C
U
G
G
GC
C
C
UUG
U
G C
GU
C
C
G
U
U
G
U
C
G
C
U
A
C
A
C
A
G
G
C G
U
G
U
CC
G
U
C
CU
U
GC
UC
UGC
Any
phae
na c
alifo
rnic
a
G U C
G
UA
GG
GG
G
G
GG
G
GG
G
G
G
G
G
A
A
A
A
A
A
A
A
A
AA
A
AU
U
UU
U U
C
C
C
C
C
C C
C
C
CC
C
C
C
C
U
G
G
GG
G G
C
C C CCC C
G
CG GG
C
CC G
G
G
U
U
G
GGU
U
A
U
U
C
AA A
A A
G
G C
G
Stip
hidi
on fa
cetu
m
G U C
C A
A
C
G
U
G
C
U
A
C
G
AU
U
UU
A
G
UC
U
U
C
U
G
U
C
CG G
G
C
G
C
G
C
A
A
U
A
U
G
G
A
G
A
G
C
C
G G G
GCCC
C
GG
G
A
A
C
C
A
A
U
U
GC
C
C G
G
AU
C
C
AGU
C
G
G
G
U C CCU
G
Zelo
tes
sp
ARM
1
95 (8
0-10
7)10
15
(101
-102
)
Am
auro
biid
ae
103
5 (9
9-11
8)
Out
grou
ps
100
2 (9
0-10
6)10
04
(93-
104)
997
(97-
102)
101
1 (1
00-1
02)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig2
Groups1
and2thesecondarystructuresofthelsquoA
rm1rsquooftheD2region
of28
SrRNAfrom
mem
bersofthelsquoaqu
aticcladersquothe
amaurobiidcladeandthreeou
tgroup
taxa
from
thecombined-
datatreeA
rm1ishigh
lyvariableevenwith
ingeneraasdemon
stratedby
thetwospeciesof
Paratheum
aGroups3nad4Arm
1sfrom
mem
bersof
theAgelenidaeandCoelotin
ae(G
roup
3)and
thecybaeidsensulatoclade(G
roup
4)Ingeneralthelatterappearmorestructurally
conservedthan
theform
erG
roups5and6
Arm
1sfrom
mem
bersof
theLathys(D
esidae)andAustralclade
spiders(row5)andforL
ycosoids
andDiony
chans(Group
6)O
fthe
form
erthe
genu
sDesishasa
particularlyconservedsecondarystructureforthisfeatureThe
armsshowninthesecondarystructure
figureswerechosen
torepresentthe
rangeof
variationBoxes
totherighto
feach
grouprepresentm
eanlength
andminim
um-m
axim
umrange(inparentheses)to
capturefullvariationin
figu
red
andunfigu
redarms
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 243
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
relationships within the Dictynidae or other 3-clawed taxa forwhich deep sampling of lycosoids and dionychawould have littleutilityTaxondeposition andGenBank accession information canbe found in Appendix 1
Three non-RTA clade taxa were included as potentialoutgroups with exemplars of Nicodamidae Simon 1897(Megadictyna thilenii Dahl 1906) Eresidae C L Koch 1950(Stegodyphus sp) andAraneidaeClerck 1757 (Zygiella x-notataClerck 1757) used to represent the groups thought to be thenearest relatives including the sister-group to the RTA clade(Griswold et al 1999 2005 Spagna and Gillespie 2008)
Molecular markers and amplification conditions
Markers chosen for this study were nuclear sequences from theD-loop sectionof the 28S rRNAgene two large sections coveringgt90 of the 18S rRNA gene plus the full sequence for theprotein-codinghistoneH3geneThesemarkerswhich are amongthebest characterised in spiderswere specifically chosen for slowaccumulation of mutations making them useful for decipheringphylogenetic relationships in arachnids at relatively deep levels(Giribet et al 1996 Wheeler and Hayashi 1998 Maddison andHedin 2003 Murphy et al 2006) Faster-evolving markerssuch as mitochondrial cytochrome c oxidase subunit I (COI)and 16S rRNA were not used as they have been shown tocontribute little unsaturated signal to analysis of divergences atthis depth (Spagna and Gillespie 2008)
Theprimer pairs 28SOand28SC(Hedin andMaddison2001)were used to amplify a section of 28S rRNA sequence (~690 bp)in length Because some RTA-clade spiders appear to have28S rRNA paralogues (Murphy et al 2006 J C Spagna andR G Gillespie unpubl data) care was taken to make sure thatno double or multiple banding was seen in PCR amplificationproducts resulting in clean interpretable chromatogramswithoutdouble peaks for all sequences used in this project Two primerpairs 18S 1F and 18S 5R and 18S 4F and 18S 9R (Giribet et al1996) were used to amplify the first ~850 and second ~950 basepairs of the 18S rRNAgene respectively A 349 base pair sectionof histone H3 was amplified using the primer pairs histone H3aFand histone H3aR (Colgan et al 1998) For 28S rRNA FailsafePCR Optimization Kit (Epicentre Technologies Madison WI)buffers were used to improve PCR reaction yields
The PCR mixtures used were 12mL total final volume with abasic mix of 25mL dNTPs 10mL buffer 02mLTaq 25mL perprimer 15mL of template DNA and 18mL H2O The bufferused for each rRNA locus was determined using a FailSafeoptimization kit while standard buffer was used for thehistone H3 amplifications For PCR cycles a basic protocol of95C melting temperature for 30 s 55C annealing temperaturefor 40 s and72Cextension temperature for 45 swas repeated for35 cycles Minor modifications were made for individual taxa interms of temperatures and cycle segment lengths (PCR buffersused and cycle optimisations are available upon request from theauthors)
Alignment secondary structure assessmentand incorporation in partitioned models
Alignments of the 18S and 28S rRNA datasets were done usingMUSCLE MUSCLE was chosen for alignment for its relative
accuracy speed and ease of use (Edgar 2004) Alignment of28S rRNAwas also conducted using secondary structure modelsto produce a second set of aligned datasets Alignment usingsecondary structure has been demonstrated to be moreappropriate than traditional alignment methods (Kjer 1995Angelini and Jockusch 2008) Also finding the preciselocation of stems and loops allows for them to be modelledseparately in phylogenetic analysis and also allows for theimplementation of a doublet model (Kjer 2004) To accountfor secondary structure in 18S rRNA we used the pre-existingmodel of Hendriks et al (1988) based on sequences from atheraphosid spider which has been shown to be generallyapplicable across a broad range of arachnids (Spagna andGillespie 2006) There was no conflict between the MUSCLE-aligned 18S rRNA data and the secondary structure model usedthat required post hoc realignment so bases could simply beassigned to lsquostemsrsquo or lsquoloopsrsquo based on the Hendriks et al (1988)model
Modelling of the secondary structure of 28S rRNA was morecomplex because the interspecific variability in this marker madefolding and self-complementary pairing more ambiguousExisting models for arthropods (Hedin and Maddison 2001)and the mfold web server (Zuker 2003) were used to findappropriate structures for the amplified sections (~690 bp) ofthe D2 and D3 segments of 28S rRNA Mfold uses a free energyalgorithm following the thermodynamic rules of SantaLucia(1998) and the default parameters were used
Assessment of behaviour of themfold algorithmdemonstratedthat lower folding temperatures resulted in consistently greaterlevels of pairingwithin each arm than at the default temperature of37C (mean increase in the number of bases forming pairs was48at 21C49at 4C)Though this default temperature settingfor the mfold algorithm might appear high for an ectothermicanimal the spiders sampled live at a wide range of temperaturesoften exceeding 37C in the case of Saltonia so it falls withinrealistic physiological limits of the animals Because we lackcorroborating crystallographic data for our sequencesrsquo secondarystructures we used 37C as a conservative approach to base-pairmodelling as any inappropriate assignment of bases to stempartitions would result in underweighting these effectivelyreducing the size of the matrix
Each helix of the D2 region for each taxon was input singlyinto the server thus producing a possible structure or structuresfor eachhelix of each specimenThe entire lsquocore +D3rsquo regionwasinput for each specimen producing a possible structure Gapswere inserted and adjusted by eye to produce the final alignmentin such a way to minimise discontinuities in the stem structuresshared across taxa assuming that unpaired loop regions are morelikely to have undergone the indel events represented by the gapsIn the final alignment it was ambiguous whether certain baseswere stems or loops and in these cases the ambiguities wereassigned to loops
The protein coding gene histone H3 (349 base pairs) wasaligned by eye after the sequences were translated to codonswhich also allowed us to determine that therewere no stop codonsin any of the sequences
With the data matrix including lineages with known longbranches inconsistent molecular clocks (Spagna and Gillespie2008) anddeep divergences (gt100million years) (Penney 2003)
240 Invertebrate Systematics J C Spagna et al
we used a Bayesian likelihood approach to analyse the data inorder to avoid possible long-branch attraction artefacts(Felsenstein 1978 Bergsten 2005) Additionally secondarystructure information for rRNA genes using both pre-existingmodels and new models developed herein was included in theBayesian analyses in order to compensate for the non-independence of base-pairing in the self-complementaryhelices of the lsquostem regionsrsquo of rRNA
Although generally considered superior to parsimonymethods for dealing with long branches Bayesian methodshave also been shown to be potentially susceptible to suchartefacts (Kolaczkowski and Thornton 2009) For this reasonwe performed accelerated maximum likelihood analysis usingRAxML-HPC 704 (Stamatakis 2006a) on the CIPRES Portal v21 (Miller et al 2009) for comparison with Bayesian outcomesThe matrix was partitioned into stems and loops for the rRNAgenes and by codon position for histone H3 using the GTRCATmodel (Stamatakis 2006b) for each partition with the number ofrate categories set to 25 Bootstrapping was conducted usingthe rapid bootstrap algorithm for webservers (Stamatakis et al2008) and these values were calculated for nodes on the best-scoring tree
Bayesian analysis was conducted on each gene separatelyand in one concatenated dataset for both the MUSCLE-aligneddatasets and the structurally aligned datasets using mixed-modelpartitioned analyses (Nylander 2004 Brandley et al 2005) inMrBayes 31 (Huelsenbeck and Ronquist 2001 Ronquist andHuelsenbeck 2003) The rRNA genes were partitioned by stemshalf-stems and loops using a doublet model of nucleotidesubstitution (Kjer 2004 Angelini and Jockusch 2008) for fullstems to account for the non-independence of bases for which wehad explicit information about complementary base positionswhile histone H3 was partitioned by codon position The doubletmodel as implemented in MrBayes 31 is highly parameterised(16 parameters total) but previous studies have shown it to besuperior to models that assume independence (Tsagkogeorgaet al 2009) The primary risk from over-parameterisation isan increase in variance of parameters (eg branch lengths)rather than decreased accuracy in consensus tree topology(Huelsenbeck and Rannala 2004 Kelchner and Thomas 2007)For all non-stem partitions models were chosen using the AIC inMrModeltest 22 (Nylander 2004 Posada and Buckley 2004)Chosen models are shown in Table 1 Gaps were treated aslsquomissing datarsquo in all analyses The 28S rRNA and H3 datasetswere run for 10million generations the 18S rRNA dataset wasrun for 9million generations and the concatenated dataset wasrun for 12million generations The cumulative and comparecommands were used in AWTY (Wilgenbusch et al 2004) todetermine burn-in and compare the two results of each MrBayesrun
Results
Modelling of secondary structure
The 18S rRNA model for spiders of Hendriks et al (1988) wasfound to be sufficient for assignment of strings of bases to looppartitions and pairs of bases to complementary stem partitionsfor model and parameter partitioning before analysis The 28SrRNAdata were considerablymore variable in terms of both base
and length polymorphisms and the secondary structure modelsdeveloped for the region sequenced are seen in Figs 1ndash5
Arm 1 is the most variable of the 28S rRNA structuresmodelled and a representation of the variation can be seen inFig 2 Variation occurs both between and within clades (seephylogenetic results) in terms of total transcript length andnumber and size of loops and stems Within-taxon variationoccurs even in species of the same genus (such as Paratheumainsulana (Banks 1902) and P armata (Marples 1964) Fig 2Group1) Of groups represented in Fig 1 the Cybaeidae sl anddictynoids+Desis appear themost structurally uniform with theAgelenidae (including the subfamily Coelotinae) appearing quitevariable
Structures for Arm 2 aremuchmore uniform thanArm 1 withmost variation occurring in the form of presence or absence of asingle loop (represented by an asterisk in Fig 3) in the terminalhelix of the arm (Fig 3 panels 1ndash4)
While longer than both Arm 1 and Arm 2 across the spidertaxa modelled the Arm 3s (Fig 4) are similar in uniformity toArm 2 with a set of 7ndash8 positionally homologisable loopstructures Though not all Arm 3s have identical numbers ofloops the task of homologising individual loops was madesimpler by the amount of asymmetry possibly resulting fromslip-strand mispairing events (Gillespie et al 2004) between the50 and 30 sides of the intervening loops as these were oftenremarkably unequal and this asymmetry was fairly consistentacross taxa
The primary sequences for the D3 + core structures were inmost cases highly similar and thus alignment was largely trivialsimplifying the modelling of stems and loops across multipletaxa with some exceptions The structure for Zygiella x-notata(Fig 5) represents the structure for conforming taxa while thethree taxa with notable variations from the primary sequencelength (Paratheuma Saltonia and Dictyna) were modelledindividually Structurally Paratheuma and Dictyna appear toshare a positionally homologous self-complementary four-base-pair motif which makes the centre of the D-loop core appearlsquopinchedrsquo into two smaller loops (top 2 panels of (F 5) which isnot seen in the other study taxa
Combined data phylogeny
The majority-rule consensus tree for 6000 post-burn-incombined-data trees (representing 6million generations after4million generations were discarded as burn-in) is shown inFig 6 The monophyly of the ingroup representing all sampledmembers of the RTA clade was strongly supported (posteriorprobability (pp) = 10) The focal taxa of this study (those with
Table 1 Partitions and models used in partitioned Bayesian analyses
Partition Selected model
18S rRNA stems HKY+I+G (doublet)18S rRNA loops GTR+I+GHistone H3 position 1 GTR+I+GHistone H3 position 2 HKYHistone H3 position 3 GTR+I+G28S rRNA stems GTR+G(doublet)28S rRNA loops GTR+G28S rRNA half-stems GTR+G
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 241
GC
G
GC
UU
A
U
C
AG
G
G
G
G
G
G
U
GC
C
A
GC
A
C
A
A
A
AU
U
G
C
UU
A
CU
C
C
CU
C
CC
C A
C
U
C
U
G
G
G
C
C
U
G
U
C
C
G
GGG
U
U
U
U
U
U
G
U
U
G
G
U
G
G
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
5
G C C
G G A
U
GG
C
C
G
U
G
C
G
C
C
C
G
C
A
C
U
U
G
U
G
C
GC
CG
G
G
G
G
G
C
G
A
C
G
C
A
GA
G
CC
G
C
G
G
C
C
C
G
G
C
G
C
C
U
U
C
A
A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
C
C
G
U
C
G
C
A
UU
A
A
C
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
UCU
CCC
GU
CCC
G A GG
ACG C U G C A
CA
CGGUC
UG
GU A U G G U AAA C G
AC
UC
UC
GG
A U GG A G A
AGUCGACA
GA C C
GUGGUGAC
G C C GG
CCC
GA A
CG
GGG A G U C G
C
UGCAGC
U
U
G GC
G
C
G
C
G
A
C
C
GUGG
U
G
A
CG
CC
GG
CC
C G A A C GG
G Y A G U CG
C
3
ARM 1
ARM 2
ARM 3
ldquoCORErdquoldquoD3rdquo
Fig 1 Hypothesis of secondary structure for the D2 and D3 regions and the core region connecting them for Saltonia incerta showing relative positions ofthe Core D3 and Arms 1ndash3 in the 50 (lower left) to 30 (upper right) Arm and core orientations are maintained in Figs 2ndash5 for comparison
242 Invertebrate Systematics J C Spagna et al
G U C
C A GU
A
A
G
G G GG
G
GGG
GG
G
G
GG
G
A
A
A
A
A
A
A
AA
AU
U
U
UU U
C
C
C
CC
C
C
C CCC CCC
C
CC
U
G G
G
G
G G
C
C C CCC C
G CG G G
C
CC
G
G
GU U UU
GG
G
U
U
U
A
A A
U
A
AA
G
G G
A
A
Arg
yron
eta
aqua
tica
G U C
C A GU
A
A
GG
GGGGG
GG
GG
GGG
G
GGG
GG
A A
A
A
AA
A
AAAA
AAA
AA
AU
U
U
U
U
U
U UU
UU
UU
U
U U
C
CC
C
C C C C
C
C C C CC
C
C CC
CC
C
C
C
C
U
G
G
G
G
G
G GG
G
G
G
C
C C
CC
C
CC C
Dic
tyna
sp
G U C
C A GU
A
A
G
G G
G
G
GGG
G
G
G
G
G
GGG
A A
A
A
AA
A
A
A
AA
A
AU U
U
U
UU
U U
CCC
C
C
C
C
C
C C
CC
C C
CCC
C
C
U
G G
G
G G G
C
C C CC
C C
G
C G G G
C
C
C
G
G
GU
U
U
U
GGG
U
UA A
UU
Par
athe
uma
arm
ata
G U C
C A GU
A
A
G
G
G
GGG
G
GG
G
GG
GG
G
GCUG
A A
A
A
AAA
A
A
A
A
AA
AU U
U
U
U
U
U U
C
C
C
C
C C
C
CC
C
C
C C
CCC
CC
C
C
U
G G
G
G
G GG
GG
G
C
C C
CC
C
C
C
G
C GG
G C CC
G
Para
theu
ma
insu
lana
G C G
G C
UU
A
U
C AG
G
G
G
G
G
GU
GCC
A GC
A
C
A
AAAU UG
C
UU
ACU
C
C
C U CCCC
A
C
U
C
U
G
G
G
C
C U GU
C
C
G G G G U
U
U
U
U
U
G
U
U
G
GU
G
G
Sal
toni
a in
certa
Aqu
atic
Cla
de
G U C
C A G
A
G G
U
UG C
U
A
GG
AU
G
UU
A
G
CCCC
U
A
C
U
G
G
G G
C
A
CG C
G
C
G
C
G
G
C
A
U
A
A
G
G
CC
A
U
U
G
G
C G GC CG C G
GU C
U
G
C
CG
G
C
U
A
A
C
G
U
A
GG
G
U
C
CC
UC
A
C
AGUUA
Am
aruo
bius
sim
ilis
Am
auro
biid
ae a
nd O
utgr
oups
G U U
C
A
C
GU
G
GC
U
GCCC
GU
U
A
G
G
G
A
G
A
A
A
U
A
A
G
A
C
AU
CC
U U
C
GU
C
G
A
UC
CC
GC
C
U
G
C
C
G
C
U
U
G
C C
CC
GG
C
CG
G
G
U
G
AA
GG
U
G
G
AG A
AC
C
G
GC
AU
AC
UU
U
Ste
gody
phus
sp
G U C
C A GG
C
A
G
G
A
U
GC
G
UC CAA
A
A
G
A
A
U
AU
U
U
C
U U
AU
U
A
CCCC C
G
G
G
C
U
G
G
C
G G
C
U
U
C
GU
A
G G G
U
U
C
A
A
U
G
G
G
G
GA
U
U
U
C
G
G
U
A
U
C
C C
C
C
G
C
G
G
AC
UG
UG
C
Meg
adic
tyna
thile
nii
G U C
C A GG
C
A
AA
G
G
GG
GG
GGC
G
GGGG
G
A
A
G
AG
A
AUAA
A
AC
AA
AU
U
UU
G
U
U
UUU
UA
C
G
U U
CC
CG
C U C C
G
U CC
CC
U
GC
U
C
C
U
G C
G
G
G CU
C
G
A
C G
C
C
C
G C
GCG U
Zygi
ella
x-n
otat
a5
3
G UC G C
U
A
G G
U
G CG
A
GGC
A
AU
U U
A
C
CC
G
G
C
U
G
G
C
G
GC
U
G
U
G
A
G
G
U
G
C
G
C
G
C
G
C
A
C
C
C
G
G
G
C
A
GGC
C
G
GA
G
A
C
C
U
CG
CA
A
U UGAA
A
CG
G
A
AAC
C
CC
CC
GG
G G
G
CC
Cry
phoe
ca s
p
G UC A G
C
A
G C
U
G
GC CG
A
CG
C
A
AU
G
U U
A
G
C
GG
G
G
C
U
G
A
C
C CG
G
C
G
U
A
A
G
GA
U
C
U
G
C
G
C
G
U
C
G
C
A
C
C
G
G
G
C
C
G
G
C G
C
UGC
G CA
G
G
AG
C
C
C
C
UC
GC
C
GA
U GCAG
A
CG
Bla
bom
ma
sp
G UC A G
C
A
U C
C
G
UG GG
A
CG
G
A
AU
G
U U
A
G
A
GA
C
G
C
U
G A
C
C CG
A
G
G
C
U
U
G
UA
U
G
G
U
A
U
A
C
G
U
C
G
C
A
C
C
G
C
GC C
G
G
C G
C
CGU
G CA
G
G
AG
C
G U
C
C
C
UC
GC
CGG
G GCCG
G
Cal
ymm
aria
sp
G UC G C
C
A
G G
A
G C
A
G
GU
U U
A
C
CG
G
C
C
U
GC G
C
G
C
G
C
G
G
G
U
G
C
U
G
G G
C
A
C
C
C
G
G
G
C
A
GCC
C
G
GA
G
A
C U
CG
C
U
C
U
AG
A
CG
G
A
A
AC CC
CG
G
G G
G
AA
AA
Rot
haei
na s
p
G UC A G
C
A
G C
A
G
G CG
A
CG
C
AAU
C
U U
A
G
C
G
G
G
C
C
U
G
A
C
C CG
C
C
G
C
G
A
G
GG
U
U
G
C
A
G
G
U
C
G
C
A
C
C
G
G
G
C C
G
G
C G
C
GCG
U CA
G
G
GG
C
C
C
C
UCG
CC
GA
U GCAG
A
CG
G
Cyb
aeus
sp
2
Cyb
aeid
ae s
l C
lade
G U C
C A GU
C
A
G GG
A
U
GC
G
G
GG GGG
A
A
G
G A
UAU
U U
C
U U
U
C
C
C
C
C
C
CC
G
CC
C
C
C
U
G
G
G
G
G
C
C
C
CG
C
G
G
G G
G
C
C
C
A
A
A
U
A
U
U
G
G
C
U
U
U
A
C
A
G
A
A
A
G
G
C
A
G
G
U
C
GG
GC
U
C
53
Text
rix d
entic
ulat
a
Age
leni
nes
+ C
oelo
tines
G U C
C A GUA
G
G
GG
G
G
GGG
A
A
G
C
A
AU
U
G
U U
A
C
C
C CCCC
C
U
C
G
C
U
G G
G
G G
C
C
C
CG
C
C
G
CC
C
G
G
U
G
U
G
G
C
C
U
G
C
A
G
G
A
A
G
C
A
A
G
G
GG
C
U
GC C
G
G G
CG
UCA
CUCG
Tege
naria
dom
estic
a
G U C
C A
A
C G
U
G
UG CC
A
G
A
AU
G
U U
G
CA
CCC
A
CU
G
G
C
U
G
G
G G
C
C
G
CG G
C
A
C
C
G
T
G
GU
GC
U
G
A
C
A
U
C
CGU
C
U
C
U
GC G G
C
G
UC
CGCA
G
C
GC
C
GU
Wad
otes
dix
iens
is
G U C
C A GUA
G
C
AG
GG GGG
A
UG
C
A
AU
U
G
U U
A
G
C CCCC
CU
C
U
C
U
G
G G
GG
C
C
C
CG
C
C
G
UC
C
G
A
U
A
U
GC
C
C
U
G
G
G
G
G
A
U
G
C
A
A
G
C
GGC
UC
A CG
G G
CG
UC
A
GCCA U
Coe
lote
s te
rres
tris
G U C
C A GU
A
A
G GG
GG
GG
G
G
GG GGG
A
A
A
A
A
A
A
AU
U
U
U
U U
C
CC
CC
C
C
CC
C
CC
C
C
C
U
G
G
G
G
G
C
C
C
CC
C
G
C
G G
G
C
C
C
G
G
G
U
U
U
UG
G
G
UU
U
A
C
A
U
A
A
A
G
G
A
A
UU
U
C
Alla
gele
na g
raci
lens
G U C
C A GU
C
A
G
G
AG
GG GGG
A
UG
C
A
AU
U
U
G
U U
A
C
C
C CCCG
CU
C
U
C
U
G
G
G
G G
C
C
C
GG
G
C
G
UC
C
G
A
U
A
U
G
G
C
U
U
G
C
A
G
G
A
C
G
C A
A
G
C
GG
C
UUC C
C
G G
CG
UC
Age
leno
psis
ape
rta
Lyco
soid
s +
Dio
nych
ans
G U C
C A GUA
C
CU
A
A
CC
G
G
G
A
U
C
A
A
AU
C
UG
U U
G
G
G
G
CCC C
A CG
C
U
G U
U G
C
C
C
U
GG
A
G
C G
G
C
A
A
GG
G
G
C
A
G
U
U
C
C
C
UA
A
G C
G
G
CG
G
U
G
C
U
G
G
G CG G
G A
CC
A AA
UU
U
Lath
ys a
lber
ta
Dic
tyno
ids
+ D
esis
G U C
C A G
A
G G
UUG CU
A
GG
AU
G
U U
A
G
CCCC
G
G
C
U
G
G
G C
C
A
C
CG U
G
C
G
C
G
A
C
A
U
A
A
G
G
A
C
C
A
A
T
G
G
C G GC
G
CG C
G
GC
C
C
G
A
C
UG G
C
C U
A
UC
G
A
A
GG
G
CU
U
GG
Met
alte
lla s
imon
i
G U C
G
A
C
G
G
G
GGC
A
GGGC
C
U
C
G
G
G
G
A
C
A
A
A
A
A
A
G
C
A
AAU G
UG
U U
C
CC
C
C
C C
G
C
CC
GCG
C
U
G
G
G
G
C
C U CCC C
CG CG
U
C
G
G
G
A
A
AG
U
U
A
U
G
C
AA
A
GC
GC
G
GC C
GUA
AA
C
Neo
ram
ia ja
nus
G U C
C A G
A
G G
AUG
C
G
A
G
C
AU
U U
G
GGG
A
A
GG
G
C
U
G
C
C C
C
C CG U
G
A
C
A
G
A
C
UC
U
G
A
U
C
U
C
A
U
U
C
A
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G
G
C
C
A
C
G
A
A
GG
G
U
G GC
C
C
C
UC
Des
is m
arin
a
G U C
C A G
A
G G
UUG CU
A
G
G
AU
A
U U
A
G
GGG
A
A
UG
G
C
U
G
C
C C
C
A CG C
G
C
G
C
G
A
C
UC
U
A
A
G
G
A
U
A
U
C
C
G
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G G
C
C
A
U
C
G
U
A
G GG
U
C G
G
C
Bad
umna
long
iqua
G U C
C A G
CC
A
G
G U
GG
G
G
G
G
C
GG
GC
G
G
GGG
A
A
A
A
A
A
A
A
A AAAAU
U
U
A
UUU U
CC U
C
C
C
U
C
UC
A G
C C
C C
C
U
C
U
G
G
G
G G
C
G
C
C
C
U
C
G
G
C UC G
C
U
U
U U UU
U
UU
A
U
Teng
ella
radi
ata
G U C
C A GU
A
A
A
CU
AA
UG G
G
G
A
U
C
A
A
AU UC
U
U U
G
A
A
C
GGG G
C
C
C
C
U
G G
UG C
C
C
U
GG U
G
U G
GC
A
A
G
GG
C
C
G
U
U
U
C
G
C
U A
C
G C
G
C
C
G
G
UG
CU
C
G
GC
C C GAGA
Alo
peco
sa k
ochi
G U C
C A GU
A
A
G
G
U
GCG
GG GGG
A
G
G
G
G
C
AU
C
AG
U U
A
C
C
C GGGC
GA
U
U
C
U
G
G
GC
C
C
UUG
U
G C
GU
C
C
G
U
U
G
U
C
G
C
U
A
C
A
C
A
G
G
C G
U
G
U
CC
G
U
C
CU
U
GC
UC
UGC
Any
phae
na c
alifo
rnic
a
G U C
G
UA
GG
GG
G
G
GG
G
GG
G
G
G
G
G
A
A
A
A
A
A
A
A
A
AA
A
AU
U
UU
U U
C
C
C
C
C
C C
C
C
CC
C
C
C
C
U
G
G
GG
G G
C
C C CCC C
G
CG GG
C
CC G
G
G
U
U
G
GGU
U
A
U
U
C
AA A
A A
G
G C
G
Stip
hidi
on fa
cetu
m
G U C
C A
A
C
G
U
G
C
U
A
C
G
AU
U
UU
A
G
UC
U
U
C
U
G
U
C
CG G
G
C
G
C
G
C
A
A
U
A
U
G
G
A
G
A
G
C
C
G G G
GCCC
C
GG
G
A
A
C
C
A
A
U
U
GC
C
C G
G
AU
C
C
AGU
C
G
G
G
U C CCU
G
Zelo
tes
sp
ARM
1
95 (8
0-10
7)10
15
(101
-102
)
Am
auro
biid
ae
103
5 (9
9-11
8)
Out
grou
ps
100
2 (9
0-10
6)10
04
(93-
104)
997
(97-
102)
101
1 (1
00-1
02)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig2
Groups1
and2thesecondarystructuresofthelsquoA
rm1rsquooftheD2region
of28
SrRNAfrom
mem
bersofthelsquoaqu
aticcladersquothe
amaurobiidcladeandthreeou
tgroup
taxa
from
thecombined-
datatreeA
rm1ishigh
lyvariableevenwith
ingeneraasdemon
stratedby
thetwospeciesof
Paratheum
aGroups3nad4Arm
1sfrom
mem
bersof
theAgelenidaeandCoelotin
ae(G
roup
3)and
thecybaeidsensulatoclade(G
roup
4)Ingeneralthelatterappearmorestructurally
conservedthan
theform
erG
roups5and6
Arm
1sfrom
mem
bersof
theLathys(D
esidae)andAustralclade
spiders(row5)andforL
ycosoids
andDiony
chans(Group
6)O
fthe
form
erthe
genu
sDesishasa
particularlyconservedsecondarystructureforthisfeatureThe
armsshowninthesecondarystructure
figureswerechosen
torepresentthe
rangeof
variationBoxes
totherighto
feach
grouprepresentm
eanlength
andminim
um-m
axim
umrange(inparentheses)to
capturefullvariationin
figu
red
andunfigu
redarms
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 243
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
we used a Bayesian likelihood approach to analyse the data inorder to avoid possible long-branch attraction artefacts(Felsenstein 1978 Bergsten 2005) Additionally secondarystructure information for rRNA genes using both pre-existingmodels and new models developed herein was included in theBayesian analyses in order to compensate for the non-independence of base-pairing in the self-complementaryhelices of the lsquostem regionsrsquo of rRNA
Although generally considered superior to parsimonymethods for dealing with long branches Bayesian methodshave also been shown to be potentially susceptible to suchartefacts (Kolaczkowski and Thornton 2009) For this reasonwe performed accelerated maximum likelihood analysis usingRAxML-HPC 704 (Stamatakis 2006a) on the CIPRES Portal v21 (Miller et al 2009) for comparison with Bayesian outcomesThe matrix was partitioned into stems and loops for the rRNAgenes and by codon position for histone H3 using the GTRCATmodel (Stamatakis 2006b) for each partition with the number ofrate categories set to 25 Bootstrapping was conducted usingthe rapid bootstrap algorithm for webservers (Stamatakis et al2008) and these values were calculated for nodes on the best-scoring tree
Bayesian analysis was conducted on each gene separatelyand in one concatenated dataset for both the MUSCLE-aligneddatasets and the structurally aligned datasets using mixed-modelpartitioned analyses (Nylander 2004 Brandley et al 2005) inMrBayes 31 (Huelsenbeck and Ronquist 2001 Ronquist andHuelsenbeck 2003) The rRNA genes were partitioned by stemshalf-stems and loops using a doublet model of nucleotidesubstitution (Kjer 2004 Angelini and Jockusch 2008) for fullstems to account for the non-independence of bases for which wehad explicit information about complementary base positionswhile histone H3 was partitioned by codon position The doubletmodel as implemented in MrBayes 31 is highly parameterised(16 parameters total) but previous studies have shown it to besuperior to models that assume independence (Tsagkogeorgaet al 2009) The primary risk from over-parameterisation isan increase in variance of parameters (eg branch lengths)rather than decreased accuracy in consensus tree topology(Huelsenbeck and Rannala 2004 Kelchner and Thomas 2007)For all non-stem partitions models were chosen using the AIC inMrModeltest 22 (Nylander 2004 Posada and Buckley 2004)Chosen models are shown in Table 1 Gaps were treated aslsquomissing datarsquo in all analyses The 28S rRNA and H3 datasetswere run for 10million generations the 18S rRNA dataset wasrun for 9million generations and the concatenated dataset wasrun for 12million generations The cumulative and comparecommands were used in AWTY (Wilgenbusch et al 2004) todetermine burn-in and compare the two results of each MrBayesrun
Results
Modelling of secondary structure
The 18S rRNA model for spiders of Hendriks et al (1988) wasfound to be sufficient for assignment of strings of bases to looppartitions and pairs of bases to complementary stem partitionsfor model and parameter partitioning before analysis The 28SrRNAdata were considerablymore variable in terms of both base
and length polymorphisms and the secondary structure modelsdeveloped for the region sequenced are seen in Figs 1ndash5
Arm 1 is the most variable of the 28S rRNA structuresmodelled and a representation of the variation can be seen inFig 2 Variation occurs both between and within clades (seephylogenetic results) in terms of total transcript length andnumber and size of loops and stems Within-taxon variationoccurs even in species of the same genus (such as Paratheumainsulana (Banks 1902) and P armata (Marples 1964) Fig 2Group1) Of groups represented in Fig 1 the Cybaeidae sl anddictynoids+Desis appear themost structurally uniform with theAgelenidae (including the subfamily Coelotinae) appearing quitevariable
Structures for Arm 2 aremuchmore uniform thanArm 1 withmost variation occurring in the form of presence or absence of asingle loop (represented by an asterisk in Fig 3) in the terminalhelix of the arm (Fig 3 panels 1ndash4)
While longer than both Arm 1 and Arm 2 across the spidertaxa modelled the Arm 3s (Fig 4) are similar in uniformity toArm 2 with a set of 7ndash8 positionally homologisable loopstructures Though not all Arm 3s have identical numbers ofloops the task of homologising individual loops was madesimpler by the amount of asymmetry possibly resulting fromslip-strand mispairing events (Gillespie et al 2004) between the50 and 30 sides of the intervening loops as these were oftenremarkably unequal and this asymmetry was fairly consistentacross taxa
The primary sequences for the D3 + core structures were inmost cases highly similar and thus alignment was largely trivialsimplifying the modelling of stems and loops across multipletaxa with some exceptions The structure for Zygiella x-notata(Fig 5) represents the structure for conforming taxa while thethree taxa with notable variations from the primary sequencelength (Paratheuma Saltonia and Dictyna) were modelledindividually Structurally Paratheuma and Dictyna appear toshare a positionally homologous self-complementary four-base-pair motif which makes the centre of the D-loop core appearlsquopinchedrsquo into two smaller loops (top 2 panels of (F 5) which isnot seen in the other study taxa
Combined data phylogeny
The majority-rule consensus tree for 6000 post-burn-incombined-data trees (representing 6million generations after4million generations were discarded as burn-in) is shown inFig 6 The monophyly of the ingroup representing all sampledmembers of the RTA clade was strongly supported (posteriorprobability (pp) = 10) The focal taxa of this study (those with
Table 1 Partitions and models used in partitioned Bayesian analyses
Partition Selected model
18S rRNA stems HKY+I+G (doublet)18S rRNA loops GTR+I+GHistone H3 position 1 GTR+I+GHistone H3 position 2 HKYHistone H3 position 3 GTR+I+G28S rRNA stems GTR+G(doublet)28S rRNA loops GTR+G28S rRNA half-stems GTR+G
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 241
GC
G
GC
UU
A
U
C
AG
G
G
G
G
G
G
U
GC
C
A
GC
A
C
A
A
A
AU
U
G
C
UU
A
CU
C
C
CU
C
CC
C A
C
U
C
U
G
G
G
C
C
U
G
U
C
C
G
GGG
U
U
U
U
U
U
G
U
U
G
G
U
G
G
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
5
G C C
G G A
U
GG
C
C
G
U
G
C
G
C
C
C
G
C
A
C
U
U
G
U
G
C
GC
CG
G
G
G
G
G
C
G
A
C
G
C
A
GA
G
CC
G
C
G
G
C
C
C
G
G
C
G
C
C
U
U
C
A
A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
C
C
G
U
C
G
C
A
UU
A
A
C
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
UCU
CCC
GU
CCC
G A GG
ACG C U G C A
CA
CGGUC
UG
GU A U G G U AAA C G
AC
UC
UC
GG
A U GG A G A
AGUCGACA
GA C C
GUGGUGAC
G C C GG
CCC
GA A
CG
GGG A G U C G
C
UGCAGC
U
U
G GC
G
C
G
C
G
A
C
C
GUGG
U
G
A
CG
CC
GG
CC
C G A A C GG
G Y A G U CG
C
3
ARM 1
ARM 2
ARM 3
ldquoCORErdquoldquoD3rdquo
Fig 1 Hypothesis of secondary structure for the D2 and D3 regions and the core region connecting them for Saltonia incerta showing relative positions ofthe Core D3 and Arms 1ndash3 in the 50 (lower left) to 30 (upper right) Arm and core orientations are maintained in Figs 2ndash5 for comparison
242 Invertebrate Systematics J C Spagna et al
G U C
C A GU
A
A
G
G G GG
G
GGG
GG
G
G
GG
G
A
A
A
A
A
A
A
AA
AU
U
U
UU U
C
C
C
CC
C
C
C CCC CCC
C
CC
U
G G
G
G
G G
C
C C CCC C
G CG G G
C
CC
G
G
GU U UU
GG
G
U
U
U
A
A A
U
A
AA
G
G G
A
A
Arg
yron
eta
aqua
tica
G U C
C A GU
A
A
GG
GGGGG
GG
GG
GGG
G
GGG
GG
A A
A
A
AA
A
AAAA
AAA
AA
AU
U
U
U
U
U
U UU
UU
UU
U
U U
C
CC
C
C C C C
C
C C C CC
C
C CC
CC
C
C
C
C
U
G
G
G
G
G
G GG
G
G
G
C
C C
CC
C
CC C
Dic
tyna
sp
G U C
C A GU
A
A
G
G G
G
G
GGG
G
G
G
G
G
GGG
A A
A
A
AA
A
A
A
AA
A
AU U
U
U
UU
U U
CCC
C
C
C
C
C
C C
CC
C C
CCC
C
C
U
G G
G
G G G
C
C C CC
C C
G
C G G G
C
C
C
G
G
GU
U
U
U
GGG
U
UA A
UU
Par
athe
uma
arm
ata
G U C
C A GU
A
A
G
G
G
GGG
G
GG
G
GG
GG
G
GCUG
A A
A
A
AAA
A
A
A
A
AA
AU U
U
U
U
U
U U
C
C
C
C
C C
C
CC
C
C
C C
CCC
CC
C
C
U
G G
G
G
G GG
GG
G
C
C C
CC
C
C
C
G
C GG
G C CC
G
Para
theu
ma
insu
lana
G C G
G C
UU
A
U
C AG
G
G
G
G
G
GU
GCC
A GC
A
C
A
AAAU UG
C
UU
ACU
C
C
C U CCCC
A
C
U
C
U
G
G
G
C
C U GU
C
C
G G G G U
U
U
U
U
U
G
U
U
G
GU
G
G
Sal
toni
a in
certa
Aqu
atic
Cla
de
G U C
C A G
A
G G
U
UG C
U
A
GG
AU
G
UU
A
G
CCCC
U
A
C
U
G
G
G G
C
A
CG C
G
C
G
C
G
G
C
A
U
A
A
G
G
CC
A
U
U
G
G
C G GC CG C G
GU C
U
G
C
CG
G
C
U
A
A
C
G
U
A
GG
G
U
C
CC
UC
A
C
AGUUA
Am
aruo
bius
sim
ilis
Am
auro
biid
ae a
nd O
utgr
oups
G U U
C
A
C
GU
G
GC
U
GCCC
GU
U
A
G
G
G
A
G
A
A
A
U
A
A
G
A
C
AU
CC
U U
C
GU
C
G
A
UC
CC
GC
C
U
G
C
C
G
C
U
U
G
C C
CC
GG
C
CG
G
G
U
G
AA
GG
U
G
G
AG A
AC
C
G
GC
AU
AC
UU
U
Ste
gody
phus
sp
G U C
C A GG
C
A
G
G
A
U
GC
G
UC CAA
A
A
G
A
A
U
AU
U
U
C
U U
AU
U
A
CCCC C
G
G
G
C
U
G
G
C
G G
C
U
U
C
GU
A
G G G
U
U
C
A
A
U
G
G
G
G
GA
U
U
U
C
G
G
U
A
U
C
C C
C
C
G
C
G
G
AC
UG
UG
C
Meg
adic
tyna
thile
nii
G U C
C A GG
C
A
AA
G
G
GG
GG
GGC
G
GGGG
G
A
A
G
AG
A
AUAA
A
AC
AA
AU
U
UU
G
U
U
UUU
UA
C
G
U U
CC
CG
C U C C
G
U CC
CC
U
GC
U
C
C
U
G C
G
G
G CU
C
G
A
C G
C
C
C
G C
GCG U
Zygi
ella
x-n
otat
a5
3
G UC G C
U
A
G G
U
G CG
A
GGC
A
AU
U U
A
C
CC
G
G
C
U
G
G
C
G
GC
U
G
U
G
A
G
G
U
G
C
G
C
G
C
G
C
A
C
C
C
G
G
G
C
A
GGC
C
G
GA
G
A
C
C
U
CG
CA
A
U UGAA
A
CG
G
A
AAC
C
CC
CC
GG
G G
G
CC
Cry
phoe
ca s
p
G UC A G
C
A
G C
U
G
GC CG
A
CG
C
A
AU
G
U U
A
G
C
GG
G
G
C
U
G
A
C
C CG
G
C
G
U
A
A
G
GA
U
C
U
G
C
G
C
G
U
C
G
C
A
C
C
G
G
G
C
C
G
G
C G
C
UGC
G CA
G
G
AG
C
C
C
C
UC
GC
C
GA
U GCAG
A
CG
Bla
bom
ma
sp
G UC A G
C
A
U C
C
G
UG GG
A
CG
G
A
AU
G
U U
A
G
A
GA
C
G
C
U
G A
C
C CG
A
G
G
C
U
U
G
UA
U
G
G
U
A
U
A
C
G
U
C
G
C
A
C
C
G
C
GC C
G
G
C G
C
CGU
G CA
G
G
AG
C
G U
C
C
C
UC
GC
CGG
G GCCG
G
Cal
ymm
aria
sp
G UC G C
C
A
G G
A
G C
A
G
GU
U U
A
C
CG
G
C
C
U
GC G
C
G
C
G
C
G
G
G
U
G
C
U
G
G G
C
A
C
C
C
G
G
G
C
A
GCC
C
G
GA
G
A
C U
CG
C
U
C
U
AG
A
CG
G
A
A
AC CC
CG
G
G G
G
AA
AA
Rot
haei
na s
p
G UC A G
C
A
G C
A
G
G CG
A
CG
C
AAU
C
U U
A
G
C
G
G
G
C
C
U
G
A
C
C CG
C
C
G
C
G
A
G
GG
U
U
G
C
A
G
G
U
C
G
C
A
C
C
G
G
G
C C
G
G
C G
C
GCG
U CA
G
G
GG
C
C
C
C
UCG
CC
GA
U GCAG
A
CG
G
Cyb
aeus
sp
2
Cyb
aeid
ae s
l C
lade
G U C
C A GU
C
A
G GG
A
U
GC
G
G
GG GGG
A
A
G
G A
UAU
U U
C
U U
U
C
C
C
C
C
C
CC
G
CC
C
C
C
U
G
G
G
G
G
C
C
C
CG
C
G
G
G G
G
C
C
C
A
A
A
U
A
U
U
G
G
C
U
U
U
A
C
A
G
A
A
A
G
G
C
A
G
G
U
C
GG
GC
U
C
53
Text
rix d
entic
ulat
a
Age
leni
nes
+ C
oelo
tines
G U C
C A GUA
G
G
GG
G
G
GGG
A
A
G
C
A
AU
U
G
U U
A
C
C
C CCCC
C
U
C
G
C
U
G G
G
G G
C
C
C
CG
C
C
G
CC
C
G
G
U
G
U
G
G
C
C
U
G
C
A
G
G
A
A
G
C
A
A
G
G
GG
C
U
GC C
G
G G
CG
UCA
CUCG
Tege
naria
dom
estic
a
G U C
C A
A
C G
U
G
UG CC
A
G
A
AU
G
U U
G
CA
CCC
A
CU
G
G
C
U
G
G
G G
C
C
G
CG G
C
A
C
C
G
T
G
GU
GC
U
G
A
C
A
U
C
CGU
C
U
C
U
GC G G
C
G
UC
CGCA
G
C
GC
C
GU
Wad
otes
dix
iens
is
G U C
C A GUA
G
C
AG
GG GGG
A
UG
C
A
AU
U
G
U U
A
G
C CCCC
CU
C
U
C
U
G
G G
GG
C
C
C
CG
C
C
G
UC
C
G
A
U
A
U
GC
C
C
U
G
G
G
G
G
A
U
G
C
A
A
G
C
GGC
UC
A CG
G G
CG
UC
A
GCCA U
Coe
lote
s te
rres
tris
G U C
C A GU
A
A
G GG
GG
GG
G
G
GG GGG
A
A
A
A
A
A
A
AU
U
U
U
U U
C
CC
CC
C
C
CC
C
CC
C
C
C
U
G
G
G
G
G
C
C
C
CC
C
G
C
G G
G
C
C
C
G
G
G
U
U
U
UG
G
G
UU
U
A
C
A
U
A
A
A
G
G
A
A
UU
U
C
Alla
gele
na g
raci
lens
G U C
C A GU
C
A
G
G
AG
GG GGG
A
UG
C
A
AU
U
U
G
U U
A
C
C
C CCCG
CU
C
U
C
U
G
G
G
G G
C
C
C
GG
G
C
G
UC
C
G
A
U
A
U
G
G
C
U
U
G
C
A
G
G
A
C
G
C A
A
G
C
GG
C
UUC C
C
G G
CG
UC
Age
leno
psis
ape
rta
Lyco
soid
s +
Dio
nych
ans
G U C
C A GUA
C
CU
A
A
CC
G
G
G
A
U
C
A
A
AU
C
UG
U U
G
G
G
G
CCC C
A CG
C
U
G U
U G
C
C
C
U
GG
A
G
C G
G
C
A
A
GG
G
G
C
A
G
U
U
C
C
C
UA
A
G C
G
G
CG
G
U
G
C
U
G
G
G CG G
G A
CC
A AA
UU
U
Lath
ys a
lber
ta
Dic
tyno
ids
+ D
esis
G U C
C A G
A
G G
UUG CU
A
GG
AU
G
U U
A
G
CCCC
G
G
C
U
G
G
G C
C
A
C
CG U
G
C
G
C
G
A
C
A
U
A
A
G
G
A
C
C
A
A
T
G
G
C G GC
G
CG C
G
GC
C
C
G
A
C
UG G
C
C U
A
UC
G
A
A
GG
G
CU
U
GG
Met
alte
lla s
imon
i
G U C
G
A
C
G
G
G
GGC
A
GGGC
C
U
C
G
G
G
G
A
C
A
A
A
A
A
A
G
C
A
AAU G
UG
U U
C
CC
C
C
C C
G
C
CC
GCG
C
U
G
G
G
G
C
C U CCC C
CG CG
U
C
G
G
G
A
A
AG
U
U
A
U
G
C
AA
A
GC
GC
G
GC C
GUA
AA
C
Neo
ram
ia ja
nus
G U C
C A G
A
G G
AUG
C
G
A
G
C
AU
U U
G
GGG
A
A
GG
G
C
U
G
C
C C
C
C CG U
G
A
C
A
G
A
C
UC
U
G
A
U
C
U
C
A
U
U
C
A
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G
G
C
C
A
C
G
A
A
GG
G
U
G GC
C
C
C
UC
Des
is m
arin
a
G U C
C A G
A
G G
UUG CU
A
G
G
AU
A
U U
A
G
GGG
A
A
UG
G
C
U
G
C
C C
C
A CG C
G
C
G
C
G
A
C
UC
U
A
A
G
G
A
U
A
U
C
C
G
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G G
C
C
A
U
C
G
U
A
G GG
U
C G
G
C
Bad
umna
long
iqua
G U C
C A G
CC
A
G
G U
GG
G
G
G
G
C
GG
GC
G
G
GGG
A
A
A
A
A
A
A
A
A AAAAU
U
U
A
UUU U
CC U
C
C
C
U
C
UC
A G
C C
C C
C
U
C
U
G
G
G
G G
C
G
C
C
C
U
C
G
G
C UC G
C
U
U
U U UU
U
UU
A
U
Teng
ella
radi
ata
G U C
C A GU
A
A
A
CU
AA
UG G
G
G
A
U
C
A
A
AU UC
U
U U
G
A
A
C
GGG G
C
C
C
C
U
G G
UG C
C
C
U
GG U
G
U G
GC
A
A
G
GG
C
C
G
U
U
U
C
G
C
U A
C
G C
G
C
C
G
G
UG
CU
C
G
GC
C C GAGA
Alo
peco
sa k
ochi
G U C
C A GU
A
A
G
G
U
GCG
GG GGG
A
G
G
G
G
C
AU
C
AG
U U
A
C
C
C GGGC
GA
U
U
C
U
G
G
GC
C
C
UUG
U
G C
GU
C
C
G
U
U
G
U
C
G
C
U
A
C
A
C
A
G
G
C G
U
G
U
CC
G
U
C
CU
U
GC
UC
UGC
Any
phae
na c
alifo
rnic
a
G U C
G
UA
GG
GG
G
G
GG
G
GG
G
G
G
G
G
A
A
A
A
A
A
A
A
A
AA
A
AU
U
UU
U U
C
C
C
C
C
C C
C
C
CC
C
C
C
C
U
G
G
GG
G G
C
C C CCC C
G
CG GG
C
CC G
G
G
U
U
G
GGU
U
A
U
U
C
AA A
A A
G
G C
G
Stip
hidi
on fa
cetu
m
G U C
C A
A
C
G
U
G
C
U
A
C
G
AU
U
UU
A
G
UC
U
U
C
U
G
U
C
CG G
G
C
G
C
G
C
A
A
U
A
U
G
G
A
G
A
G
C
C
G G G
GCCC
C
GG
G
A
A
C
C
A
A
U
U
GC
C
C G
G
AU
C
C
AGU
C
G
G
G
U C CCU
G
Zelo
tes
sp
ARM
1
95 (8
0-10
7)10
15
(101
-102
)
Am
auro
biid
ae
103
5 (9
9-11
8)
Out
grou
ps
100
2 (9
0-10
6)10
04
(93-
104)
997
(97-
102)
101
1 (1
00-1
02)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig2
Groups1
and2thesecondarystructuresofthelsquoA
rm1rsquooftheD2region
of28
SrRNAfrom
mem
bersofthelsquoaqu
aticcladersquothe
amaurobiidcladeandthreeou
tgroup
taxa
from
thecombined-
datatreeA
rm1ishigh
lyvariableevenwith
ingeneraasdemon
stratedby
thetwospeciesof
Paratheum
aGroups3nad4Arm
1sfrom
mem
bersof
theAgelenidaeandCoelotin
ae(G
roup
3)and
thecybaeidsensulatoclade(G
roup
4)Ingeneralthelatterappearmorestructurally
conservedthan
theform
erG
roups5and6
Arm
1sfrom
mem
bersof
theLathys(D
esidae)andAustralclade
spiders(row5)andforL
ycosoids
andDiony
chans(Group
6)O
fthe
form
erthe
genu
sDesishasa
particularlyconservedsecondarystructureforthisfeatureThe
armsshowninthesecondarystructure
figureswerechosen
torepresentthe
rangeof
variationBoxes
totherighto
feach
grouprepresentm
eanlength
andminim
um-m
axim
umrange(inparentheses)to
capturefullvariationin
figu
red
andunfigu
redarms
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 243
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
GC
G
GC
UU
A
U
C
AG
G
G
G
G
G
G
U
GC
C
A
GC
A
C
A
A
A
AU
U
G
C
UU
A
CU
C
C
CU
C
CC
C A
C
U
C
U
G
G
G
C
C
U
G
U
C
C
G
GGG
U
U
U
U
U
U
G
U
U
G
G
U
G
G
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
5
G C C
G G A
U
GG
C
C
G
U
G
C
G
C
C
C
G
C
A
C
U
U
G
U
G
C
GC
CG
G
G
G
G
G
C
G
A
C
G
C
A
GA
G
CC
G
C
G
G
C
C
C
G
G
C
G
C
C
U
U
C
A
A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
C
C
G
U
C
G
C
A
UU
A
A
C
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
UCU
CCC
GU
CCC
G A GG
ACG C U G C A
CA
CGGUC
UG
GU A U G G U AAA C G
AC
UC
UC
GG
A U GG A G A
AGUCGACA
GA C C
GUGGUGAC
G C C GG
CCC
GA A
CG
GGG A G U C G
C
UGCAGC
U
U
G GC
G
C
G
C
G
A
C
C
GUGG
U
G
A
CG
CC
GG
CC
C G A A C GG
G Y A G U CG
C
3
ARM 1
ARM 2
ARM 3
ldquoCORErdquoldquoD3rdquo
Fig 1 Hypothesis of secondary structure for the D2 and D3 regions and the core region connecting them for Saltonia incerta showing relative positions ofthe Core D3 and Arms 1ndash3 in the 50 (lower left) to 30 (upper right) Arm and core orientations are maintained in Figs 2ndash5 for comparison
242 Invertebrate Systematics J C Spagna et al
G U C
C A GU
A
A
G
G G GG
G
GGG
GG
G
G
GG
G
A
A
A
A
A
A
A
AA
AU
U
U
UU U
C
C
C
CC
C
C
C CCC CCC
C
CC
U
G G
G
G
G G
C
C C CCC C
G CG G G
C
CC
G
G
GU U UU
GG
G
U
U
U
A
A A
U
A
AA
G
G G
A
A
Arg
yron
eta
aqua
tica
G U C
C A GU
A
A
GG
GGGGG
GG
GG
GGG
G
GGG
GG
A A
A
A
AA
A
AAAA
AAA
AA
AU
U
U
U
U
U
U UU
UU
UU
U
U U
C
CC
C
C C C C
C
C C C CC
C
C CC
CC
C
C
C
C
U
G
G
G
G
G
G GG
G
G
G
C
C C
CC
C
CC C
Dic
tyna
sp
G U C
C A GU
A
A
G
G G
G
G
GGG
G
G
G
G
G
GGG
A A
A
A
AA
A
A
A
AA
A
AU U
U
U
UU
U U
CCC
C
C
C
C
C
C C
CC
C C
CCC
C
C
U
G G
G
G G G
C
C C CC
C C
G
C G G G
C
C
C
G
G
GU
U
U
U
GGG
U
UA A
UU
Par
athe
uma
arm
ata
G U C
C A GU
A
A
G
G
G
GGG
G
GG
G
GG
GG
G
GCUG
A A
A
A
AAA
A
A
A
A
AA
AU U
U
U
U
U
U U
C
C
C
C
C C
C
CC
C
C
C C
CCC
CC
C
C
U
G G
G
G
G GG
GG
G
C
C C
CC
C
C
C
G
C GG
G C CC
G
Para
theu
ma
insu
lana
G C G
G C
UU
A
U
C AG
G
G
G
G
G
GU
GCC
A GC
A
C
A
AAAU UG
C
UU
ACU
C
C
C U CCCC
A
C
U
C
U
G
G
G
C
C U GU
C
C
G G G G U
U
U
U
U
U
G
U
U
G
GU
G
G
Sal
toni
a in
certa
Aqu
atic
Cla
de
G U C
C A G
A
G G
U
UG C
U
A
GG
AU
G
UU
A
G
CCCC
U
A
C
U
G
G
G G
C
A
CG C
G
C
G
C
G
G
C
A
U
A
A
G
G
CC
A
U
U
G
G
C G GC CG C G
GU C
U
G
C
CG
G
C
U
A
A
C
G
U
A
GG
G
U
C
CC
UC
A
C
AGUUA
Am
aruo
bius
sim
ilis
Am
auro
biid
ae a
nd O
utgr
oups
G U U
C
A
C
GU
G
GC
U
GCCC
GU
U
A
G
G
G
A
G
A
A
A
U
A
A
G
A
C
AU
CC
U U
C
GU
C
G
A
UC
CC
GC
C
U
G
C
C
G
C
U
U
G
C C
CC
GG
C
CG
G
G
U
G
AA
GG
U
G
G
AG A
AC
C
G
GC
AU
AC
UU
U
Ste
gody
phus
sp
G U C
C A GG
C
A
G
G
A
U
GC
G
UC CAA
A
A
G
A
A
U
AU
U
U
C
U U
AU
U
A
CCCC C
G
G
G
C
U
G
G
C
G G
C
U
U
C
GU
A
G G G
U
U
C
A
A
U
G
G
G
G
GA
U
U
U
C
G
G
U
A
U
C
C C
C
C
G
C
G
G
AC
UG
UG
C
Meg
adic
tyna
thile
nii
G U C
C A GG
C
A
AA
G
G
GG
GG
GGC
G
GGGG
G
A
A
G
AG
A
AUAA
A
AC
AA
AU
U
UU
G
U
U
UUU
UA
C
G
U U
CC
CG
C U C C
G
U CC
CC
U
GC
U
C
C
U
G C
G
G
G CU
C
G
A
C G
C
C
C
G C
GCG U
Zygi
ella
x-n
otat
a5
3
G UC G C
U
A
G G
U
G CG
A
GGC
A
AU
U U
A
C
CC
G
G
C
U
G
G
C
G
GC
U
G
U
G
A
G
G
U
G
C
G
C
G
C
G
C
A
C
C
C
G
G
G
C
A
GGC
C
G
GA
G
A
C
C
U
CG
CA
A
U UGAA
A
CG
G
A
AAC
C
CC
CC
GG
G G
G
CC
Cry
phoe
ca s
p
G UC A G
C
A
G C
U
G
GC CG
A
CG
C
A
AU
G
U U
A
G
C
GG
G
G
C
U
G
A
C
C CG
G
C
G
U
A
A
G
GA
U
C
U
G
C
G
C
G
U
C
G
C
A
C
C
G
G
G
C
C
G
G
C G
C
UGC
G CA
G
G
AG
C
C
C
C
UC
GC
C
GA
U GCAG
A
CG
Bla
bom
ma
sp
G UC A G
C
A
U C
C
G
UG GG
A
CG
G
A
AU
G
U U
A
G
A
GA
C
G
C
U
G A
C
C CG
A
G
G
C
U
U
G
UA
U
G
G
U
A
U
A
C
G
U
C
G
C
A
C
C
G
C
GC C
G
G
C G
C
CGU
G CA
G
G
AG
C
G U
C
C
C
UC
GC
CGG
G GCCG
G
Cal
ymm
aria
sp
G UC G C
C
A
G G
A
G C
A
G
GU
U U
A
C
CG
G
C
C
U
GC G
C
G
C
G
C
G
G
G
U
G
C
U
G
G G
C
A
C
C
C
G
G
G
C
A
GCC
C
G
GA
G
A
C U
CG
C
U
C
U
AG
A
CG
G
A
A
AC CC
CG
G
G G
G
AA
AA
Rot
haei
na s
p
G UC A G
C
A
G C
A
G
G CG
A
CG
C
AAU
C
U U
A
G
C
G
G
G
C
C
U
G
A
C
C CG
C
C
G
C
G
A
G
GG
U
U
G
C
A
G
G
U
C
G
C
A
C
C
G
G
G
C C
G
G
C G
C
GCG
U CA
G
G
GG
C
C
C
C
UCG
CC
GA
U GCAG
A
CG
G
Cyb
aeus
sp
2
Cyb
aeid
ae s
l C
lade
G U C
C A GU
C
A
G GG
A
U
GC
G
G
GG GGG
A
A
G
G A
UAU
U U
C
U U
U
C
C
C
C
C
C
CC
G
CC
C
C
C
U
G
G
G
G
G
C
C
C
CG
C
G
G
G G
G
C
C
C
A
A
A
U
A
U
U
G
G
C
U
U
U
A
C
A
G
A
A
A
G
G
C
A
G
G
U
C
GG
GC
U
C
53
Text
rix d
entic
ulat
a
Age
leni
nes
+ C
oelo
tines
G U C
C A GUA
G
G
GG
G
G
GGG
A
A
G
C
A
AU
U
G
U U
A
C
C
C CCCC
C
U
C
G
C
U
G G
G
G G
C
C
C
CG
C
C
G
CC
C
G
G
U
G
U
G
G
C
C
U
G
C
A
G
G
A
A
G
C
A
A
G
G
GG
C
U
GC C
G
G G
CG
UCA
CUCG
Tege
naria
dom
estic
a
G U C
C A
A
C G
U
G
UG CC
A
G
A
AU
G
U U
G
CA
CCC
A
CU
G
G
C
U
G
G
G G
C
C
G
CG G
C
A
C
C
G
T
G
GU
GC
U
G
A
C
A
U
C
CGU
C
U
C
U
GC G G
C
G
UC
CGCA
G
C
GC
C
GU
Wad
otes
dix
iens
is
G U C
C A GUA
G
C
AG
GG GGG
A
UG
C
A
AU
U
G
U U
A
G
C CCCC
CU
C
U
C
U
G
G G
GG
C
C
C
CG
C
C
G
UC
C
G
A
U
A
U
GC
C
C
U
G
G
G
G
G
A
U
G
C
A
A
G
C
GGC
UC
A CG
G G
CG
UC
A
GCCA U
Coe
lote
s te
rres
tris
G U C
C A GU
A
A
G GG
GG
GG
G
G
GG GGG
A
A
A
A
A
A
A
AU
U
U
U
U U
C
CC
CC
C
C
CC
C
CC
C
C
C
U
G
G
G
G
G
C
C
C
CC
C
G
C
G G
G
C
C
C
G
G
G
U
U
U
UG
G
G
UU
U
A
C
A
U
A
A
A
G
G
A
A
UU
U
C
Alla
gele
na g
raci
lens
G U C
C A GU
C
A
G
G
AG
GG GGG
A
UG
C
A
AU
U
U
G
U U
A
C
C
C CCCG
CU
C
U
C
U
G
G
G
G G
C
C
C
GG
G
C
G
UC
C
G
A
U
A
U
G
G
C
U
U
G
C
A
G
G
A
C
G
C A
A
G
C
GG
C
UUC C
C
G G
CG
UC
Age
leno
psis
ape
rta
Lyco
soid
s +
Dio
nych
ans
G U C
C A GUA
C
CU
A
A
CC
G
G
G
A
U
C
A
A
AU
C
UG
U U
G
G
G
G
CCC C
A CG
C
U
G U
U G
C
C
C
U
GG
A
G
C G
G
C
A
A
GG
G
G
C
A
G
U
U
C
C
C
UA
A
G C
G
G
CG
G
U
G
C
U
G
G
G CG G
G A
CC
A AA
UU
U
Lath
ys a
lber
ta
Dic
tyno
ids
+ D
esis
G U C
C A G
A
G G
UUG CU
A
GG
AU
G
U U
A
G
CCCC
G
G
C
U
G
G
G C
C
A
C
CG U
G
C
G
C
G
A
C
A
U
A
A
G
G
A
C
C
A
A
T
G
G
C G GC
G
CG C
G
GC
C
C
G
A
C
UG G
C
C U
A
UC
G
A
A
GG
G
CU
U
GG
Met
alte
lla s
imon
i
G U C
G
A
C
G
G
G
GGC
A
GGGC
C
U
C
G
G
G
G
A
C
A
A
A
A
A
A
G
C
A
AAU G
UG
U U
C
CC
C
C
C C
G
C
CC
GCG
C
U
G
G
G
G
C
C U CCC C
CG CG
U
C
G
G
G
A
A
AG
U
U
A
U
G
C
AA
A
GC
GC
G
GC C
GUA
AA
C
Neo
ram
ia ja
nus
G U C
C A G
A
G G
AUG
C
G
A
G
C
AU
U U
G
GGG
A
A
GG
G
C
U
G
C
C C
C
C CG U
G
A
C
A
G
A
C
UC
U
G
A
U
C
U
C
A
U
U
C
A
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G
G
C
C
A
C
G
A
A
GG
G
U
G GC
C
C
C
UC
Des
is m
arin
a
G U C
C A G
A
G G
UUG CU
A
G
G
AU
A
U U
A
G
GGG
A
A
UG
G
C
U
G
C
C C
C
A CG C
G
C
G
C
G
A
C
UC
U
A
A
G
G
A
U
A
U
C
C
G
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G G
C
C
A
U
C
G
U
A
G GG
U
C G
G
C
Bad
umna
long
iqua
G U C
C A G
CC
A
G
G U
GG
G
G
G
G
C
GG
GC
G
G
GGG
A
A
A
A
A
A
A
A
A AAAAU
U
U
A
UUU U
CC U
C
C
C
U
C
UC
A G
C C
C C
C
U
C
U
G
G
G
G G
C
G
C
C
C
U
C
G
G
C UC G
C
U
U
U U UU
U
UU
A
U
Teng
ella
radi
ata
G U C
C A GU
A
A
A
CU
AA
UG G
G
G
A
U
C
A
A
AU UC
U
U U
G
A
A
C
GGG G
C
C
C
C
U
G G
UG C
C
C
U
GG U
G
U G
GC
A
A
G
GG
C
C
G
U
U
U
C
G
C
U A
C
G C
G
C
C
G
G
UG
CU
C
G
GC
C C GAGA
Alo
peco
sa k
ochi
G U C
C A GU
A
A
G
G
U
GCG
GG GGG
A
G
G
G
G
C
AU
C
AG
U U
A
C
C
C GGGC
GA
U
U
C
U
G
G
GC
C
C
UUG
U
G C
GU
C
C
G
U
U
G
U
C
G
C
U
A
C
A
C
A
G
G
C G
U
G
U
CC
G
U
C
CU
U
GC
UC
UGC
Any
phae
na c
alifo
rnic
a
G U C
G
UA
GG
GG
G
G
GG
G
GG
G
G
G
G
G
A
A
A
A
A
A
A
A
A
AA
A
AU
U
UU
U U
C
C
C
C
C
C C
C
C
CC
C
C
C
C
U
G
G
GG
G G
C
C C CCC C
G
CG GG
C
CC G
G
G
U
U
G
GGU
U
A
U
U
C
AA A
A A
G
G C
G
Stip
hidi
on fa
cetu
m
G U C
C A
A
C
G
U
G
C
U
A
C
G
AU
U
UU
A
G
UC
U
U
C
U
G
U
C
CG G
G
C
G
C
G
C
A
A
U
A
U
G
G
A
G
A
G
C
C
G G G
GCCC
C
GG
G
A
A
C
C
A
A
U
U
GC
C
C G
G
AU
C
C
AGU
C
G
G
G
U C CCU
G
Zelo
tes
sp
ARM
1
95 (8
0-10
7)10
15
(101
-102
)
Am
auro
biid
ae
103
5 (9
9-11
8)
Out
grou
ps
100
2 (9
0-10
6)10
04
(93-
104)
997
(97-
102)
101
1 (1
00-1
02)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig2
Groups1
and2thesecondarystructuresofthelsquoA
rm1rsquooftheD2region
of28
SrRNAfrom
mem
bersofthelsquoaqu
aticcladersquothe
amaurobiidcladeandthreeou
tgroup
taxa
from
thecombined-
datatreeA
rm1ishigh
lyvariableevenwith
ingeneraasdemon
stratedby
thetwospeciesof
Paratheum
aGroups3nad4Arm
1sfrom
mem
bersof
theAgelenidaeandCoelotin
ae(G
roup
3)and
thecybaeidsensulatoclade(G
roup
4)Ingeneralthelatterappearmorestructurally
conservedthan
theform
erG
roups5and6
Arm
1sfrom
mem
bersof
theLathys(D
esidae)andAustralclade
spiders(row5)andforL
ycosoids
andDiony
chans(Group
6)O
fthe
form
erthe
genu
sDesishasa
particularlyconservedsecondarystructureforthisfeatureThe
armsshowninthesecondarystructure
figureswerechosen
torepresentthe
rangeof
variationBoxes
totherighto
feach
grouprepresentm
eanlength
andminim
um-m
axim
umrange(inparentheses)to
capturefullvariationin
figu
red
andunfigu
redarms
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 243
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
G U C
C A GU
A
A
G
G G GG
G
GGG
GG
G
G
GG
G
A
A
A
A
A
A
A
AA
AU
U
U
UU U
C
C
C
CC
C
C
C CCC CCC
C
CC
U
G G
G
G
G G
C
C C CCC C
G CG G G
C
CC
G
G
GU U UU
GG
G
U
U
U
A
A A
U
A
AA
G
G G
A
A
Arg
yron
eta
aqua
tica
G U C
C A GU
A
A
GG
GGGGG
GG
GG
GGG
G
GGG
GG
A A
A
A
AA
A
AAAA
AAA
AA
AU
U
U
U
U
U
U UU
UU
UU
U
U U
C
CC
C
C C C C
C
C C C CC
C
C CC
CC
C
C
C
C
U
G
G
G
G
G
G GG
G
G
G
C
C C
CC
C
CC C
Dic
tyna
sp
G U C
C A GU
A
A
G
G G
G
G
GGG
G
G
G
G
G
GGG
A A
A
A
AA
A
A
A
AA
A
AU U
U
U
UU
U U
CCC
C
C
C
C
C
C C
CC
C C
CCC
C
C
U
G G
G
G G G
C
C C CC
C C
G
C G G G
C
C
C
G
G
GU
U
U
U
GGG
U
UA A
UU
Par
athe
uma
arm
ata
G U C
C A GU
A
A
G
G
G
GGG
G
GG
G
GG
GG
G
GCUG
A A
A
A
AAA
A
A
A
A
AA
AU U
U
U
U
U
U U
C
C
C
C
C C
C
CC
C
C
C C
CCC
CC
C
C
U
G G
G
G
G GG
GG
G
C
C C
CC
C
C
C
G
C GG
G C CC
G
Para
theu
ma
insu
lana
G C G
G C
UU
A
U
C AG
G
G
G
G
G
GU
GCC
A GC
A
C
A
AAAU UG
C
UU
ACU
C
C
C U CCCC
A
C
U
C
U
G
G
G
C
C U GU
C
C
G G G G U
U
U
U
U
U
G
U
U
G
GU
G
G
Sal
toni
a in
certa
Aqu
atic
Cla
de
G U C
C A G
A
G G
U
UG C
U
A
GG
AU
G
UU
A
G
CCCC
U
A
C
U
G
G
G G
C
A
CG C
G
C
G
C
G
G
C
A
U
A
A
G
G
CC
A
U
U
G
G
C G GC CG C G
GU C
U
G
C
CG
G
C
U
A
A
C
G
U
A
GG
G
U
C
CC
UC
A
C
AGUUA
Am
aruo
bius
sim
ilis
Am
auro
biid
ae a
nd O
utgr
oups
G U U
C
A
C
GU
G
GC
U
GCCC
GU
U
A
G
G
G
A
G
A
A
A
U
A
A
G
A
C
AU
CC
U U
C
GU
C
G
A
UC
CC
GC
C
U
G
C
C
G
C
U
U
G
C C
CC
GG
C
CG
G
G
U
G
AA
GG
U
G
G
AG A
AC
C
G
GC
AU
AC
UU
U
Ste
gody
phus
sp
G U C
C A GG
C
A
G
G
A
U
GC
G
UC CAA
A
A
G
A
A
U
AU
U
U
C
U U
AU
U
A
CCCC C
G
G
G
C
U
G
G
C
G G
C
U
U
C
GU
A
G G G
U
U
C
A
A
U
G
G
G
G
GA
U
U
U
C
G
G
U
A
U
C
C C
C
C
G
C
G
G
AC
UG
UG
C
Meg
adic
tyna
thile
nii
G U C
C A GG
C
A
AA
G
G
GG
GG
GGC
G
GGGG
G
A
A
G
AG
A
AUAA
A
AC
AA
AU
U
UU
G
U
U
UUU
UA
C
G
U U
CC
CG
C U C C
G
U CC
CC
U
GC
U
C
C
U
G C
G
G
G CU
C
G
A
C G
C
C
C
G C
GCG U
Zygi
ella
x-n
otat
a5
3
G UC G C
U
A
G G
U
G CG
A
GGC
A
AU
U U
A
C
CC
G
G
C
U
G
G
C
G
GC
U
G
U
G
A
G
G
U
G
C
G
C
G
C
G
C
A
C
C
C
G
G
G
C
A
GGC
C
G
GA
G
A
C
C
U
CG
CA
A
U UGAA
A
CG
G
A
AAC
C
CC
CC
GG
G G
G
CC
Cry
phoe
ca s
p
G UC A G
C
A
G C
U
G
GC CG
A
CG
C
A
AU
G
U U
A
G
C
GG
G
G
C
U
G
A
C
C CG
G
C
G
U
A
A
G
GA
U
C
U
G
C
G
C
G
U
C
G
C
A
C
C
G
G
G
C
C
G
G
C G
C
UGC
G CA
G
G
AG
C
C
C
C
UC
GC
C
GA
U GCAG
A
CG
Bla
bom
ma
sp
G UC A G
C
A
U C
C
G
UG GG
A
CG
G
A
AU
G
U U
A
G
A
GA
C
G
C
U
G A
C
C CG
A
G
G
C
U
U
G
UA
U
G
G
U
A
U
A
C
G
U
C
G
C
A
C
C
G
C
GC C
G
G
C G
C
CGU
G CA
G
G
AG
C
G U
C
C
C
UC
GC
CGG
G GCCG
G
Cal
ymm
aria
sp
G UC G C
C
A
G G
A
G C
A
G
GU
U U
A
C
CG
G
C
C
U
GC G
C
G
C
G
C
G
G
G
U
G
C
U
G
G G
C
A
C
C
C
G
G
G
C
A
GCC
C
G
GA
G
A
C U
CG
C
U
C
U
AG
A
CG
G
A
A
AC CC
CG
G
G G
G
AA
AA
Rot
haei
na s
p
G UC A G
C
A
G C
A
G
G CG
A
CG
C
AAU
C
U U
A
G
C
G
G
G
C
C
U
G
A
C
C CG
C
C
G
C
G
A
G
GG
U
U
G
C
A
G
G
U
C
G
C
A
C
C
G
G
G
C C
G
G
C G
C
GCG
U CA
G
G
GG
C
C
C
C
UCG
CC
GA
U GCAG
A
CG
G
Cyb
aeus
sp
2
Cyb
aeid
ae s
l C
lade
G U C
C A GU
C
A
G GG
A
U
GC
G
G
GG GGG
A
A
G
G A
UAU
U U
C
U U
U
C
C
C
C
C
C
CC
G
CC
C
C
C
U
G
G
G
G
G
C
C
C
CG
C
G
G
G G
G
C
C
C
A
A
A
U
A
U
U
G
G
C
U
U
U
A
C
A
G
A
A
A
G
G
C
A
G
G
U
C
GG
GC
U
C
53
Text
rix d
entic
ulat
a
Age
leni
nes
+ C
oelo
tines
G U C
C A GUA
G
G
GG
G
G
GGG
A
A
G
C
A
AU
U
G
U U
A
C
C
C CCCC
C
U
C
G
C
U
G G
G
G G
C
C
C
CG
C
C
G
CC
C
G
G
U
G
U
G
G
C
C
U
G
C
A
G
G
A
A
G
C
A
A
G
G
GG
C
U
GC C
G
G G
CG
UCA
CUCG
Tege
naria
dom
estic
a
G U C
C A
A
C G
U
G
UG CC
A
G
A
AU
G
U U
G
CA
CCC
A
CU
G
G
C
U
G
G
G G
C
C
G
CG G
C
A
C
C
G
T
G
GU
GC
U
G
A
C
A
U
C
CGU
C
U
C
U
GC G G
C
G
UC
CGCA
G
C
GC
C
GU
Wad
otes
dix
iens
is
G U C
C A GUA
G
C
AG
GG GGG
A
UG
C
A
AU
U
G
U U
A
G
C CCCC
CU
C
U
C
U
G
G G
GG
C
C
C
CG
C
C
G
UC
C
G
A
U
A
U
GC
C
C
U
G
G
G
G
G
A
U
G
C
A
A
G
C
GGC
UC
A CG
G G
CG
UC
A
GCCA U
Coe
lote
s te
rres
tris
G U C
C A GU
A
A
G GG
GG
GG
G
G
GG GGG
A
A
A
A
A
A
A
AU
U
U
U
U U
C
CC
CC
C
C
CC
C
CC
C
C
C
U
G
G
G
G
G
C
C
C
CC
C
G
C
G G
G
C
C
C
G
G
G
U
U
U
UG
G
G
UU
U
A
C
A
U
A
A
A
G
G
A
A
UU
U
C
Alla
gele
na g
raci
lens
G U C
C A GU
C
A
G
G
AG
GG GGG
A
UG
C
A
AU
U
U
G
U U
A
C
C
C CCCG
CU
C
U
C
U
G
G
G
G G
C
C
C
GG
G
C
G
UC
C
G
A
U
A
U
G
G
C
U
U
G
C
A
G
G
A
C
G
C A
A
G
C
GG
C
UUC C
C
G G
CG
UC
Age
leno
psis
ape
rta
Lyco
soid
s +
Dio
nych
ans
G U C
C A GUA
C
CU
A
A
CC
G
G
G
A
U
C
A
A
AU
C
UG
U U
G
G
G
G
CCC C
A CG
C
U
G U
U G
C
C
C
U
GG
A
G
C G
G
C
A
A
GG
G
G
C
A
G
U
U
C
C
C
UA
A
G C
G
G
CG
G
U
G
C
U
G
G
G CG G
G A
CC
A AA
UU
U
Lath
ys a
lber
ta
Dic
tyno
ids
+ D
esis
G U C
C A G
A
G G
UUG CU
A
GG
AU
G
U U
A
G
CCCC
G
G
C
U
G
G
G C
C
A
C
CG U
G
C
G
C
G
A
C
A
U
A
A
G
G
A
C
C
A
A
T
G
G
C G GC
G
CG C
G
GC
C
C
G
A
C
UG G
C
C U
A
UC
G
A
A
GG
G
CU
U
GG
Met
alte
lla s
imon
i
G U C
G
A
C
G
G
G
GGC
A
GGGC
C
U
C
G
G
G
G
A
C
A
A
A
A
A
A
G
C
A
AAU G
UG
U U
C
CC
C
C
C C
G
C
CC
GCG
C
U
G
G
G
G
C
C U CCC C
CG CG
U
C
G
G
G
A
A
AG
U
U
A
U
G
C
AA
A
GC
GC
G
GC C
GUA
AA
C
Neo
ram
ia ja
nus
G U C
C A G
A
G G
AUG
C
G
A
G
C
AU
U U
G
GGG
A
A
GG
G
C
U
G
C
C C
C
C CG U
G
A
C
A
G
A
C
UC
U
G
A
U
C
U
C
A
U
U
C
A
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G
G
C
C
A
C
G
A
A
GG
G
U
G GC
C
C
C
UC
Des
is m
arin
a
G U C
C A G
A
G G
UUG CU
A
G
G
AU
A
U U
A
G
GGG
A
A
UG
G
C
U
G
C
C C
C
A CG C
G
C
G
C
G
A
C
UC
U
A
A
G
G
A
U
A
U
C
C
G
G
C U CC
A
CG C
A
G
GC C
CG
A
C
U
G G
C
C
A
U
C
G
U
A
G GG
U
C G
G
C
Bad
umna
long
iqua
G U C
C A G
CC
A
G
G U
GG
G
G
G
G
C
GG
GC
G
G
GGG
A
A
A
A
A
A
A
A
A AAAAU
U
U
A
UUU U
CC U
C
C
C
U
C
UC
A G
C C
C C
C
U
C
U
G
G
G
G G
C
G
C
C
C
U
C
G
G
C UC G
C
U
U
U U UU
U
UU
A
U
Teng
ella
radi
ata
G U C
C A GU
A
A
A
CU
AA
UG G
G
G
A
U
C
A
A
AU UC
U
U U
G
A
A
C
GGG G
C
C
C
C
U
G G
UG C
C
C
U
GG U
G
U G
GC
A
A
G
GG
C
C
G
U
U
U
C
G
C
U A
C
G C
G
C
C
G
G
UG
CU
C
G
GC
C C GAGA
Alo
peco
sa k
ochi
G U C
C A GU
A
A
G
G
U
GCG
GG GGG
A
G
G
G
G
C
AU
C
AG
U U
A
C
C
C GGGC
GA
U
U
C
U
G
G
GC
C
C
UUG
U
G C
GU
C
C
G
U
U
G
U
C
G
C
U
A
C
A
C
A
G
G
C G
U
G
U
CC
G
U
C
CU
U
GC
UC
UGC
Any
phae
na c
alifo
rnic
a
G U C
G
UA
GG
GG
G
G
GG
G
GG
G
G
G
G
G
A
A
A
A
A
A
A
A
A
AA
A
AU
U
UU
U U
C
C
C
C
C
C C
C
C
CC
C
C
C
C
U
G
G
GG
G G
C
C C CCC C
G
CG GG
C
CC G
G
G
U
U
G
GGU
U
A
U
U
C
AA A
A A
G
G C
G
Stip
hidi
on fa
cetu
m
G U C
C A
A
C
G
U
G
C
U
A
C
G
AU
U
UU
A
G
UC
U
U
C
U
G
U
C
CG G
G
C
G
C
G
C
A
A
U
A
U
G
G
A
G
A
G
C
C
G G G
GCCC
C
GG
G
A
A
C
C
A
A
U
U
GC
C
C G
G
AU
C
C
AGU
C
G
G
G
U C CCU
G
Zelo
tes
sp
ARM
1
95 (8
0-10
7)10
15
(101
-102
)
Am
auro
biid
ae
103
5 (9
9-11
8)
Out
grou
ps
100
2 (9
0-10
6)10
04
(93-
104)
997
(97-
102)
101
1 (1
00-1
02)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig2
Groups1
and2thesecondarystructuresofthelsquoA
rm1rsquooftheD2region
of28
SrRNAfrom
mem
bersofthelsquoaqu
aticcladersquothe
amaurobiidcladeandthreeou
tgroup
taxa
from
thecombined-
datatreeA
rm1ishigh
lyvariableevenwith
ingeneraasdemon
stratedby
thetwospeciesof
Paratheum
aGroups3nad4Arm
1sfrom
mem
bersof
theAgelenidaeandCoelotin
ae(G
roup
3)and
thecybaeidsensulatoclade(G
roup
4)Ingeneralthelatterappearmorestructurally
conservedthan
theform
erG
roups5and6
Arm
1sfrom
mem
bersof
theLathys(D
esidae)andAustralclade
spiders(row5)andforL
ycosoids
andDiony
chans(Group
6)O
fthe
form
erthe
genu
sDesishasa
particularlyconservedsecondarystructureforthisfeatureThe
armsshowninthesecondarystructure
figureswerechosen
torepresentthe
rangeof
variationBoxes
totherighto
feach
grouprepresentm
eanlength
andminim
um-m
axim
umrange(inparentheses)to
capturefullvariationin
figu
red
andunfigu
redarms
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 243
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Aquatic Clade
Amaurobiidae and Outgroups
GG
C
C
U
G
G
G
G
G
U
G
A
C
C
C
U
G
G
CG
U
CG
A
C
U
G
GG
G
U
C
G
G
G
C
C
A
G
U
GU
U
GC
GG C
C
U
G
C
C
U
A
C
U
A
G
C
G
A
G
U
U
5 3
Argyroneta aquatica
GG
C
G
U
G
G
G
C
C
G
G
C
U
G
A
U
C
C
U
C
G
C
C
A
C
GC
A
C
U
G
GG
G
U
G
U
C
C
A
C
G
U
C
U
C
C G
U
GU
C
GC
G
A
GGG C
C
U
G
C
C
A
A A
G
C
G
U
UG
AG
CC
C
GG
Paratheuma armata
U
C
C
G
G
U
C
A
G
G
C
CC
G
GA
G
U
C
C
U
U
C
A
AU
C
G
G
GG
AU
C
U
U
G
AG
G
G
U
C
U
G
G
G
G
C
G U
G
C
G
U
UG
CGC
GGC
U A
A
C
A
G
Saltonia incerta
GG
C
G
G
GG
G
C
G
A
G
C
U
G
A
C
C
C
U
C
C
C
C
G
U
G
U
C
C
U
GC
GG
U
U
G
G
C
C
G
CGU
C
U
C
A
G
G
U
GU
G
GC
C
A
GGG C
C
UG
C
C
Dictyna sp
C
G
U
G
G
U
A
A
G
U
G
U
CC
G
G
G
G
U
C
A
G
A
C
G
U
GU
G
GA
CG
AC
C
C
U
G
A
G
G
G
C
C
C
U
C
G
C
C
G C
A
A
G
G U
A
Amaurobius similis
G C
U
G
C
C
G
U
A
A
C
C
U
C
G
C
CG C
G
GG
G
U
C
G
G
G
U
C
A
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
G
C
C
C
G
C
UG
U
A U
U
G
A U
G C
Megadictyna thilenii
Cybaeidae sl Agelenines + Coelotines
G
G
U
CG
G
G
G
G
G
U
A
A
U
C
U
G
CCG
U
CG
A
G
U
G
G
G
G
U
C
G
G
G
U
C
A G
U
GU
C
G
C
GC G
C
U
G
C
C
C
G
A
U
A
G
U
G
A
C
U
U
Textrix denticulata
G C
G
G
A
C
C
U
G
A
C
C
U
A
C
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
G
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
G
C
U
C
A
C
A
G
A
G
Calymmaria sp
G
C
U
C
U
G
A
C
C
U
C
CG
C
C
G
G
G
G
G
U
C
G
G
G
G
C
U
GU
G
GG
CG
C
C
G
C
C
G
G
A
G
G
G
A
G
U
C
C
G
C
GC
A
A
C
C
C
Cybaeota sp1
Lycosoids + Dionychans Dictynoids + Desis
GG
C
G
U
C
U
G
A
G
U
U
U
C
UG
U
CG
C
G
G
G
C
G
UC
G
G
C
G
U
G
U
GU
G
GG
CC
G
C
U
C
C
G
G
A
G
G
G
G
C
C
G
U
A
C
U
A
A
A
U
C
U
Tengella radiata
U
A
U
A
U
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
U
C
G
U
GU
G
GG
CG
G
C
G
C
C
C
C
A
G
C
G
G
A
U
U
C
G
UG
U
G C
U
A
A U
G C
G U
Alopecosa kochi
C
G
U
G
G
U
A
A
U
C
U
G
C
CG C
G
G
G
G
U
C
U
G
G
C
C
G
U
GU
G
GG
CG
U
C
G
C
C
G
C
A
G
C
G
A
A
G
U
C
G
U
G
U
A U
U
A U
G C
G C
Lathys alberta
G C
G
G
U
G
C
U
G
A
C
C
U
C
C
U
G
U
UG
G C
G
G
G
G
U
C
G
G
G
C
C
G
U
GU
A
GG
CG
CC
G
C
C
G
G
C
A
G
C
G
A
G
U
UA
U
G
U
U
G
U
Metaltella simoni
G C
G
G
U
G
C
U
A
A
C
C
U
CA
U
G
U
CG
G C
G
G
G
G
U
C
G
G
G
U
C
G
U
GU
A
GG
CG
C
C
G
C
C
G
G
C
A
G
C
G
C
C
U
A
U
A
G
U
G
U
Desis marina
ARM 2
703 (67-81)
61 (58-61)
Amaurobiidae
658 (64-70)
Outgroups627 (59-63) 612 (57-63)
642 (61-67) 63 (62-65)
5 3 5 3
5 3 5 3 5 35 3 5 3
5 3
5 3 5 3 5 3 5 3 5 3
Fig 3 Groups 1 and 2 the secondary structures of the lsquosecond armrsquo of the D2 region of 28S rRNA from members of the lsquoaquatic cladersquo+Dictyna(Group 1) an Amaurobiid an outgroup taxon a representative from the Agelenidae and theCybaeidae slClade (Group 2) Groups 3 and 4 Arm 2s fromLycosoids Dictynoids and Desis There is little length variation in this region overall Arm 2 secondary structure appears to be quite conserved acrosstaxa with most variation seen in the presence of a single loop structure (marked with an asterisk) in some taxa Boxes as in Fig 2
244 Invertebrate Systematics J C Spagna et al
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Aqu
atic
Cla
deA
mau
robi
idae
and
Out
grou
ps
G U C
C C GU
G
G
A
UGC
G
C
G
C
G AGG
G
A
C
U
U
U
A
C
C
G
G
G
A
G
C
AG
AA
G
G
C
U
C
G
C
G
A
C
U
G
G G
U C
CG
C
C
A
C
GG
G UCG G
CG C
G
U
A
C
C
G G
G
G C
C
C G G
C
C
GG
G
C
U
AC
G
A
AC
CC
G
G
G
C
A
U C
G
C
G
G U
C C
CGGG
C
C CCU
A
G
GC
Arg
yron
eta
aqua
tica
C C U
C G GC
A
A
G
G
UGG
G
CG
C
G
G
G
GC
U
C
U
C
A
A
A
G
A
G
A
A
G
C
A
G
A
U
A
C
G
G
U
C
G
CG
A
C
C CCU
C
G
C
G
GA
C
A
G
CC
A
C
G
G
C
UC
U
G
G
C
G
C
C
C
U
C
A
G
G
G
G
G
CG
C
G GC
C C
G
G G
GC
C
GG
CC
G
G CU
A
C G
C
G
AG
C CG
C
U
C
G
Par
athe
uma
arm
ata
G C C
C G GU
A
A
G
G U
G
G
G
G GG
C
G
G
GC
G
G
C
U
C
A
A
A
A
A
A
A A
A
A
A
A
A
A
U
U
U
G U
U
CG
CG
C
C
C CC
U
C
G C
A G
A
C
C
C
G
CC
U
G
GG
G
C
GG
G
G
G
C
G C
C
C
C
C
A
G
C
G
G
G
G
CC
C
G G G
C C AA
G GCG
C
G
C
G
GG
C CC
G
C G
C
G
AU
C CG
C
G
CG
C
Par
athe
uma
insu
lana
G C CG G A
U
GG C
C
GU
G
C
G
C
C
C
GC
A
C
U
U
G UGC
GC
CG G G
G
G
G
C
G
A
C
G
C
A
G A
GCC G
CG
G
C
C
C G G
C
GC
C
U
U
C
A A
G
U
C
C
G
A
G
U
G
A
G
G
G
G
G
G
A
A
G
U
C
A
A
GC
C
G
G
A
C
C
U
C
U
C
G
G
G
CC
G
U
C
G
C
A
UUA
A
C
Sal
toni
a in
certa
G C C
U G GU
G
G
A
C
U
GG
C
GC G
A
C U
C
C
A
A
G
G
G
G
U
A
G
U
G
A
UG
U
G
C
G
A
C
U
CG CG
G
U G
G
U
A
C
G
G
U
C
AC G
G
GC
UA
G C
A U
C
U
C
G
C CU
C
C
G
CU G
U
G
C G
G
C
GC
C
U
CC
C C
G
GG
C C G
A
A
C G U
U
A
C
G
G G
C G
G
A
G
GA
A
CCA
Pim
us
G C C
U G UU
A
G
G
C
C
GG
CGC
C
A
C
U
C
C
C
C
GG
U
C
U
G
U
G
U
C
G
G
CG AG
GCGC
G
G
C
G
CC G G
U UGC
C
U
G
C
A
U
G
C
U
U
C
G A
GC
C
CU
C
U
G
G
U
G
U
A CU
C
U
C
C G
C G
G
G
A
G
C
G
G
U
GA
A UG
C
G
A
G GG
A U
A
CA
U
UG
A
U
GGG
CG
GU
CC
Meg
adic
tyna
thile
nii
G U C
C G UG
A UG
C
A
GU
GC
G C
C
C
G
C
A
CU
U
U
G
A
U
G
GC
CG GG
G
C
G
C
U
A
C
G
C
A
G C
U G
C
C
U
CG G
C
C C G
G
G
C
GC
C
U
UC
U
C
U
U
A
U
G
U
C
C
U
G
G
A
G
U G
A
G
GG
A UG G
A
A
A UC
U
A
A
CG
G
GG
A U
A
CU
C
UG
C
GCG
G G
G
CC
G
UU
CG
CA
UU
A
A
Ste
gody
phus
sp
G C C
C G UU
A
A
G
C
G
UG
C
GC C
A
C
G
C
C
C
A
GU
U
U U
C
C
C
G
U
U
G
A
CG CG
GCG
CC
U
C
C
A
CG C G U
U
G
C
G
G
G
C
A
U
G
C
U
U
GG
G A
GC
G
U
GC
G
A
C
C
G
C
A
U C G
A
U
C
U
U
C
G
A
G
C
GG
U
C
G
G
C
G
G
G
A A
U
CA
G
U G
CG
A
GG CA
A
A
U
U
U
A GA
U
C
Zygi
ella
x-n
otat
a
Cyb
aeid
ae s
l C
lade
Age
leni
nes
+ C
oelo
tines
G U C
C G GU
G
C
A
U
GU
GG
C
C
A
GC
G
A
C U
U
C
G
C
G
G
A
G
A
AC
GG
A
G
G
G
C
U
CG
G
C
U
U
GGU
CCG C
C
G
U G
C
G U
G
C
C
G
A
G
C
U
A
U
A
C G
C
U
GC
U
G
C
G
C
U
AC
G
A
A
C
C
C
G
G
G
CU
C
C
G
G
U U
A
C
GCGG
C CCG
U
GC
U
UUU
GG
G CU G GC
U
Text
rix d
entic
ulat
a
G U C
U C GU
A
G
G C
C
GG
C
GC
G
A
C
U
U
C
C
A
G
G
U A
C
G
G
C
C
U
U
G
C
G
G
CCG CG
G
C
G
CC
C
C
C C
GC
AG C G
C
U
C
U CG
C
C
CG
C
C
U
GU
G
C
C
G
A
GC
G
U
UC
G C
G
C
G
U
G
G
G G
G
G
A
C
C
G
U
G
GG
G
U GU
AG CA
A
G
G
CUG
G
U
Cry
phoe
ca s
p
G C C
C G GU
G
A
A
C
U
GG
C
GC
G
A
C
U
G
C
C
A
A
C
U A
C
G
G
C
C
G
U
G
C
G
G
CCG CG
G
C G
A
A
C C
C C
GC
G
G
G C G
C
U
C
C CG
C
U
CU
C
CC
GU
G
A
C
G
C
GC
G
A
CC
G
C G
G
C U
G
G
G
C
G
G
G
A
C
G
G
C
G
G
A
G
G
C G
U
A
U
G
G
G
GUUG
Bla
bom
ma
sp
Lyco
soid
s +
Dio
nych
ans
Dic
tyno
ids
+ D
esis
G U C
A C GU
A
A
G
CU
C
G
CGC G AC G
A
C
G
A
A
G
A
AU
A
G
C
G
C
CG
U
G
C
G A
G U
CG CG
GC G
C
A
C
G
A
C
AC G
A
CC
C G C
U
C
C
C
G C
C C C
G
C
U G
UG
A
C
G A
GC
G
C
UC
G GGC
C
G
G
U
C
C G
G
U
AC
G
C U
C
G
G
AA
C
A
CG
UAG
G
G
A A
Teng
ella
radi
ata
G U C
U G GU
G GA
A
U
GG
CGC
G
AC
A
C
C
C
A
G
U
C
UG
U
G
G
C
AU GG
GG
C
C
G
G
C
G
C
C G
UGC
C
C
G
U
A
U
G
CC
G
A
GC
C
U
UC
A
G
C
U
C
U
A
C
C
C
C G
C C
G
G
G
G
C
G
G
G
C
A
GGC
G
A
G CC
A
A
A
CG
G
A
UG
G
G
C
G
G U
C
C
A
G A G
Alo
peco
sa k
ochi
G U C
U C GU
A GG
C
C
GU
CGC
G C
C
C
A
C
A
UG
U
U
G
G
U
G
U
GC
CG GA
G
C
G
C
G
G
C
G
C
A
G U
U GCU
C
UG C
C
G CC
G
G C
GC
C
U
UC
U
G
G
U
U
ACUC
U
C
C G
C
G
G
G
A
G
C
G
U
C
GU
A
U
G C
G
A
G GC
C
A
A
UG U
AG
A
A
G
UG
C
G
GG
U
AG
GG
Lath
ys a
lber
ta
G U C
U C GU
A
A
G
U
CC
GG
C
GC G
A
C U
G
C
G
A
G
G
U
G
AU
A
G
U
G
U
UG
U
G
C
U G
C
A
C
CG CG
G
C
G
G
UU
C
G
G
U
C
AC G
A
GC
UC
G
C
U U
G
C
C
GC
C
U
C C
C
A
C G
UG
G
G
G
C
GC
U
C
CC
C C
C
GG
C C
G
U
G
C G A
C
A
CU
G
G G
GU
U
G
A
Met
alte
lla s
imon
i
G U C
U C GU
G
C
G
G
C
GG
C
GC
G
A
C
U
C
C
C
A
G
C
U
A
G
G
U
C
G
U
G
G
CCG UU
GC
G
C
U
C
C
C
C
A C
G C G
C
U
C
G C
G
C
U
C G
G
C
C U
U
AG
G C
GC
G
U
CC
G
C
C
G
G
G
G
G G C
G
U
C
G
C
U
G
G
G
C GU
G
GC
C
GG
C
U
A
A
U
C
CU
AA
G
U
G
G
Des
is m
arin
a
ARM
3
123
7 (1
17-1
33)
124
5 (1
22-1
27)
Am
auro
biid
ae
126
8 (1
16-1
34)
Out
grou
ps
121
5 (1
10-1
30)
124
1 (1
21-1
29)
122
6 (1
19-1
21)
123
3 (1
07-1
29)
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
53
Fig4
Groups1
and2thesecondarystructuresofthelsquothirdarmrsquoofthe
D2region
of28
SrRNAfrom
mem
bersoftheaquatic
clade(G
roup
1)the
amaurobiidsand
outgroup
taxa
(Group
2)G
roups3
and4Arm
3sfrom
arepresentativ
eof
theAgelenidaetheCybaeidae
slcladealycosoidsandDictyno
idsplus
DesisB
oxes
asin
Fig2
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 245
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
extreme microhabitat affinities) are found in two distinct cladesconsistent with geography but not with current taxonomy Theseare represented by the thick green blue and red branches inFig 6 The first clade includes two species (Desis formidabilis(O Pickard-Cambridge 1890) and D marina (Hector 1877))that are nested within a group including the Austral cribellatespiders of New Zealand while the second includes all exemplarsof Argyroneta Saltonia and Paratheuma as well asDictyna sp(pp = 10)
Additionallargercladesapproximatingtaxaatthelsquofamilyrsquo levelcomprise a cybaeid group including Cybaeus L Koch 1868Calymmaria Chamberlin amp Ivie 1937 Cybaeina Chamberlin ampIvie 1932 Cybaeota Chamberlin amp Ivie 1933 New genus 4(Bennett 1991 p 156 Ubick et al 2005) BlabommaChamberlinamp Ivie 1937 Yorima Chamberlin amp Ivie 1942 New genus 1(Bennett 1991 p 168 Ubick et al 2005) and CryphoecaThorell 1870 and an agelenid group including all Holarctictaxa plus monophyletic Agelenidae (including Coelotines)(both supported with ppgt 095) Other nodes with strongstatistical support include a clade consisting of all cribellateAmaurobiidae sampled (pp = 10) and a sister relationship
between the aforementioned Holarctic agelenids and Cybaeidaesl (pp = 096)
Comparing the topology from the combined-data structurallymodelled analysis with the topology from the MUSCLE-alignedanalysis shows 26 nodes statistically supported (pp gt 095 redasterisks in Fig 6) in both analyses while 10 additional nodes(marked with black asterisks) are supported at this level onlyin the structural analysis Two nodes (grey asterisks) receivestatistical support only in the analyses of the MUSCLE-aligneddata while one node (Dictyna Sundevall 1833 as the sistertaxon to Paratheuma+Argyroneta) receives support in theMUSCLE analysis but remains unresolved at the 050 level inthe structurally modelled tree and thus does not appear in thefigure
The tree topology yielded by the combined-data RAxMLanalysis was largely congruent with the Bayesian trees witha major difference in the placement of the genera PimusChamberlin 1947 Amaurobius C L Koch 1837 andCallobius Chamberlin 1947 These cribellate Amaurobiidaewhich emerge as a clade nested well within the RTA clade inthe Bayesian tree (see nodemarkedwith dagger in Fig 6) appear
CU
A
GG
G
A U
A
U
C
C
U
G
C
C
G
C
G
U
GG
UU GU
A
G
A
G
GU
G
CA
U
C
G
CC
A
U
G
C A
A
A
CC
G
A
A
G
C
U C
G
C
A
CG
C
A
A
G
C
A
U
G
C
G
C
G
GA
GU
U
U
U
G
U
G
G
C
G
A
A
C
A
A
A A
UUU
CAC
G CA
C GG CC
GA
A
AA
CC
G
AC
G
G
C
U
C
A UCG
C
C GG
C
G G
U
U
G
C
G
U C
A
C U
G
U
G
C
GU
CG
C
C
G
G
GU U U
GA
G
GGCC
C
UU
5 3
Dictyna sp
CUA
GGU
AG
G
U
C
A
G
C
C
G
G
A
G
UG
G
C
A
U
G
C
A
C
C
C
C
C
CA
G
C
G
U
A G
C
C
CC
C
GG
U
A
U
A
CA
C G
G
G
C
G
U
A
C
CA
G
C
G
A
UGU
ACA
GGC
CCGGC
G
A
A
CG
G
AG
G
G
C
U
A
A
UC
C GG
C
G U
C
C
C
C
G CC G
G
C
C
G
U
C
A
GU
G
C
G
G
A
A
G A
GG
G
CG
GA
CA
G G
G
A
C
A
U
A
5 3
Saltonia incerta
G
C
C
C
G
C
G
C
A
G
G
A
C
A
C U G
G
C
A
G
U G
G
C
C
G
C
C
G
C
C
C
C
G
C AA C
C
C
C
A
C
G
U
G C
G
C
G
CA
C
A
G
U
A
C
G
C
C
C
G
CU
U
G
C
A
A
U
A
C
C
G
A
A
G
C
AA
A
GU
A
AA
CC
G
AGG
U
G
G
G A
AU
G
C GC
C
C A
C
G
U
A
C G
A
C U
G
C
G
GU
G
C
UC
G
C
G
GU G U
A
G
C
GG
G G
GG
GG
C
GG
A
Paratheuma insulana
C
A
U
G
C C
G
G
C
G
GA
G C U
A
C
G
U U
A
C
A
U
A
C A
C
A
G
U
G
G
AG
C
C
G
CC
C
C
G
G
UG
A GU
U
C
C C
C
G
C
U
A
C
G
G
A
G
C
A
G
U
U
UU
G
G A
A U
CC
G
A C G
G
A
U
G
C
CG
A C G
C
A G
U
G
CC
G
C
G
UG
U
U
G
CG GCA
AC
AU
A
A
A U
A
C
GC
CG
A
A
G GG
UG
CA
C
U
CG
G
GG C
C
CAU
UU
GG
5 3
Zygiella x-notata
CUA
GGU
AGU
CCA
UGU
ACA
GGC
CCGGC
5 3
UA
GU
AGU
CCA
UGU
ACA
G
CORE + D3
Fig 5 The secondary structures for the lsquocorersquo plus lsquoD3rsquo regions of 28S rRNA This region was highly conserved across taxa thus only a few divergentrepresentatives are shown These includeDictyna Saltonia andParatheuma from the lsquoaquaticrsquo clade and two outgroup taxa S incerta is fairly divergent inthis region of 28S rRNA which is typically highly conserved Several deletions appear to have taken place in the Saltonia 28S rRNA gene
246 Invertebrate Systematics J C Spagna et al
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
as a basal clade (bootstrap value 27 for the node separating thePimus species from Amaurobius +Callobius) in the RAxMLanalysis Bootstrap values for the RAxML analysis can beseen in Fig 6
All combined-data trees show a branch to Saltonia that islonger than the rest of the branches in the tree though this effect ismuch more extreme in the RAxML tree than in the Bayesiandoublet-modelled tree
aquac
clade
Austral
clade
Agelenidae sl
Cybaeidae sl
Cribellate A
maurobiidae
Coelonae A
geleninae
98
48
91
100
88
99
86
72
100
98
100
45
49 100
100
dagger
100
100
100
100
100
100
74
100
49
81
98100
9199
98
92
6071
100
34
89
Fig 6 Combined-data tree Majority-rule consensus tree of 6million generations after stationarity wasachieved in Bayesian analysis of the combined-data matrix Asterisks () denote branches receiving posteriorprobability support of95 in theMUSCLE-alignment only (grey asterisks) structurally aligned and doublet-modelled only (black asterisks) or in both alignments (red asterisks) Coloured branches denote habitat affinityas followsblue = aquatic green = intertidal and red = salt-flatsalt crustsLargemonophyletic groups supportedin combined analysis and discussed in text are labelled with clade names Numbers represent bootstrap valuesfrom theRAxMLanalysis onnearest nodes receiving statistical support inBayesian analysiswith thedagger (dagger)representing the single supported node not supported in the best ML tree (see text for details)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 247
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Single-gene trees
18S rRNA
The single-gene tree from the 18S rRNA data (Fig 7) largelyreflects the relationships seen in the combined-data tree with afew notable exceptions The split between the Holarctic andAustral intertidal desids is present in the 18S rRNA tree (thickgreen branches) however within the Holarctic clade theintertidal taxa (Paratheuma) form a tritomy with Argyronetaand (Saltonia +Dictyna) instead of being sister to Argyroneta
The Holarctic agelenid clade is present but with poor resolutionbetween the coelotine agelenine tegenariine and textricinetaxa Most (21 of 22) nodes receiving statistical support in thestructurally partitioned treewere also supported in theMUSCLE-aligned 18S rRNA tree while five nodes were supported in theMUSCLE-aligned tree only Two of these five nodes do notappear in Fig 5 ndash (Cicurina Menge 1871 + CheiracanthiumC L Koch 1839 + Amaurobius +Callobius+ lsquoaquatic cladersquoand Saltonia+Paratheuma) as the tree in the figure representsthe majority-rule consensus of structurally modelled trees The
Cybaeota sp 2
Fig 7 18S rRNA gene tree Majority-rule consensus of 4million post-stationarity generations of the 18SrRNA matrix Labelling and colour conventions as in Fig 6
248 Invertebrate Systematics J C Spagna et al
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
branch leading to the Saltonia taxa is the longest in the tree as inthe combined-data tree
28S rRNA
The majority-rule consensus 28S rRNA tree (Fig 8) reflectsmany of the same general relationships as the combined-data and18S rRNA gene trees but shows less support for some key nodesof interest anddoesnot provide statistically significant (95pp)support for some of the nodes in the combined-data tree that arethe focus of this study Individual differences include paraphyly
of the RTA clade ingroup dissolution of the Austral clade andthe presence of the Saltonia clade in an unsupported sister-relationship with Neoantistea Gertsch 1934 (HahniidaeBertkau 1878) The relationships of the focal taxa ndash SaltoniaArgyronetaParatheuma andDesisndash are similar to the combined-data tree including the geographic split between intertidalspecies however neither the node supporting those inhabitingthe northern hemisphere (Paratheuma spp) nor the nodesupporting the Austral association is statistically supported atthe 95 (pp = 091 and 088 respectively)
Fig 8 28S rRNA gene tree D2 +D3 and core regions Majority-rule consensus of 11 thousand treessampled from 11million post-stationarity generations of the 28S rRNA matrix Labelling and colourconventions as in Fig 6
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 249
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Neither the Holarctic Agelenidae nor the cybaeid group(Fig 6) achieve statistical significance (pp = 089 and 091respectively) in the 28S rRNA gene tree and both containunresolved and poorly supported internal nodes Eighteen of19 supported nodes in the structurally modelled 28S rRNAgene tree were also supported in the MUSCLE-alignedanalysis while 10 (including 2 not in the majority-ruleconsensus structural tree- Stegodyphus+Megadictyna Dahl1906 + Zygiella F O P-Cambridge 1902 and an Australclade including Stiphidion ndash not figured) fell below the 095pp threshold when the structural partitions were used
Histone H3
Contrasting with the gene trees from rRNA matrices theprotein-coding histone H3 gene tree (not included as a figure)based on 349 bp of data offers little resolution at the 95 pplevel with the exception of some non-controversial groupings ofclosely related taxa These stable groupings include monophylyof three focal genera (as genera not as a single inclusive group)DesisParatheuma and Saltonia andmonophyly of all agelenineexemplars (Agelena Walckenaer 1805 + Agelenopsis Geibel1869 + Barronopsis Chamberlin amp Ivie 1941) and coelotineexemplars (Eurocoelotes Wang 2002 +Coelotes Blackwall1841)
Discussion
Secondary structure patterns in the RTA clade
TheArm 1 structures from 28S rRNA (Fig 2) for members of theaquatic clade (Fig 6) appear reduced in length relative to otherexemplars but there is no clear trend towards increase orreduction in size across taxa Of the large clades represented inFig 1 the Cybaeidae sl and Dictynidae appear the moststructurally uniform while the Agelenidae and Coelotinaestructures appear quite variable For Arm 2 (Figs 1 3) theaquatic-clade exemplars vary little in general secondarystructurewith visually homologisable loops and stems in all taxa
Overall the structural conservation of the loops and stemsof all three arms of the D2-loop sequenced suggest thatour methods produced reasonable hypotheses of primaryhomology for modelling stems and loops and can be used as abasis for similar work at a variety of levels across the RTA cladeand possibly in related spider taxa Though modelling genes andarms individually was tractable for this study for a larger dataset(either more sequence data or more taxa) a more automatedapproach might be desirable In such a case a variety of methodsfor determination of secondary structure based on two or moreRNA sequences are available such as Dynalign (Mathews andTurner 2002 as implemented in the software package RNAStructure ver 503 Mathews 2010) and RNAalifold (Hofackeret al 2002 Bernhart et al 2008)
Doublet modelling
While doublet models have been shown to perform well withsingle genes (Kim et al 2003Angelini and Jockusch 2008) thereare few published studies addressing the performance of doubletmodels when used with multiple gene partitions (16S and 28SrRNA genes ndashDeans et al 2006 18S and 28S rRNA genes ndashRixet al 2008) and none that deal with divergences at the temporal
scale addressed here in which the divergence point between theingroup and outgroup taxa has been estimated to be mid-Jurassic(Penney and Ortuntildeo 2006) this deep divergence is reflected inthe high degree of variation in the arms of the D-loop region of28S rRNA Despite this variation both within and between taxaparticularly for Arm 1 of the D-loop region of 28S rRNA therewas sufficient similarity to allow primary homology assessmentand assignment of stretches of primary sequence to partitions
It might appear that the application of the doublet results in alsquoworsersquo outcome if one focusses only on support values andresolution of individual clades in single-gene trees The 28SrRNA tree (Fig 8) has considerably less support (10 of 28shared nodes fall below the statistical threshold of 95posterior probability) when the stemndashloop partition anddoublet model are applied than when the computer-generatedalignment is used The effect is less stark in the 18S rRNA treewith only 3 of 25 shared nodes losing statistical support when thedoublet is applied However to understand whether the loss ofsupport in single-gene trees really represents a loss of analyticalpower it is important to consider the changes in the topology andsupport of the combined-data trees (Fig 6) The tree remains wellresolved and well supported when the doublet model is appliedand in fact the difference between theMUSCLE-aligned trees andthe doublet-modelled trees shows the opposite effect than whenapplying the model to the single-gene datasets instead of losingstatistical support 10 additional nodes gain support throughapplication of the doublet model leading to a consensus treein which 36 nodes are supported by statistically significantposterior probabilities
This effectndash lossof support in single-gene trees but increase insupport in combined-data trees ndash has been explained by Kjer(2004) in his study using 18S data of holometabolous insectsreduction in support may result when pairs of non-independentcharacters are effectively down-weighted when recognised andtreated as non-independent With pairs of bases as the unitcharacter the total number of stem characters is reduced byhalf and fewer characters can reduce resolution Even if theoutcome of paired-sites modelling is not higher support or anincrease in some other measure of confidence in the phylogenetichypotheses generated it does correct the violations of assumedindependence inherent inmostmodels used in likelihood analysisof self-complementary genes reduced support may in fact be animprovement in the analytical outcome if the alternative is highsupport for an incorrect clade
In the context of a multi-gene analysis such as this study thesupport lost in the single-gene trees can be made up for by strongsignal fromother partitions but this pattern canonlybe seenwhenmultiple genes are used This is not always the case though as acomparable study of a different clade of spiders (Rix et al 2008)showed decreased support in a combined-data 18S rRNA+28SrRNA tree This might be due to a smaller dataset (720 bp in theformer study versus ~1800 used here and no protein-coding datain that study) and underlines the importance of alternate lines ofevidence (multiple genes with a variety of genomic sources) in acombined analysis
The length of the Saltonia branch in the combined-dataanalyses which appears to be contributed primarily by the 18SrRNA data (Fig 6) may result from either rapid evolution of the18S rRNA gene or a disproportionate amount of speciation
250 Invertebrate Systematics J C Spagna et al
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
extinction in the Saltonia lineage (Bergsten 2005) Though gene-duplication and functional divergence in paralogues has beenshown to result in rapid genetic change leading to long branches(Fares et al 2006) examination of secondary structure relative tomodels from other spiders shows no particular violations ofstemndashloop expectations which one would expect to find in apseudogene Post hoc examination of base composition revealsno significant deviation in theSaltonia18S rRNAgene (A = 24T = 26 C= 22G= 28) from the percentages found in othertaxa (A= 24 T = 25 C= 23 G= 28 c2 = 089 df = 3P gt 08) We remain confident that the Saltonia 18S rRNAgenes sequenced are orthologues of those from other spidersboth their adherence to secondary-structure rules for eukaryotesgenerally and spiders specifically (Hendriks et al 1988 Spagnaand Gillespie 2006) and the fact that we sequenced threeindividuals from distant populations and found them to bequite similar The branchrsquos phylogenetic position was robust tochanges in method (Bayesian withwithout doublet andRAxML) and it showed no tendency to group with otherlong branches such as the outgroup branches or the hahniidexemplar Neoantistea Use of the doublet model reduced therelative length of the branch probably due to down-weightingcomplementary changes in paired stem bases which furthersupports maintenance of function in the gene (stems would notbe under selective pressure to maintain pairing integrity in apseudogene) although the forces driving the relatively rapidaccumulation of base differences remain unknown
Taxonomic affinities of target taxa
An unexpected outcome of the study was the close phylogeneticaffinity among three of the target taxa the arachnid extremophilesArgyroneta Paratheuma and Saltonia (Fig 6) The combinedanalysis placed these taxa together in a clade with DictynaThough the grouping of Paratheuma and Argyroneta has beenpredicted based on previous molecular analyses (Spagna andGillespie 2008Miller et al 2010) the close relationship betweenthese two taxa and Saltonia the position of which has long beenenigmatic in spider systematics (Roth and Brown 1975b Foelix1996) is unexpected Indeed the combined-data tree supportsnone of the current family-rank groupings of spiderswith extrememicrohabitat affinitiesArgyroneta does not fall with others in theCybaeidae as has long been classified based on morphologyand fossil evidence (Simon 1892 Lehtinen 1967 Selden 2002)but rather must be derived from a dictynid-like ancestor whileprobably being sister to Paratheuma The new position of thelatter taxon ndash traditionally and currently considered a desid(Platnick 2010) ndash would make Desidae polyphyleticDictynidae is also polyphyletic in these analyses with fourrepresentatives (Dictyna Lathys Simon 1884 Cicurina andSaltonia) never forming a clade in the combined or single-gene trees
Despite conflicts with morphological taxonomy the resultsof these analyses are robust and are largely consistent withother recent studies including RTA-clade taxa (Griswold et al1999 2005 Spagna and Gillespie 2008 Miller et al 2010)The increase in support when the data are modelled to accountfor base-pairing and data from all three genes are included(Fig 6) suggests that we have developed a hypothesis that is
more reliable than the consensus result for the matrix whensecondary structure is not considered Thus we consider ourapproach to be a major methodological improvement for thestudy of ancient divergences in spiders although future workshould benefit not only from use of secondary structurebut also from the development of newer possibly better-parameterised likelihood models incorporating doubletinformation Such models include the RNA7a and RNA7bmodels as implemented in the program PHASE (Gowri-Shankar and Jow 2006) and their incorporation into partitionedBayesian analyses Comparison of Bayesian likelihood andaccelerated likelihood underlines the consistency of the wellsupported clades here with the exception of the lsquoCribellateAmaurobiidaersquo which are placed in a basal position and arenot monophyletic within the RTA clade only in the RAxMLanalysis Though this difference does little to settle the difficultproblems of the phylogenetic composition and relationships ofa monophyletic Amaurobiidae it has little effect on the evolutionof the dictynoid extremophiles and parallel evolution betweenAustral and worldwide species which are the foci for this study
Geographic convergence
It has been suggested that the spider fauna of New Zealand andAustralia represents an assortment of relict taxa with familiessuch as Gradungulidae Forster 1955 and Austrochilidae Zapfe1955 maintaining plesiomorphic features not common in morewidely distributed spider taxa (Forster 1987 Forster and Forster1999) This idea has affected studies of spider evolution byinfluencing taxon sampling in datasets assembled to addressdeep phylogenetic questions For example based on thepremise that Austral taxa often represent basal lineages ofworldwide families the taxon Neoramia has been used as thesole lsquoplesiomorphicrsquo exemplar of the family Agelenidae in avariety of studies (Forster and Wilton 1973 Coddington andLevi 1991 Griswold et al 1999 Blackledge et al 2009)Previously this approach has been shown to be flawed as theAustral and worldwide Agelenidae are not monophyletic(Spagna and Gillespie 2008 Miller et al 2010) The currentstudy also renders the family Desidae polyphyletic and showsindependent origins for derived similarities in widely separatedgenera rather than simple plesiomorphy in the basalAustralrestricted branches of a worldwide family Within theAustralian desid taxa the intertidal species has shifted toecribellate silk production supporting the general suppositionthat Austral taxa exhibit shifts between these cribellate andecribellate morphologies (Forster and Wilton 1973 Forsterand Gray 1979 Griswold et al 1999) while reinforcing themore recent hypothesis that the inclusion of Austral taxa inworldwide families may make them polyphyletic (Spagna andGillespie 2008)
The expansion of the lsquoAustral cladersquo to include members ofanother worldwide family (Desidae) makes it clear that a majorunsolved aspect of spider phylogeny is the true extent of this clade(or clades as the casemay turn out to be under broader sampling)and its relation to other RTA-clade spiders While some Australtaxa clearly belong in worldwide families (eg AustralianLycosidae Sundevall 1833 Australian Selenopidae Simon1897 Australian orb-weavers and derived orb-weavers) this
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 251
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
work casts doubt on the proper phylogenetic placement of a largenumber of taxa particularly those restricted to New Zealandmany of which were described as and remain in Agelenidae(Forster and Wilton 1973 Platnick 2010)
An lsquoaquaticrsquo clade
In the current study one shift back to the aquatic habitat (theArgyroneta+Paratheuma+ Saltonia line) led to freshwatersaltwater and salt-flat species Relative to taxonomy the closerelationship between these taxa is a novel result However giventhe documented similarities in some characters including welldeveloped tracheae and an oversized colulus this result is notaltogether surprising The most intriguing aspect of this result isthatwhile these environments are fundamentally different (underfresh water intertidal dry salt lakes) and present very differentphysiological challenges it may be that the spiders share anancient history and are an early offshoot of Dictynidae a familywith the oldest known fossils being dated to the mid-Cretaceous(Penney et al 2003 Penney and Ortuntildeo 2006) By contrastdictynids of more typical terrestrial environments are mostcommonly associated with vegetation making small irregularwebs under tree bark below (or in curled) leaves or in clustersof sticks or flowers although some may be found in leaf litter orsoil (Chamberlin and Gertsch 1958 Hagley and Allen 1989)Accordingly these taxamay be relictualmembers of a previouslymore widespread and more physiologically homogeneouslineage adapted to an aquatic existence Although intriguingthis result might not be surprising as aquatic habitats generallyappear to foster relictualism (Humphreys 2006) The currentoccurrence of one of the aquatic taxa in what are now dryalkaline salt-flats is likely the result of these relictual taxaadapting over time as their aquatic environment dried upleaving the current disjunct distribution of Saltonia(S C Crews and R G Gillespie unpubl data) This highlightsthe extreme niche conservatism found among spiders at leastwith regards to the terrestrial versus aquatic lifestyle
It should be noted that although freshwater and saltwaterphysiologies are highly divergent in many organisms due tothe opposite osmotic pressures presented by the twoenvironments this may not be true for spiders in which freshand salt-water spiders share many characteristics The differenceis that a spiderrsquos method for dealing with both kinds of aquaticenvironment is similar employing strategies to minimiseinteractions with the water directly The spiders more or lessinsulate themselves from contact with the water using somebehavioural or mechanical trick such as carrying around anair bubble (Argyroneta) (Shirtcliffe et al 2006) or sealingthemselves in barnacle shells during high tide (Paratheuma)(Roth and Brown 1975a) The close relationships among thedifferent aquatic extremophiles shown in this study raise someimportant questions regarding the underlying behavioural orecological adaptations that allow spiders to adapt to largefluctuations of moisture temperature and salinity
Conclusions
By providing evidence for the relative rarity of shifts by spidersto aquatic semi-aquatic and historically aquatic environmentsthis study contributes to the general picture of long-term
morphological stasis in arachnids and extreme age (gt90MY)for their major adaptations (such as evolution of venoms use ofsilks as prey snares and the shift from hackled-silk webs to glue-coated webs (Eberhard 1990 Coddington and Levi 1991 Opell1999)Moreover thiswork suggests that convergent evolution ofsome aquatic features hinder a correct understanding of thehistorical pattern of spider evolution In particular it appearsthat Austral lineages may not provide a basal framework of theoverall entelegyne tree (Forster andWilton 1973 Griswold et al1999) Rather theremay be anAustral versusworldwide divisiondeepwithin the RTA clade followed by independent sets of shiftsof important characters such as production of cribellate silk andintertidal habitat affinity The study provides strong evidence foran important role for conservative evolution of aquaticspecialisation followed by ancient relictualism allowing thesignature of aquatic lifestyle to be maintained long after thegeological environment itself has been modified
Understanding of deep divergences between spider lineageslike those we have explored here has heretofore been hinderedby the relatively small number of genetic markers of provenutility for phylogenetic study (Ayoub et al 2007) We havedemonstrated methodological improvements in particularenhanced partitioned analysis through modelling of secondarystructure thatndashwhen applied to themost common lsquoslowrsquomarkers(18S rRNA and 28S rRNA) ndash can help compensate for thisshortage of data sources
Authorsrsquo contributions
JCS and SCC designed the study modelled the secondarystructures and drafted the manuscript SCC amplified andsequenced genetic loci and analysed the data RGG assistedwith drafts of the manuscript and provided project guidance andmaterial support for the laboratory work All authors read editedand approved the final manuscript
Acknowledgements
The authors would like to thank Charles Griswold Joel Ledford and BrentOpell for loan of specimens crucial to this work All new specimens werecollected under appropriate permits We thank Jim McGuire and MattBrandley for use of and assistance with computing resources at UCB andMark Miller and Lucie Chan for assistance with the CIPRES web portal Wethank David Gilley for comments leading to many improvements in themanuscript Funding for the project was provided by the SchlingerFoundation with additional support from the Division of Insect Biology atUC Berkeley and the National Science Foundation (Grant DGE 0231877)
References
Angelini D R and Jockusch E L (2008) Relationships among pest flourbeetles of the genus Tribolium (Tenebrionidae) inferred from multiplemolecularmarkersMolecular Phylogenetics and Evolution 46 127ndash141doi101016jympev200708017
Ayoub N A and Riechert S E (2004) Molecular evidence for Pleistoceneglacial cycles driving diversification of a North American desert spiderAgelenopsis aperta Molecular Ecology 13 3453ndash3465 doi101111j1365-294X200402335x
AyoubN A Garb J E HedinM andHayashi C Y (2007) Utility of thenuclear protein-coding gene elongation factor-1 gamma (EF-1g) forspider systematics emphasizing family level relationships of tarantulasand their kin (Araneae Mygalomorphae) Molecular Phylogenetics andEvolution 42 394ndash409 doi101016jympev200607018
252 Invertebrate Systematics J C Spagna et al
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Beatty J A and Berry J W (1988a) The spider genus ParatheumaBryant(Araneae Desidae) The Journal of Arachnology 16 47ndash54
Beatty J A and Berry J W (1988b) Four new species of Paratheuma(Araneae Desidae) from the Pacific The Journal of Arachnology 16339ndash347
Bennett R G (1991) lsquoThe Systematics of the North American CybaeidSpiders (Araneae Dictynoidea Cybaeidae)rsquo (University of GuelphGuelph Ontario)
Bergsten J (2005) A review of long-branch attraction Cladistics 21163ndash193 doi101111j1096-0031200500059x
Bernhart S Hofacker I Will S Gruber A and Stadler P (2008)RNAalifold improved consensus structure prediction for RNAalignments BMC Bioinformatics 9 474 doi1011861471-2105-9-474
Blackledge T A Scharff N Coddington J A Szuumlts T Wenzel J Wet al (2009) Reconstructing web evolution and spider diversificationin the molecular era Proceedings of the National Academy of Sciencesof the United States of America 106 5229ndash5234 doi101073pnas0901377106
Brandley M C Shmitz A and Reeder T W (2005) Partitioned Bayesiananalyses partitionchoice and thephylogenyof scincid lizardsSystematicBiology 54 373ndash390 doi10108010635150590946808
Chamberlin RV andGertschW J (1958) The spider familyDictynidae inAmerica north of Mexico Bulletin of the American Museum of NaturalHistory 116 1ndash152
Coddington J A and Levi H W (1991) Systematics and evolution ofspiders (Araneae) Annual Review of Ecology and Systematics 22565ndash592 doi101146annureves22110191003025
Coddington J A Giribet G Harvey M S Prendini L andWalter D E(2004) Arachnida In lsquoAssembling the Tree of Lifersquo (Eds J Cracraft andM Donoghue) pp 296ndash318 (Oxford University Press New York)
Colgan D McLauchlan A Wilson G and Livingston S (1998) HistoneH3 and U2 snRNA DNA sequences and arthropod molecular evolutionAustralian Journal of Zoology 46 419ndash437 doi101071ZO98048
Crews S C Puente-Rolon A R Rutstein E and Gillespie R G (2010)A comparison of populations of island and adjacent mainland speciesof Caribbean Selenops (Araneae Selenopidae) spiders MolecularPhylogenetics and Evolution 54 970ndash983 doi101016jympev200910012
Deans AR Gillespie J J andYoderM J (2006) An evaluation of ensignwasp classification (Hymenoptera Evaniidae) based on molecular dataand insights from ribosomal RNA secondary structure SystematicEntomology 31 517ndash528 doi101111j1365-3113200600327x
DixonMT andHillisDM (1993)Ribosomal-RNAsecondary structurendashcompensatory mutations and implications for phylogenetic analysisMolecular Biology and Evolution 10 256ndash267
Eberhard W G (1990) Function and phylogeny of spider webs AnnualReview of Ecology and Systematics 21 341ndash372 doi101146annureves21110190002013
Edgar R C (2004) MUSCLE multiple sequence alignment with accuracyand high throughput Nucleic Acids Research 32 1792ndash1797doi101093nargkh340
ErpenbeckDNichols SAVoigtODohrmannMDegnanBM et al(2007) Phylogenetic analyses under secondary structure-specificsubstitution models outperform traditional approaches case studieswith diploblast LSU Journal of Molecular Evolution 64 543ndash557doi101007s00239-006-0146-3
Fares M A Byrne K P and Wolfe K H (2006) Rate asymmetry aftergenome duplication causes substantial long-branch attraction artifactsin the phylogeny of saccharomyces species Molecular Biology andEvolution 23 245ndash253 doi101093molbevmsj027
Felsenstein J (1978) Cases in which parsimony or compatibility methodswill be positively misleading Systematic Zoology 27 401ndash410doi1023072412923
Foelix R F (1996) lsquoBiology of Spidersrsquo 2nd edn (Oxford UniversityPress New York)
Forster R R (1987) A review of the spider superfamilies Hypochiloidea andAustrochiloidea (Araneae Araneomorphae) Bulletin of the AmericanMuseum of Natural History 185 1ndash116
Forster R R and Forster L (1999) lsquoSpiders of New Zealand and TheirWorldwide Kinrsquo (University of Otago Press Dunedin)
Forster R R andGrayMR (1979)Progradungula a new cribellate genusof the spider family Gradungulidae (Araneae) Australian Journal ofZoology 27 1051ndash1071 doi101071ZO9791051
Forster R R andWilton C L (1973) The spiders of NewZealand Part IVOtago Museum Bulletin 4 1ndash309
Gillespie J J Cannone JGutell R andCognatoA I (2004)A secondarystructural model of the 28S rRNA expansion segments D2 and D3 fromrootworms and related leaf beetles (Coleoptera ChrysomelidaeGalerucinae) Insect Molecular Biology 13 495ndash518 doi101111j0962-1075200400509x
Giribet G Carranza S Baguntildea N RiutortM andRibera C (1996) Firstmolecular evidence for the existence of a Tardigrada +Arthropoda cladeMolecular Biology and Evolution 13 76ndash84
Givnish T J and Systma K J (Eds) (1997) lsquoMolecular Evolutionand Adaptive Radiationrsquo (Cambridge University Press CambridgeUK)
Glesne R (1998) Terrestrial riparian arthropod investigations in the BigBeaver Creek Research Natural Area North Cascades National ParkService Complex 1995ndash1996 Part III Arachnida Araneae TechnicalReport NPSNRNOCANRTR98-03 United States Department ofInterior ndash National Park Service ndash Pacific West Region
Gowri-Shankar V and Jow H (2006) PHASE A Software Package forPhylogenetics and Sequence EvolutionManual version 20Available athttpwwwcsmanchesteracukaiSoftwarephasephase-20-manualpdf [Accessed 1 October 2009]
Griswold C E Coddington J A Platnick N I and Forster R R (1999)Towards a phylogeny of entelegyne spiders (Araneae AraneomorphaeEntelegynae) The Journal of Arachnology 27 53ndash63
Griswold C E Ramirez M J Coddington J A and Platnick N I (2005)Atlas of phylogenetic data for entelegyne spiders Proceedings of theCalifornia Academy of Sciences 56(Suppl II) 1ndash324
HagleyEAC andAllenWR (1989) Prey of the cribellate spiderDictynaannulipes (Araneae Dictynidae) on apple tree foliage The Journal ofArachnology 17 366ndash377
HedinMC andMaddisonWP (2001)Acombinedmolecularapproach tophylogeny of the jumping spider subfamily Dendryphantinae (AraneaeSalticidae) Molecular Phylogenetics and Evolution 18 386ndash403doi101006mpev20000883
Hendriks L Van Broeckhoven C Vandenberghe A Van de Peer Y andDe Wachter R (1988) Primary and secondary structure of the 18Sribosomal RNA of the bird spider Eurypelma californica andevolutionary relationships among eukaryotic phyla European Journalof Biochemistry 177 15ndash20 doi101111j1432-10331988tb14339x
Hofacker I L Fekete M and Stadler P F (2002) Secondary structureprediction for aligned RNA sequences Journal of Molecular Biology319 1059ndash1066 doi101016S0022-2836(02)00308-X
Huelsenbeck J P and Rannala B (2004) Frequentist properties ofBayesian posterior probabilities of phylogenetic trees under simple andcomplex substitution models Systematic Biology 53 904ndash913doi10108010635150490522629
Huelsenbeck J P and Ronquist F (2001)MRBAYES Bayesian inferenceof phylogenetic trees Bioinformatics 17 754ndash755 doi101093bioinformatics178754
Humphreys W F (2006) Aquifers the ultimate groundwater-dependentecosystems Australian Journal of Botany 54 115ndash132 doi101071BT04151
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 253
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Jeram A J Selden P A and Edwards D (1990) Land Animals in theSilurian arachnids and myriapods from Shropshire England Science250 658ndash661 doi101126science2504981658
Kelchner S A and ThomasM A (2007)Model use in phylogenetics ninekey questions Trends in Ecology amp Evolution 22 87ndash94 doi101016jtree200610004
Kim S Kjer K M and Duckett C N (2003) Comparison betweenmolecular and morphological-based phylogenies of galerucinealticineleaf beetles Insect Systematics amp Evolution 34 53ndash64
Kjer K M (1995) Use of rRNA secondary structure in phylogenetic studiesto identify homologous positions an example of alignment and datapresentation from the frogs Molecular Phylogenetics and Evolution 4314ndash330 doi101006mpev19951028
Kjer K M (2004) Aligned 18S and insect phylogeny Systematic Biology53 506ndash514 doi10108010635150490445922
Kolaczkowski B and Thornton J W (2009) Long-branch attraction biasand inconsistency in Bayesian phylogenetics PLoS ONE 4 e7891doi101371journalpone0007891
Lee M S Y (1998) Convergent evolution and character correlation inburrowing reptiles towards a resolution of squamate relationshipsBiological Journal of the Linnean Society Linnean Society of London65 369ndash453 doi101111j1095-83121998tb01148x
Lehtinen P (1967) Classification of the cribellate spiders and some alliedfamilies Annales Zoologici Fennici 5 199ndash468
MacNaughton R Cole J Dalrymple R Braddy S Briggs D and LukieT (2002) First steps on land arthropod trackways in Cambrian-Ordovician eolian sandstone southeastern Ontario CanadaGeology 30391ndash394 doi1011300091-7613(2002)030lt0391FSOLATgt20CO2
Maddison W P and Hedin M C (2003) Jumping spider phylogenyInvertebrate Systematics 17 529ndash549 doi101071IS02044
Mallatt J M Garey J R and Shultz J W (2004) Ecdysozoan phylogenyand Bayesian inference first use of nearly complete 28S and 18S rRNAgene sequences to classify the arthropods and their kin MolecularPhylogenetics and Evolution 31 178ndash191 doi101016jympev200307013
Martins E P (2000) Adaptation and the comparative method Trends inEcologyampEvolution 15 296ndash299 doi101016S0169-5347(00)01880-2
Mathews D H (2010) RNA Structure version 503 Available at httprnaurmcrochestereduRNAstructurehtml
Mathews D H and Turner D H (2002) Dynalign an algorithm for findingthe secondary structure common to two RNA sequences Journalof Molecular Biology 317 191ndash203 doi101006jmbi20015351[Verified July 2010]
McQueenD J andMcLayCL (1983)Howdoes the intertidal spiderDesismarina (Hector) remain under water for such a long time New ZealandJournal of Zoology 10 383ndash392
MillerM A HolderM T Vos RMidford P E Liebowitz T Chan LHoover P and Warnow T (2009) lsquoThe CIPRES Portalsrsquo Availableat httpwwwphyloorgsub_sectionsportal [Accessed 23 December2009]
Miller J A Carmichael A Ramiacuterez M J Spagna J C Haddad C RRezac M Johannesen J Kral J Wang X-P and Griswold C E(2010) Phylogeny of entelegyne spiders affinities of the familyPenestomidae (NEW RANK) generic phylogeny of Eresidae andasymmetric rates of change in spinning organ evolution (AraneaeAraneoidea Entelegynae) Molecular Phylogenetics and Evolution55 786ndash804 doi101016jympev201002021
MurphyNPFramenauVWDonnellanSCHarveyMS ParkY-Cand Austin A D (2006) Phylogenetic reconstruction of the wolf spiders(Araneae Lycosidae) using sequences from the 12S rRNA 28S rRNAandNADH1 genes implications for classification biogeography and theevolution of web building behavior Molecular Phylogenetics andEvolution 38 583ndash602 doi101016jympev200509004
Nylander J A A (2004) lsquoMrModeltest (Version 22)rsquo Program distributedby the author Uppsala University Available at httpwwwabcse~nylandermrmodeltest2mrmodeltest2html [Verified July 2010]
Opell B D (1999) Changes in spinning anatomy and thread stickinessassociated with the origin of orb-weaving spiders Biological Journalof the Linnean Society Linnean Society of London 68 593ndash612doi101111j1095-83121999tb01190x
Penney D (2003) Does the fossil record of spiders track that of theirprincipal prey the insects Transactions of the Royal Society ofEdinburgh Earth Sciences 94 275ndash281 doi101017S0263593300000675
Penney D and Ortuntildeo V M (2006) Oldest true orb-weaving spider(Araneae Araneidae) Biology Letters 2 447ndash450 doi101098rsbl20060506
Penney DWheater C P and Selden P A (2003) Resistance of spiders toCretaceousndashTertiary extinction events Evolution 57 2599ndash2607
Pisani D Poling L L Lyons-Weilier M and Hedges S B (2004) Thecolonization of land by animals molecular phylogeny and divergencetimes among arthropods BMCBiology 2 1ndash10 doi1011861741-7007-2-1
Platnick N I (2010) The world spider catalog version 105 AmericanMuseum of Natural History Available at httpresearchamnhorgentomologyspiderscatalogindexhtml [Accessed April 10 2010]
PoinarGKerpH andHassH (2007)Palaeonemaphyticumgen n sp n(Nematoda Palaeonematidae fam n) a Devonian nematode associatedwith early land plants Nematology 10 9ndash14
Posada D and Buckley T R (2004) Model selection and model averagingin phylogenetics advantages of Akaike Information Criterion andBayesian approaches over likelihood ratio tests Systematic Biology53 793ndash808 doi10108010635150490522304
RixMGHarveyMS andRoberts J D (2008)Molecular phylogeneticsof the spider family Micropholcommatidae (Arachnida Araneae) usingnuclear rRNA genes (18S and 28S) Molecular Phylogenetics andEvolution 46 1031ndash1048 doi101016jympev200711001
Ronquist F and Huelsenbeck J (2003)MrBayes 3 Bayesian phylogeneticinference under mixed models Bioinformatics 19 1572ndash1574doi101093bioinformaticsbtg180
Roth V D (1967) A review of the South American spiders of the familyAgelenidae (Arachnida Araneae) Bulletin of the American Museum ofNatural History 134 297ndash346
Roth V D and Brown W L (1975a) A new genus of Mexican intertidalzone spider (Desidae) with biological and behavioral notes AmericanMuseum Novitates 2568 1ndash7
Roth V D and Brown W L (1975b) Comments on the spiderSaltonia incerta Banks (Agelenidae) The Journal of Arachnology 353ndash56
SantaLucia J J (1998) A unified view of polymer dumbbell andoligonucleotide DNA nearest-neighbor thermodynamics Proceedingsof the National Academy of Sciences of the United States of America95 1460ndash1465 doi101073pnas9541460
Schoumlniger M and von Haeseler A (1994) A stochastic model for theevolution of autocorrelated DNA sequences Molecular Phylogeneticsand Evolution 3 240ndash247 doi101006mpev19941026
Selden P A (1990) Terrestrialization invertebrates In lsquoPaleobiologyA Synthesisrsquo (Eds D E G Briggs and P R Crowther) pp 64ndash68(Blackwell Oxford UK)
Selden P A (2002) Missing links between Argyroneta and Cybaeidaerevealed by fossil spiders The Journal of Arachnology 30 189ndash200doi1016360161-8202(2002)030[0189MLBAAC]20CO2
Shirtcliffe N J McHale G Newton M I Perry C C and Pyatt F B(2006) Plastron properties of superhydrophobic surface Applied PhysicsLetters 89 104ndash106 doi10106312347266
Simon E (1892) lsquoHistoire naturelle des araignees (Vol 1)rsquo (Paris)
254 Invertebrate Systematics J C Spagna et al
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Spagna JC andGillespieRG (2006)Unusually longHyptiotes (AraneaeUloboridae) sequence for small subunit (18S) ribosomal RNA supportssecondary structure model utility in spiders The Journal of Arachnology34 557ndash565 doi101636H05-541
Spagna J C and Gillespie R G (2008) More data fewer shifts molecularinsights into the evolution of the spinning apparatus in non-orb-weavingspiders Molecular Phylogenetics and Evolution 46 347ndash368doi101016jympev200708008
Stamatakis A (2006a) RAxML-VI-HPC Maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed modelsBioinformatics 22 2688ndash2690 doi101093bioinformaticsbtl446
Stamatakis A (2006b) Phylogenetic models of rate heterogeneity a highperformance computing perspective In lsquoProceedings of the 20th IEEEACM International Parallel and Distributed Processing Symposium(IPDPS2006) Rhodos Greecersquo
Stamatakis A Hoover P and Rougemont J (2008) A fast bootstrappingalgorithm for the RAxMLweb-servers Systematic Biology 57 758ndash771doi10108010635150802429642
Thomas B A (1972) A probablemoss from the Lower Carboniferous of theForest of Dean Gloucestershire Annals of Botany 36 155ndash161
Thorp J H and Kovich A P (1991) lsquoEcology and Classification of NorthAmerican Freshwater Invertebratesrsquo (Academic Press New York)
TsagkogeorgaGTuronXHopcroftRRTilakMK FeldsteinT et al(2009) An updated 18S rRNA phylogeny of tunicates based on mixtureand secondary structure models BMC Evolutionary Biology 9 187doi1011861471-2148-9-187
Ubick D Paquin P Cushing P E and Roth V (2005) lsquoSpiders ofNorth America An Identification Manualrsquo (American ArachnologicalSociety Gainesville FL)
WakeD B (1991) Homoplasy the result of natural selection or evidence ofdesign limitations American Naturalist 138 543ndash567 doi101086285234
Wheeler W C and Hayashi C Y (1998) The phylogeny of the extantchelicerate orders Cladistics 14 173ndash192 doi101111j1096-00311998tb00331x
Wheeler W C Whiting M F Carpenter J C andWheeler Q D (2001)The phylogeny of the insect orders Cladistics 12 1ndash57 doi101111j1096-00311996tb00189x
Wilgenbusch J C Warren D L and Swofford D L (2004) AWTYa system for graphical exploration of MCMC convergence in Bayesianphylogenetic inference Available at httpking2scsfsueduCEBProjectsawtyawty_startphp [Accessed 1 May 2008]
Wise D H (1993) lsquoSpiders in Ecological Websrsquo (Cambridge UniversityPress Cambridge UK)
ZukerM (2003)Mfoldweb server for nucleic acid folding and hybridizationprediction Nucleic Acids Research 31 3406ndash3415 doi101093nargkg595
Manuscript received 8 January 2010 accepted 3 June 2010
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 255
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Appendix 1 Specimen and genetic data information for exemplars usedNS indicates no sequence was available for a particular locus
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Agelena labyrinthica Agelenidae NJNU03002 AY633851 AY633862 NSAgelenopsis aperta Agelenidae Spagna and Gillespie (2008) DQ628659 DQ628695 DQ628632
DQ628732Allagelena gracilens Agelenidae Spagna and Gillespie (2008) DQ628661 DQ628697 DQ628634
DQ628733Barronopsis barrowsi Agelenidae Spagna and Gillespie (2008) DQ628664 DQ628700 DQ628636
DQ628737Huangyuannia tibetana Agelenidae NJNU02001 AY633857 AY633859 NSNeoramia janus Agelenidae Spagna and Gillespie (2008) DQ628680 DQ628716 DQ628645
DQ628753Novalena intermedia Agelenidae Spagna and Gillespie (2008) DQ628679 DQ628715 NS
DQ628752Tegenaria domestica Agelenidae NJNU03003 AY633852 AY633852 NSTegenaria domestica Agelenidae Spagna and Gillespie (2008) DQ628683 DQ628719 DQ628648
DQ628756Textrix denticulata Agelenidae Spagna and Gillespie (2008) DQ628682 DQ628718 DQ628647
DQ628755Amaurobius similis Amaurobiidae Spagna and Gillespie (2008) DQ628663 DQ628699 NS
DQ628736Callobius gertschi Amaurobiidae Spagna and Gillespie (2008) DQ628668 DQ628704 NS
DQ628741Coelotes terrestris (specimen 1) Amaurobiidae Spagna and Gillespie (2008) DQ628688 DQ628724 DQ628651
DQ628761Coelotes terrestris (specimen 2) Amaurobiidae Spagna and Gillespie (2008) DQ628689 DQ628725 DQ628652
DQ628762Eurocoelotes inermis Amaurobiidae Spagna and Gillespie (2008) DQ628690 DQ628726 DQ628653
DQ628763Pimus sp 1 Amaurobiidae Spagna and Gillespie (2008) DQ628675 DQ628711 DQ628642
DQ628748Wadotes dixiensis Amaurobiidae Spagna and Gillespie (2008) DQ628685 DQ628721 NS
DQ628758Pimus sp 2 Amaurobiidae Spagna and Gillespie (2008) DQ628681 DQ628717 DQ628646
DQ628754Metaltella simoni Amphinectidae Spagna and Gillespie (2008) DQ628677 DQ628713 NS
DQ628750Anyphaena californica Anyphaenidae Spagna and Gillespie (2008) DQ628660 DQ628696 DQ628633
DQ628733Zygiella x-notata Araneidae Spagna and Gillespie (2008) DQ628658 DQ628694 NS
DQ628731New genus 4 Cybaeidae JSSC_004 HM576653 HM576638 HM576672New genus 4 Cybaeidae JSSC_012 HM576652 HM576637 HM576671Argyroneta aquatica Cybaeidae Spagna and Gillespie (2008) DQ628687 DQ628723 DQ628650
DQ628760Argyroneta aquatica Cybaeidae ARAQ1 HM576655 NS NSCybaeina sp 1 Cybaeidae JSSC_006 HM576647 HM576631 HM576666Cybaeina sp 2 Cybaeidae JSSC_009 HM576648 HM576632 HM576667Cybaeota sp 2 Cybaeidae JSSC_005 HM576656 NS HM576674Cybaeota sp 3 Cybaeidae JSSC_007 HM576645 HM576629 HM576664Cybaeus morosus Cybaeidae Spagna and Gillespie (2008) DQ628671 DQ628744 DQ628641
DQ628707Cybaeus sp 1 Cybaeidae JSSC_001 NS HM576635 HM576669Cybaeus sp 2 Cybaeidae JSSC_003 HM576654 HM576639 HM576673Cybaeus sp 3 Cybaeidae JSSC_013 HM576651 HM576636 HM576670New genus 1 Cybaeidae JSSC_008 HM576646 HM576630 HM576665Badumna longinqua Desidae Spagna and Gillespie (2008) DQ628665 DQ628701 DQ628637
DQ628738Desis marina Desidae DesisNZ HM576644 HM576628 HM576663Desis formidabilis Desidae DDf1 HM576643 HM576627 HM576662
(continued next page )
256 Invertebrate Systematics J C Spagna et al
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
Appendix 1 (continued )
OTU Family Voucherreference 28S accession 18S accessions Histone H3 accession
Paratheuma armata Desidae Spagna and Gillespie (2008) DQ628674 DQ628710 NSDQ628747
Paratheuma insulana Desidae TCI_045 HM576642 HM576626 HM576661Blabomma sp Dictynidae JSSC_011 HM576649 HM576633 NSDictyna sp Dictynidae Spagna and Gillespie (2008) DQ628673 DQ628709 NS
DQ628746Lathys alberta Dictynidae Spagna and Gillespie (2008) DQ628676 DQ628712 DQ628643
DQ628749Saltonia incerta ndash CA Dictynidae salt_033 HM576657 NS HM576675Saltonia incerta ndash CA Dictynidae salt_053 HM576640 HM576624 HM576659Saltonia incerta ndash NM Dictynidae salt_085 HM576641 HM576625 HM576660Yorima sp Dictynidae JSSC_010 HM576650 HM576634 HM576668Cicurina sp Dicynidae Spagna and Gillespie (2008) DQ628699 DQ628705 DQ628640
DQ628742Stegodyphus sp Eresidae Spagna and Gillespie (2008) DQ628691 DQ628727 DQ628654
DQ628764Zelotes sp 1 Gnaphosidae Spagna and Gillespie (2008) DQ628686 DQ628722 NS
DQ628759Calymmaria sp 1 Hahniidae Spagna and Gillespie (2008) DQ628666 DQ628702 DQ628638
DQ628739Calymmaria sp 2 Hahniidae Spagna and Gillespie (2008) DQ628667 DQ628703 DQ628639
DQ628740Cryphoeca sp Hahniidae Spagna and Gillespie (2008) DQ628672 DQ628708 NS
DQ628614Neoantistea agilis Hahniidae Spagna and Gillespie (2008) DQ628678 DQ628714 DQ628644
DQ628751Alopecosa kochi Lycosidae Spagna and Gillespie (2008) DQ628662 DQ628698 DQ628635
DQ628735Cheiracanthium mildei Miturgidae Spagna and Gillespie (2008) DQ628670 DQ628706 NS
DQ628743Neolana dalmasi Neolanidae Spagna and Gillespie (2008) NS DQ628728 DQ628655
DQ628765Megadictyna thilenii Nicodamidae Spagna and Gillespie (2008) DQ628692 DQ628729 DQ628656
DQ628766Stiphidion facetum Stiphidiidae Spagna and Gillespie (2008) DQ628693 DQ628730 DQ628657
DQ628767Tengella radiata Tengellidae Spagna and Gillespie (2008) DQ628684 DQ628720 DQ628649
DQ628757
Appendix 2 Primers used
Genetic locusprimer name Primer sequence 50 30 Reference
28S28S O ACT GCT CAA AGG TAA ACG G Hedin and Maddison (2001)28S28S C GGT TCG ATT AGT CTT TCG CC Hedin and Maddison (2001)18S18S 1F TAC CTG GTT GAT CCT GCC AGT AG Giribet et al (1996)18S18S 5R CTT GGC AAA TGC TTT CGC Giribet et al (1996)18S18S 5F GCG AAA GCA TTT GCC AAG AA Giribet et al (1996)18S18S 9R ATG GCT CGT ACC AAG CAGACV GC Giribet et al (1996)Histone H3H3aF GCT CGT ACC AAG CAG ACV GC Colgan et al (1998)Histone H3H3aR ATA TCC TTR GGC ATR ATR GTG AC Colgan et al (1998)
Parallelism and aquatic habitats in dictynoid spiders Invertebrate Systematics 257
httpwwwpublishcsiroaujournalsis
![Adaptive Affinity Fields for Semantic Segmentationstellayu/publication/doc/2018aafECCV.pdfAdaptive Affinity Fields for Semantic Segmentation Tsung-Wei Ke* [00000003 1315 3834], Jyh-Jing](https://img.dokumen.tips/doc/110x75/600d21f124b11f24f414f7c9/adaptive-afinity-fields-for-semantic-segmentation-stellayupublicationdoc2018aafeccvpdf.jpg)
![Index []– methods 925–927 affinity purification. See chromatography affinity-selected material analysis 943–944 Affymetrix Integrated Genome Browser 724 AFM. See atomic force](https://img.dokumen.tips/doc/110x75/5f88e9d8f347645f20775d47/index-a-methods-925a927-afinity-puriication-see-chromatography-afinity-selected.jpg)
