Molecular Phylogenetics and Evolution 57 (2010) 1017–1025

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier .com/ locate /ympev

Pulmonate phylogeny based on 28S rRNA gene sequences: A frameworkfor discussing habitat transitions and character transformation

W.E. Holznagel a,�, D.J. Colgan b,⇑, C. Lydeard c

a Department of Biological Sciences, University of Alabama, Box 870345, Tuscaloosa, AL 35487, USAb The Australian Museum, Sydney 2010, Australiac American University, Department of Biology, 4400 Massachusetts Avenue, NW, Washington, DC 20016, USA

a r t i c l e i n f o

Article history:Received 6 October 2009Revised 24 September 2010Accepted 24 September 2010Available online 3 November 2010

Keywords:AmphiboloideaGlacidorbidaePulmonataTerrestrialityFreshwater colonization

1055-7903/$ - see front matter Crown Copyright � 2doi:10.1016/j.ympev.2010.09.021

⇑ Corresponding author. Address: Research BranchCollege St., Sydney N.S.W. 2010, Australia. Fax: +612

E-mail addresses: don.colgan@austmus.gov.au, c(D.J. Colgan).

� Deceased author.

a b s t r a c t

Pulmonate snails occupy a wide range of marine, estuarine, freshwater and terrestrial environments.Non-terrestrial forms are supposed to be basal in pulmonate evolution but the group’s phylogeny isnot well resolved either morphologically or on the basis of available DNA sequence data. The lack of arobust phylogeny makes it difficult to understand character polarization and habitat transformation inpulmonates.

We have investigated pulmonate relationships using 27 new sequences of 28S rRNA from pulmonatesand outgroups, augmented with data from GenBank. The complete alignments comprised about 3.8 kb.Maximum parsimony, maximum likelihood and Bayesian analyses of alignments generated under differ-ent assumptions are reported. Complete alignments appear to have a degree of substitution saturation sowhere there is conflict between hypothesised relationships more weight is given to analyses whereregions of random similarity are excluded and which are not affected by this complication.

Monophyly of the five main pulmonate groups was robustly supported in almost all analyses. The mar-ine group Amphiboloidea and the freshwater Glacidorbidae are the most basal. The remaining pulmo-nates (Siphonariidae, Hygrophila and Eupulmonata) form a moderately-supported monophyletic groupin all analyses bar one probably affected by saturation of substitutions. Siphonariidae, a predominantlymarine and intertidal family, and Eupulmonata (mainly terrestrial with marine, estuarine and freshwaterspecies) form a strongly supported clade that is the sister group to Hygrophila (freshwater).

Multiple colonizations of freshwater and terrestrial habitats by pulmonate snails are suggested. Noanalyses strongly support the possibility of habitat reversions. The colonizations of freshwater by Hygro-phila and of land by Stylommatophora were apparently phylogenetically independent although it cannotyet be excluded that there were transient terrestrial phases in the history of the former group or fresh-water phases in the latter.

Crown Copyright � 2010 Published by Elsevier Inc. All rights reserved.

1. Introduction

Although familiar primarily from their terrestrial representa-tives, pulmonate gastropods occupy an extremely wide range ofhabitats. They can be found living for part or all of the life cyclein intertidal zones on rocky or sandy shores or estuaries, in fresh-waters, in the littoral or as fully terrestrial in damp or even quitedry habitats. The non-terrestrial forms have generally beenthought to be basal in pulmonate evolution (Hubendick, 1945,1978; Morton, 1955a; Nordsieck, 1992). This is broadly supportedby the cladistic analyses of the group based on morphology

010 Published by Elsevier Inc. All r

, The Australian Museum, 69320 6020.olgan2153@optusnet.com.au

(Nordsieck, 1992; Barker, 2001; Dayrat and Tillier, 2002). But thereis little consistency in these studies in the suggested relationships.Many relationships are unresolved. Mordan and Wade (2008)describe the morphological understanding of pulmonate phyloge-netics as ‘‘sketchy”. The incongruence in the distribution of infor-mative character states makes polarising their changes difficult.A large proportion of important characters (such as the possessionof an osphradium in aquatic forms or the location of the eyes attentacle tips in terrestrial species) are expected to be highly adap-tative for particular habitats and consequently liable to homoplas-ious transformations. The significance of such characters, andindeed the pattern of habitat transitions can only be understoodin the context of a robust phylogenetic framework.

The Bouchet and Rocroi (2005) classification of the group basedon Nordsieck’s (1992) system is shown in Table 1. The most spec-iose group is Eupulmonata (sensu Nordsieck, 1992). This includes

ights reserved.

Table 1Classification of the principal candidate pulmonate groups based on Bouchet andRocroi (2005). Common habitats and shell or animal body form are indicated forfamilies, the term ‘‘snail” describing typically coiled shells.

Incertae sedisGlacidorboidea Ponder, 1986

Glacidorbidae Ponder, 1986 – freshwater snailsInformal group PULMONATAInformal group BASOMMATOPHORA Keferstein, 1864

Amphiboloidea Gray, 1840Amphibolidae Gray, 1840 - mangrove snails

Siphonarioidea Gray, 1827Siphonariidae Gray, 1827 - intertidal and subtidal marine limpets

Clade HygrophilaAcroloxoidea Thiele, 1931

Acroloxidae Thiele, 1931 – freshwater limpetsChilinoidea Dall, 1870

Chilinidae Dall, 1870 – freshwater snailsLatiidae Hutton, 1882 – freshwater limpets

Lymnaeoidea Rafinesque, 1815Lymnaeidae Rafinesque, 1815 – freshwater snails and limpets (Lancinae)

Planorboidea Rafinesque, 1815Planorbidae Rafinesque, 1815 - freshwater snails and limpets (Ancylini)Physidae Fitzinger, 1833 – freshwater snails

Clade EUPULMONATAEllobioidea L. Pfeifferi, 1854

Ellobiidae L. Pfeifferi, 1854 - estuarine or terrestrial snailsTrimusculoidea J. Q. Burch, 1845

Trimusculidae J. Q. Burch, 1945 - marine intertidal limpetsOtinoidea Adams & Adams, 1855

Otinidae Adams & Adams, 1855 - marine intertidal snailsSmeagolidae Climo, 1980 – marine meiofaunal slugs

Clade SystellommatophoraOnchidoidea Rafinesque, 1815

Onchidiidae Rafinesque, 1815 - mangroves, marine intertidal, terrestrialor freshwater slugs

Veronicelloidea Gray, 1840Rathouisiidae Heude, 1885 – terrestrial slugsVeronicellidae Gray, 1840 - terrestrial slugs

Clade Stylommatophora Terrestrial snails and slugs

1018 W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025

all families with terrestrial species, but also has marine and estua-rine representatives. Hygrophila, another large group, is dependenton freshwater for reproduction although there are species that asadults can survive for extended periods without immersion.Glacidorbidae is entirely restricted to freshwater but whether thisis a pulmonate group is doubtful. Thalassophila combines Sipho-nariidae that live predominantly in the marine intertidal as adultsand Amphiboloidea that are found mostly in estuaries andmangroves.

Pulmonata is presently included with another traditional group,Opisthobranchia, in Heterobranchia (Haszprunar, 1985, 1988;Ponder and Lindberg, 1997). Heterobranchia also includes a num-ber of families, of which Pyramidellidae is the largest, informallygrouped as ‘‘lower heterobranchs”. The boundaries of Opisthobran-chia, Pulmonata and ‘‘the lower heterobranchs” are unclear(Mikkelsen, 1996, 2002; Dayrat and Tillier, 2002; Grande et al.,2008). The morphological cladistic analysis of Dayrat and Tillier(2002) recovered a monophyletic Pulmonata (excepting Glacidor-bidae) but it is not monophyletic in molecular analyses (Colganet al., 2000, 2003; Dayrat et al., 2001; Grande et al., 2004, 2008;Klussmann-Kolb et al., 2008; Dinapoli and Klussmann-Kolb, 2010).

Klussmann-Kolb et al. (2008) investigated heterobranch phy-logeny using DNA sequences from nuclear (18S rRNA and 28SrRNA) and mitochondrial (16S rRNA and cytochrome c oxidasesubunit I) genes including data from 29 pulmonates in 19 families.In their Bayesian analysis, Klussmann-Kolb et al. (2008) found thatSiphonariidae and the opisthobranch group Sacoglossa formed aclade that was the sister group to a clade comprising Hygrophilaand, as its sister group, the remainder of the pulmonates, a pyrami-dellid and the opisthobranch group Acochlidia. A similar topology

was found by Dinapoli and Klussmann-Kolb (2010) although thenumber of included pulmonates was reduced to 13 since their fo-cus was on lower heterobranchs.

The position of Glacidorbidae, for which sequence data werefirst reported in Dinapoli and Klussmann-Kolb (2010) whilst thispaper was under review, has been controversial. This group isfound only in temperate Australia and southern South America(Ponder and Avern, 2000). Ponder (1986) recognised it as a distinctsuperfamily, possibly representing a freshwater invasion quiteindependent of other gastropods. He considered the group to bepulmonates (Ponder, 1986). This was supported by studies ofsperm ultrastructure (Healy, 1996) and DNA sequence analysesthat place the group as the sister of Pyramidellidae in a derived po-sition amongst pulmonates (Dinapoli and Klussmann-Kolb, 2010)but has been disputed by many morphologists (Bandel, 1997;Dayrat and Tillier, 2002; Haszprunar, 1988). Most current opinionplaces the group in the lower heterobranchs (Barker, 2001;Bouchet and Rocroi, 2005).

Although molecular studies have allotted anomalous positionsto some taxa, this data source appears to offer the best chance ofresolving pulmonate phylogeny given that progress in morpholog-ical analyses is currently slow. Being doubtful of the use of COI and16S rRNA at this phylogenetic level, we have chosen to collect datafrom longer sections of the 28S rRNA gene than have previouslybeen obtained from pulmonates. These data were used to testthe monophyly of proposed higher level groups possibly belongingto Pulmonata and to infer their phylogenetic relationships. Thephylogeny provides a framework for discussing character transfor-mations and habitat transitions in Pulmonata. Possibly, some ofthese transitions are reversals, but which these might be cannotyet be determined.

2. Materials and methods

2.1. Data collection

Details of specimens sequenced for this study, including prove-nance and voucher numbers, are presented in Table 2. Most datawere collected at the University of Alabama. These are identifiedbelow with the letters UA. Data collected at the AustralianMuseum are identified by AM.

Genomic DNA was extracted using a standard CTAB/Proteinase Kmethod (UA) (Saghai-Maroof et al., 1984) or the DNAeasy kit (QIA-GEN, Hilden, Germany) (AM). Approximately half of the 28S rRNADNA (at the 50end) was amplified by Long PCR performed with aTaKaRa LA PCR Kit (Takara Bio USA, Madison, Wisconsin) usingthe 2X GC Buffer II following the vendor’s recommended protocol(UA) and the 28S-F and 28S-R primers (Table 2). The cycling param-eters were one treatment of 94 �C for 2 min 45 s, 30 cycles of dena-turation at 94� for 30 s, annealing at 55� for 30 s and extension at72 �C for 2 min 30 s, followed by a final extension at 72 �C for5 min. DNA amplification of AM material used standard PCR withvarious combinations of the AM or UA series primers in Table 2 togenerate overlapping fragments. Cycling parameters were similarto the UA protocol except that 32 cycles of denaturation, annealingand extension were run, with the extension time reduced to 1 minand variable annealing temperatures generally set to 1–2 �C belowthe lower TM of the primer pair, where this was less than 55� or 3 to4 �C below where the lower TM was above 55 �C.

Long PCR products (UA) were cleaned using a QIAquick GelExtraction Kit (Qiagen) and sequenced directly, using BigDyeTerminator v3.1 Cycle Sequencing Kit by Applied Biosystems, fol-lowing the vendor’s recommended protocol, on an ABI3100 auto-mated sequencer. Both strands of the 28S rRNA gene weresequenced using the primers listed in Table 2. Standard PCRproducts (AM) were purified and sequenced in both directions by

Table 2PCR amplification primers and internal sequencing primers.

Primer Sequence Positiona TM

External primers PCR28S-F: GATCGGACGAGATTACCCGCTGAA 6828S-R: CAGATGTACCGCCCCAGTCAAACT 66Internal sequencing primers28S-F1 GCAGAGCGCTACGATCGRTT 539 6028S-F2 GTGCAAATCGATCGTCAAACTTGG 1075 6528S-F3 CGAATCAACCAGCCCTGAAAATGG 1482 6828S-F4 GAACAATGTAGGTAAGGGAAGTCG 2119 5828S-F5 TACTATCTAGCGAAACCACAGCC 2607 5828S-R1 CCAGTTCTGAGTTGGCTGTTGAATGC 2327 6728S-R2 TGTCCAGGTTCGGGAATATTAACC 1797 6228S-R3 CAGTTCTGCTTACCAAAAGTGGC 1291 6028S-R4 CTTGGTCCGTGTTTCAAGACGG 782 6428S-R5 CATTCCCAAACAACCCGACTCC 235 65Australian Museum Primers28SAM1F: ATCRSTCGGCTCRNGNRTCGATGAAG 66 6228SAM2F: CAAGTACCGTRAGGGAAAGTTG 339 5428SAM4F: GTCTTGAAACACGGACCRMGAAGT 808 5728SAM5F: ACCCGAAAGATGGTGAACTAT 995 5028SAM7F: ANCCGYTAAGGAGTGTGTAACA 1448 5328SAM9F: CAGATCKTRRTRGWAGTAGCAA 1639 5128SAM11F: AAKGYAGGYAAGGGAAGTCGGCAA 2128 5828SAM13F: GGAATCCGACTGTYTAATTAAA 2235 4828SAM3R TCACGYNCTNTTKRACTCTCTATTCA 376 5528SAM4R ACTTCKYGGTCCGTGTTTCAAGAC 784 5728SAM6R GAACCAGCTACTAGATGGTTCGAT 1095 5628SAM8R AGCGYCATCCATTTTMAGGGCT 1469 5528SAM10R CTGTTGACRYGRAAMCCTYNTCCRCTTC 1791 6228SAM12R AGAGCCAATCCTTNTMCCGAAG 2142 5528SAM14R ACATTCKRAGSACTGGGCAGAAATCA 2427 58

a The position numbers refer to the 30 end of the primer in the Marstonia argasequence (DQ256749) obtained here. All primers were designed specifically for thisstudy.

W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025 1019

Macrogen Inc. (Geumchun-Gu, Korea) with the individual primersused to generate the original PCR product (Table 2).

2.2. Data analysis

Sequences were assembled and edited using Sequencher ver-sions 4.7 to 4.9. The newly collected data were augmented withrelevant GenBank accessions where these were more than twothirds of the length of the present sequences (Table 3).

Sequences were compiled in BioEdit (Hall, 1999) and file formatconversions were made with ForCon (Raes and Van de Peer, 1999).Alignments were initially made using the default parameters inClustal X (Thompson et al., 1997). This alignment was divided intotwo sections (at a highly conserved region) and submitted to theM-Coffee web server (Moretti et al., 2007) to generate the opti-mum alignment from a variety of alternative methods. The divisionwas necessary to accommodate the length restrictions currentlyoperating for the web server. The alignments were inspected forevidence of obvious anomalies but none were observed. Aliscore(Misof and Misof, 2009) was applied to the M-coffee alignmentto identify areas of random similarity caused by mutational satura-tion and alignment ambiguity. The default sliding windowsize(w = 6) was used, and gaps were treated as ambiguities (–N op-tion). Gblocks (Castresana, 2000) was used to identify poorlyaligned and divergent regions, employing the webserver (Talaveraand Castresana, 2007) with relaxed conditions allowing smaller fi-nal blocks, gap positions within these and less strict flanking posi-tions. The degree of substitution saturation in the alignments wasassessed using the method of Xia et al. (2003) implemented in theDAMBE package (Xia and Lemey, 2009).

Maximum likelihood analyses were conducted by RAxML(Stamatakis, 2006) using a web server (Stamatakis et al., 2008)with the default parameters. In particular, these assume a GTR

model of evolution and conduct 100 bootstrap replicates using afast search algorithm.

Bayesian analyses were conducted with MrBayes 3.1.2(Huelsenbeck and Ronquist, 2001) used default parameter valuesexcept as specified here. The ‘‘ratepr” was set to variable, a GTRmodel of evolution was modelled by setting nst = 6 and a gammadistribution was used to model variation in substitution rates be-tween base positions in the alignment. The Metropolis-CoupledMarkov Chain Monte Carlo was run with four differentially-heatedchains for ten million ‘generations’, sampling every 1000 genera-tions. Tracer, version 1.4 (Rambaut and Drummond, 2004) wasused to graph the log-likelihoods of sampled trees. The point atwhich chains reached stationarity was judged by visual inspection.Earlier trees in the chains were discarded as ‘‘burn-in”. The remain-ing trees from each of the two runs (the default in MrBayes 3.1.2)were included in calculations of posterior probabilities.

Parsimony analyses were conducted by PAUP version 4.0 b10(Swofford, 2001) assuming the default values for parameters,unless specified below. Heuristic searches with random taxonaddition were conducted for 1000 replicates. For bootstrap pseu-do-resampling, 500 replicates were used, with 20 random taxonaddition replicates in each bootstrap replicate. In analyses of theAliscore and Gblocks alignments, no more than 200 trees greaterthan length 200 were kept in each replicate.

Abbreviations of the reported analyses are: MP – maximumparsimony M-Coffee alignment; BY – Bayesian inference, M-Coffeealignment; ML – maximum likelihood, M-Coffee alignment. Theaddition of ‘-ALI’ to abbreviations indicates analyses in which areasof random similarity as defined by Aliscore (Misof and Misof, 2009)were excluded. ‘-GB’ indicates analyses in which areas of ambigu-ous alignment as defined by Gblocks (Castresana, 2000) were ex-cluded. Bootstrap support percentages are identified below bythe abbreviation BS for both parsimony and likelihood analyses.Posterior probabilities are abbreviated as ‘‘PP”). In describing theresults, PP values over 0.90 are regarded as ‘‘moderate support”,those over 0.95 as ‘‘significant support” and those over 0.975 as‘‘strong support”. Often BS values of 70% and above have been re-garded as strong support. Here, BS values between 65% and 85%are described as moderate support and those over 85% as strongsupport, as recent evaluations (Soltis and Soltis, 2003; Taylor andPiel, 2004) suggest that the assumption that bootstrap valuesgreatly underestimate support is not generally correct.

3. Results

3.1. Analysis characteristics

The M-Coffee alignment had 3783 base positions in total, ofwhich 2156 were constant, 602 variable but parsimony-uninfor-mative and 1025 parsimony-informative. The Aliscore alignmenthad 2142 characters, of which 1382 characters were constant,273 variable but parsimony-uninformative and 487 parsimony-informative. The Gblocks alignment had 2660 characters, of which1662 characters were constant, 328 variable but parsimony-unin-formative and 670 parsimony-informative. The v2 test of thehomogeneity of base composition among taxa returned a P valueof 0.5802 when all characters were included, 1.000 for the Aliscorealignment and 0.999 for the Gblocks alignment.

In the DAMBE analysis of the complete M-coffee alignment, thesubstitution saturation index (ISS) was below or near the criticalvalue (ISSc) for numbers of taxonomic units (NOTU) of 16 or less.For NOTU levels of 32, however, the ISS of 1.045 was notably largerthan the ISSc of 0.809, suggesting substantial saturation in thisalignment when all taxa were included. For the Aliscore alignmentthe ISS at NOTU of 32 (the maximum calculated in DAMBE) was0.581 which is significantly less than the ISSc of 0.793 (P = 0.000)

Table 3Composition of the dataset.

Higher taxon or family Species Author GenBank accession Localitya or referenceb

VetigastrodaLepetodrilidae Lepetodrilus elevatus (Maclean, 1988) AY145413 2NeritopsinaNeritidae Nerita funiculata (Menke, 1851) DQ279976 1CaenogastropodaCyclophoridae Aperostoma palmeri (Bartsch & Morrison, 1942) DQ279983 1Pleuroceridae Juga acutifilosa (Stearns, 1890) DQ256748 Willow Cr., CA USA

Pleurocera canaliculatum undulatum (Say, 1829) DQ256747 Nolichucky R., TN USAHydrobiidae Marstonia arga (Thompson, 1977) DQ256749 Mussel Camp Rd. AL USAMuricidae Bolinus brandaris (Linnaeus, 1758) DQ279986 1Nassariidae Ilyanassa obsoleta (Say, 1822) AY145411 2Heterobranchia‘Lower heterobranchs’Aclididae Graphis sp. FJ917230 4Cimidae Cima sp. FJ917228 4Valvatidae Valvata piscinalis (Müller, 1774) FJ917224 4

Cornirostra pellucida (Laseron, 1954) FJ917225 4Pyramidellidae Boonea seminuda (Adams, 1839) AY145395 2Rissoellidae Rissoella elongatospira (Ponder, 1966) FJ917226 4Acteonoideaunclassified Rictaxis punctocaelatus (Carpenter, 1864) FJ917243 4Glacidorbidae Striadorbis spiralis (Ponder & Avern, 2000) DQ256746 Fitzroy R., VIC Aust

Glacidorbis rusticus (Ponder, 2000) 4OpisthobranchiaCephalaspideaHaminoeidae Haminoea solitaria (Say, 1822) AY145408 2Philinidae Philine aperta (Linnaeus, 1767) DQ279988 1NotaspideaUmbraculidae Umbraculum umbraculum (Lightfoot, 1786) FJ917246 4AplysiomorphaAplysiidae Aplysia californica (Cooper, 1863) AY026366 5PulmonataAmphiboloidea

Phallomedusa solida (Martens, 1878) DQ279991 1Phallomedusa solida (Martens, 1878) HQ156213 AMc Narooma, N.SW., AustPhallomedusa solida (Martens, 1878) HQ156220 AM Middle Harbour, Sydney, NSW, AustSalinator tecta (Golding et al., 2007) HQ156214 AM Salt Pan Creek, Sydney, NSW, AustSalinator tecta (Golding et al., 2007) HQ156215 AM Salt Pan Creek, Sydney, NSW, Aust

ChilinoideaLatiidae Latia neritoides (Gray, 1850) FJ917245 4LymnaeoideaLymnaeidae Austropeplea tomentosa (Pfeiffer, 1855) HQ156217 AM Terrigal, NSW, Aust

Fossaria obrussa (Say, 1825) DQ256737 Mallard Pt., AL USAPhysidae Physa acuta (Draparnaud, 1805) DQ256738 Northport, AL USAPlanorbidae Laevapex fuscus (C.B. Adams, 1841) DQ256734 Nolichucky R., TN USA

Glyptophysa gibbosa (Gould, 1846) DQ256736 Darlott Ck.,VIC AustMicromenetus dilatatus (Gould, 1841) DQ256735 Nolichucky R., TN USA

SiphonarioideaSiphonariidae Siphonaria funiculata (Reeve, 1856) DQ256743 Wapengo Lagoon, NSW Aust

Siphonaria pectinata (Linnaeus, 1758) DQ256744 Sebastian Inlet, FL USAEllobioideaEllobiidae Ophicardelus ornatus (Ferussac, 1821) DQ279994 1

Ophicardelus ornatus (Ferussac, 1821) DQ256740 Nangudga Lake, NSW, AustOphicardelus quoyi (H&A Adams, 1855) DQ256739 Blackfellows Lagoon, NSW, Aust

OtinoideaSmeagolidae Smeagol phillipensis (Tillier & Ponder, 1992) FJ917229 4SystellomatophoraOnchidiidae Onchidella patelloides (Quoy & Gaimard, 1832) HQ156218 AM Georgetown, Tasmania, Aust

Onchidium verruculatum (Cuvier, 1830) DQ256742 Wapengo Lagoon, NSW AustOnchidium damelii (Cuvier, 1830) HQ156219 AM Middle Harbour, Sydney, NSW, Aust

Veronicellidae Vaginulus plebeius (Fisher, 1868) DQ256745 Rio Emilio Carranza, Veracruz, MexStylommatophoraAgriolimacidae Deroceras reticulatum (Müller, 1774) AY145404 2Arionidae Arion silvaticus (Lohmander, 1937) AY145392 2Hygromiidae Prietocella barbarae (Linnaeus, 1758) HQ156216 AM Terrigal, NSW, AustHelicodiscidae Helicodiscus parallelus (Say, 1817) DQ256731 Tuscaloosa, AL USAPolygyridae Praticolella martensi (Pilsbry, 1907) DQ256730 Agua Buena, San Luis Potosi, MexSubulinidae Lamellaxis gracilis (Hutton, 1834) DQ256733 Tuscaloosa, AL USAZonitidae Mesomphix globosus (MacMillan, 1940) DQ256732 Tuscaloosa, AL USA

a Standard state abbreviations are used for the United States of America (USA). Mexico is abbreviated as ‘‘Mex”, Australia as ‘‘Aust”, and its states New South Wales andVictoria as ‘‘NSW” and ‘‘VIC” respectively.

b References are 1: Giribet et al. (2006); 2: Passamaneck et al. (2004); 3: Dong et al. (unpubl.); 4: Dinapoli and Klussmann-Kolb (2010); 5: Medina et al. (2001).c AM indicates a sequence determined at the Australian Museum.

1020 W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025

W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025 1021

suggesting that saturation has little effect on this dataset. A similarobservation was made for the Gblocks alignment where the ISSwas 0.536 compared to an ISSc value of 0.801 (P = 0.000).

The optimised likelihoods for the RAxML analyses were�31425.678 for ML, �14286.504 for ML–ALI and �20077.249 forML–GB.

In BY, the average standard deviation of the splits frequencywas 0.0026. The first one million generations were omitted in eachrun. The average log likelihood of the remaining trees was�3.135E4 in each run and the ESS respectively 3473.161 and3559.122. In BY–ALI, the average standard deviation of the splitsfrequency was 0.0020. The first one million generations were omit-ted for each run and the average log likelihood of the remainingtrees was �1.427E4 in each of the runs and the ESS respectively2238.42 and 2271.81. In BY–GB, the average standard deviationof the splits frequency was 0.0020. The first 500,000 generationswere omitted for each run and the average log likelihood of the

0.1 changes per site

90

59

97

70

80

96

9682

93

86

58

93

5877

8274

9996

94

68

95

81

70

54

89

85

79

94 Cima sp

Boonea seminudaStriadorbis spiraGlacidorbis rustic

Phallomedusa

PhallomeduSalinator tecta

Salinator tectPhallomedus

Aplysia californica Umbraculum umbHaminoea solitaPhiline aperta

Latia neritoides

Laevapex Microm

Glyptophysa

Physa acutaFossaria obAustropeple

Siphonaria pectinataSiphonaria funiculata

Lamellaxis graciPraticolella m

Prietocolella baHelicodiscus paMesomphix glo

Arion sylvaticDeroceras re

Smeagol phillipensisOphicardelus orna

Ophicardelus ornatOphicardelus quoyi

Vaginulus plebOnchidella pa

Onchidium verrucuOnchidium damelii

Eupulmonata

LepeNerita funiculata

MarstoAperostoma palmeri

Pleurocera canaliculatumJuga acutifilosa

Ilyanassa obsoletaBolinus brandarus

Rictaxis punRis

A

a

B

b

C

c

D

d

Fig. 1. The maximum likelihood tree produced by RAxML by analysis of the completeaccession number. Taxonomic groups are indicated by bars to the right of the specimenbelow branches or to the right of nodes according to available space. Nodes highlighted b50 percent support. The topology of analyses of this dataset was identical except that in Mwas moved to d, with PP of 0.52.

remaining trees was �2.003E4 in each of the runs and the ESSrespectively 2959.005 and 2614.437.

MP analyses recovered 15 trees of length 6054, each with a con-sistency index (‘‘CI”) of. 0.454. For MP–ALI, there were 304 trees oflength 2309 and CI 0f 0.489. For MP–GB there were 539 trees oflength 3455 and CI 0.459.

3.2. Clade monophyly

Heterobranchia was monophyletic with strong support in allanalyses. There were differences between analytical methods inthe hypothesized positions of the three pairs of ‘‘lower hetero-branchs” when all alignment positions were included (Fig. 1). In87% of the most parsimonious trees from the MP–GB analysis,the Graphis and Cima species and Cornirostra pellucida and Valvatapiscinalis formed a clade that was not observed in other analyses ofthis alignment (Fig. 2).

Pyramidellidae

Glacidorboidea

Amphiboloidea

OPISTHOBRANCHIA

Hygrophila

Siphonariidae

Stylommatophora

Otinoidea

Ellobioidea

Systellommatophora

“LOWERHETEROBRANCHIA”

Graphis spCorn. pellucida

Valvata piscinalis

lisus solida DQ279991

sa solida HQ156213 HQ156215

a HQ156214a solida HQ156220

raculumria

fuscusenetus dilatatus

gibbosa

russaa tomentosa

lisartensirbararallelusbosum

usticulatum

tus DQ279994us DQ256740

eiustelloideslatum

todrilus elevatus

nia arga

ctocaelatussoella elongatospira

alignment. Multiple individuals of the same species are distinguished by GenBankidentifiers. Bootstrap support values greater than 50 percent are written above or

y filled circles have 100 percent support. Branches without figures received less thanP, branch A in this illustration was moved to a, branch B to b and C to c and in BY, D

0.1 changes per site

0.590.98

0.99

0.610.72

0.98

0.92

0.980.52

0.95

0.96

0.72

0.54

0.61

0.76

0.99

0.80

0.91

0.64

0.51

0.74

0.97

0.85

Pyramidellidae

Glacidorboidea

Amphiboloidea

OPISTHOBRANCHIA

Hygrophila

Siphonariidae

Stylommatophora

Otinoidea

Ellobioidea

Systellommatophora

“LOWERHETEROBRANCHIA”

Lepetodrilus elevatus

Marstonia arga

IIlyanassa obsoletaBolinus brandaris

Aperostoma palmeriPleurocera canaliculatumJuga acutifilosa

Graphis sp.

Nerita funiculata

Valvata piscinalisRictaxis punctocaelatus

Rissoella elongatospira

Boonea seminudaPhallomedusa solida DQ279991

Phallomedusa solida HQ156220Salinator tecta HQ156214

Phallomedusa solida HQ156213

Striadorbis spiralis Glacidorbis rusticus

Aplysia californicaUmbraculum umbraculum

Haminoea solitariaPhiline aperta

Laevapex fuscusMicromenetus dilatatus

Glyptophysa gibbosa

Physa acutaFossaria obrussaAustropeplea tomentosa

Siphonaria pectinataSiphonaria funiculata

Vaginulus plebeiusOnchidella patelloides

Onchidium verruculatumOnchidium damelii

Smeagol phillipensisOphicardelus ornatus DQ279994

Ophicardelus ornatus DQ256940Ophicardelus quoyi

Lamellaxis gracilisPraticolella martensiPrietocolella barbaraHelicodiscus parallelus

Mesomphix globosusArion sylvaticus

Deroceras reticulatum

Cima sp.Corn. pellucida

Salinator tecta HQ156215

Latia neritoides

Eupulmonata

Fig. 2. The Bayesian 50% majority rule consensus tree with mean branch lengths for the alignment excluding ambiguous regions defined by Aliscore. Multiple individuals ofthe same species are distinguished by GenBank accession number. Taxonomic groups are indicated by bars to the right of the specimen identifiers. Posterior probabilityvalues greater than 0.5 are written above or below branches or to the right of nodes according to available space. Nodes highlighted by filled circles have 1.00 posteriorprobability.

1022 W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025

Opisthobranchia was monophyletic in all analyses of the M-Cof-fee alignment with moderate to strong support in the completedataset and BY–ALI (PP = 0.95) and BY–GB (PP = 0.93) but with lessthan 50% bootstrap support for other analyses excluding ambigu-ous data.

Pulmonata (if taken to include Amphiboloidea and Glacidorbi-dae) was weakly contradicted by the inclusion of Opisthobranchiain all analyses. There was moderate support in ML (BS = 70) for aclade comprising Opisthobranchia, Hygrophila, Siphonariidae andEupulmonata. Pulmonate monophyly was also contradicted bythe recovery of Pyramidellidae as the sister group of Glacidorbidaeplus Amphiboloidea in BY (PP = 0.52 only) or as the sister group ofAmphiboloidea only (ML–ALI BS = 85, BY–ALI PP = 0.98, MP–ALIBS = 56).

Eupulmonata, Hygrophila, Stylommatophora, Systellommato-phora (excluding Otinoidea, in the ‘‘strict” sense of Nordsieck,

1992 and sensu Bouchet and Rocroi, 2005) and all families (exceptPhallomedusidae) with two or more representatives were mono-phyletic with strong support in at least some analyses (and gener-ally most, or all) of each of the three alignments (complete, Aliscoreand Gblocks).

Amphiboloidea and Siphonariidae were each monophyletic withmaximum support in all analyses. Notably, however, Amphibolidaeand Phallomedusidae were not separated into monophyletic groups.Thalassophila was never recovered in any analysis.

3.3. Relationships between possible pulmonate clades

Siphonariidae was sister to Eupulmonata in all analyses withconsiderable support in some (ML BS = 81, ML–ALI BS = 91, MP–GB BS = 94, PP = 0.95 for BY–ALI and PP = 1.00 for both BY andBY–GB). Hygrophila was the sister group of (Siphonariidae plus

W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025 1023

Eupulmonata) in all analyses except MP where it was the sister ofOpisthobranchia. There was generally moderate support for theclade including Siphonariidae, Eupulmonata and Hygrophila (MLBS = 70, ML–GB BS = 72, BY PP = 0.68, BY–ALI PP = 0.91 and BY–GBPP = 1.00). In all analyses, except MP–GB, MP–ALI. ML–ALI andBY–ALI, Amphiboloidea and Glacidorbidae were sister groups, of-ten with strong support, for example ML BS = 96). There was nosupport in any analysis for the placement of either or both of thesegroups as the sister group to any clade (except Pyramidellidae) thatdid not include all other pulmonates.

Within Eupulmonata, monophyly of the non-Stylommatophorataxa (resolved as the sister group of Stylommatophora) receivedweak to moderate support in analyes of the complete alignment(ML BS = 68, BY PP = 0.86, MP BS = 55). In analyses of the alignmentexcluding ambiguous data Stylommatophora was resolved as thesister group of Ellobiidae plus Smeagolidae, but with only weaksupport (strongest: BY–ALI PP = 0.61). The sister pair grouping ofSmeagolidae and Ellobiidae was supported, albeit generallyweakly, in all analyses except ML.

There was topological variation between analyses within Hygro-phila. In most analyses Lymnaeidae and Physidae were recovered assister groups with moderate (ML–GB BS = 63, BY–GB PP = 0.80) orstrong (ML BS = 86, BY PP = 0.98, MP BS = 88, MP–ALI BS = 100) sup-port. In these analyses, this pair of families was the sister group ofPlanorbidae with strong support. In the two remaining analyses,however, Physidae was recovered as the sister group to allHygrophila except Latiidae (ML–ALI BS = 82, BY–ALI PP = 1.00).

4. Discussion

4.1. Clade membership and inter-relations in the Pulmonata

The phylogenetic analyses here suggest that Pulmonata as pres-ently understood comprises five groups, two that are basal(Amphiboloidea and Glacidorbidae), and three that are more derived(Hygrophila, Siphonariidae and Eupulmonata). Monophyly of thesefive groups is also generally supported (where taxonomic represen-tation allows this to be tested) in morphological (Nordsieck, 1992;Barker, 2001) or molecular investigations (Klussmann-Kolb et al.,2008; Dinapoli and Klussmann-Kolb, 2010).

DAMBE (Xia et al., 2003) analyses suggest that there may be adegree of substitution saturation in the complete alignment butthose from which potentially ambiguous regions were excludedare not apparently affected by this complication. There are somedifferences in the topological relationships suggested by analysesof the various alignments. These may be due to the larger amountsof data in the complete alignments and/or the level of saturationassociated with this. In cases of conflict, however, more weight ishere accorded the results from analyses that are free of saturationeffects.

If supported by other data, the moderate support here for theplacement of Opisthobranchia in a clade that excludes Amphiboloi-dea and Glacidorboidea but includes Hygrophila, Siphonariidaeand Eupulmonata may require re-definition of Pulmonata. Previousauthors have also questioned the membership of Pulmonata. Mostinclude Amphiboloidea (e.g. Nordsieck, 1992; Dayrat et al., 2003;Bouchet and Rocroi, 2005), but only some add Glacidorbidae(Ponder and Avern, 2000). The group is placed in the ‘‘lower het-erobranchs” in the morphologically-based phylogenies of Barker(2001) and Dayrat and Tillier (2002). Klussmann-Kolb et al.(2008) and Dinapoli and Klussmann-Kolb (2010) both findPyramidellidae and the opisthobranch groups Sacoglossa andAcochlidia, which have generally been regarded as opisthobranchs(Neusser et al., 2006; Sommerfeldt and Schrödl, 2005; Wägeleet al., 2008), in the smallest monophyletic clade includingEupulmonata, Hygrophila and Siphonariidae. Pyramidellidae was

included owing to its close association, also supported in manyof our analyses, with Amphiboloidea which occupied a derived po-sition in Klussmann-Kolb et al. (2008).

The basal position of Amphiboloidea and Glacidorbidae with re-spect to the remaining pulmonates is supported by the morpholog-ical cladistic analyses of Barker (2001). Glacidorbidae is basal butAmphiboloidea is not resolved within an anomalous derived groupincluding Otinidae, Latiidae, and Ellobiidae in Dayrat and Tillier(2002). Previous molecular analyses place Siphonariidae (withthe Sacoglossa) as the sister group of the rest of the Pulmonatas.l. (Klussmann-Kolb et al., 2008) with high posterior probabilitybut little bootstrap support. This placement is rejected in all anal-yses here, generally by strong support values for alternativerelationships.

Siphonariidae and Eupulmonata are shown as sister groups tothe exclusion of Hygrophila in all analyses, except the parsimonyanalysis of the complete alignment which we consider liable to sat-uration effects. This contrasts with recent morphological classifica-tions in which Hygrophila and Siphonariidae are supposed tobelong to a clade excluding Eupulmonta (Nordsieck, 1992) orHygrophila and Eupulmonata have been supposed to be more clo-sely related to each other than either is to Siphonariidae (Barker,2001; Dayrat and Tillier, 2002). Some molecular analyses(Klussmann-Kolb et al., 2008; Dinapoli and Klussmann-Kolb,2010) suggest a closer affinity of Hygrophila and Eupulmonatathan between either of these and Siphonariidae. In these, however,the interpolation of other taxa (opisthobranchs, pyramidellids andamphiboloids) suggests that the relationship between Hygrophilaand Eupulmonata is quite distant.

The current conception of Eupulmonata (Nordsieck, 1992;Bouchet and Rocroi, 2005) which adds Systellommatophora tothe group (comprising Stylommatophora, Ellobioidea, Trimusculoi-dea and Otinoidea) proposed by Haszprunar and Huber (1990) issupported by most molecular analyses (Klussmann-Kolb et al.,2008; Dinapoli and Klussmann-Kolb, 2010, and here. Among mor-phological analyses, it was anomalously contradicted by the re-moval of Otinoidea in Barker (2001) and Ellobiidae and Otinidaein Dayrat and Tillier, 2002). Within Eupulmonata, the monophylyof Stylommatophora is well-supported here and in previous analy-ses (Wade et al., 2001, 2006; Klussmann-Kolb et al., 2008; Dinapoliand Klussmann-Kolb, 2010). There was no strong support for theidentification of its sister group in the present analyses. InKlussmann-Kolb et al. (2008), Dinapoli and Klussmann-Kolb(2010), Stylommatophora is resolved as the sister group to all othereupulmonates whereas in morphological analyses (e.g. Barker,2001; Dayrat and Tillier, 2002), it occupies a derived position asthe sister group to only some other eupulmonates. This is sup-ported by characters in Dayrat and Tillier’s (2002) analysis suchas the loss of heterostrophy, the addition of an unpaired jaw andthe presence of a pedal gland. Nordsieck (1992) found two possiblesynapomorphies (one of which, ‘‘shell size reduction”, is of doubt-ful phylogenetic value) associating Otinoidea with Systellommato-phora. Here, with weak support in saturation-free analyses,and with stronger support in previous molecular analyses(Klussmann-Kolb et al., 2008; Dinapoli and Klussmann-Kolb,2010), Otinoidea has been recovered as the sister group of Ellobii-dae. It probably belongs to Eupulmonata but its sister group is notyet definitely resolved.

There was support in all analyses except BY–ALI and ML–ALI forthe sister group pairing of Physidae and Lymnaeidae, a possibilitysuggested by morphological investigations (Barker, 2001) and,without statistical support, analyses based on cytochrome c oxi-dase subunit I sequences (Remigio and Hebert, 2001). Bouchetand Rocroi (2005) do not suggest reasons for supporting an alter-native hypothesis that includes Physidae in the superfamilyPlanorboidea as suggested by Hubendick (1978). Albrecht et al.

1024 W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025

(2007) have dismissed this inclusion on molecular grounds. Theexceptions in ML–ALI and BY–ALI, and their conflict with analysesof the Gblocks alignments, show that more research is needed onrelationships within Hygrophila.

Although our analyses represent the collection of a largely inde-pendent and novel data set for examining pulmonate relationshipsthey are based on a single gene. Klussmann-Kolb et al. (2008) andDinapoli and Klussmann-Kolb (2010) have previously used 28SrRNA in multigene analyses of such relationships. Those analysesapparently offer stronger support at lower phylogenetic levels(particularly within Eupulmonata). However, at higher levels, sub-stitution saturation in the relatively rapidly evolving genes COI and16S rRNA may be obscuring the phylogenetic signal of 28S rRNA inthese studies, particularly in regards to the lowering of bootstrapsupport values for some of the relationships of Siphonariidae,Amphiboloidea and Glacidorboidea.

4.2. Character evolution

The broad congruence (and occasional divergences) of the pres-ent analyses allows discussion of their implications for polarizingcharacter states during morphological evolution. For example, thegenerally close association (as sister groups) of Amphiboloideaand Glacidorbidae could reduce the number of major morphologicaltransformationa that are required to explain their possession ofadult opercula and the lack of this feature in any other pulmonates.Because of the basal position of these taxa, the character polaritywill finally require consideration in the context of the overall distri-bution of adult opercula in Heterobranchia which is known to besporadic, and does not include Pyramidellidae (Dayrat et al., 2003).

The eyes of Ellobiidae are at the bases of their cephalic tentacles incontrast to the placement at the tips of tentacles in Onchidiidae, Vag-inulidae and Stylommatophora (Mordan and Wade, 2008). The eyesof Otinidae are also at the base of the tentacles (Morton, 1955b). Gi-ven the topological relationships shown in molecular analyses(Klussmann-Kolb et al., 2008; Dinapoli and Klussmann-Kolb, 2010;here), eye position in Otinidae and Ellobiidae could represent areversion or there may have been independent evolution of tentacletip placement in Systellommatophora and Stylommatophora.

Contractile pneumostomes are found in all Eupulmonata (Bar-ker, 2001; Mordan and Wade, 2008) but otherwise the conditionis observed only in the Siphonariidae in which both contractileand non-contractile forms occur (Barker, 2001). Contractility is as-sumed to be adaptive to terrestrial environments but the occur-rence of this state in Otinoidea and Siphonariidae suggests it mayalso be advantageous in other environments. The distribution ofthis character possibly supports the association of Siphonariidaeand Eupulmonata in the present analyses, although this would im-ply that it can also be lost or that Siphonariidae is paraphyletic tothe Eupulmonata, which is an unlikely scenario.

The morphological association of Siphonaria and Amphiboloi-dea in the Thalassophila (Hubendick, 1945; van Mol, 1967; Nord-sieck, 1992) was principally based in the cladistic analysis ofNordsieck (1992) on supposed shared secondary monauly (fusionof the male and female reproductive tracts extending from a com-mon genital opening). Recent studies have shown that theamphiboloid genus Phallomedusa is diaulic (Golding et al., 2007),suggesting that within this clade the evolution of monauly is inde-pendent from that in Siphonariidae and Stylommatophora (pre-sumed secondary as diauly is the condition for Ellobiidae,Trimusculidae, Onchidiidae, etc. Mordan and Wade, 2008).

4.3. Habitat transitions

Previous phylogenetic studies have suggested that the evolu-tion of eupulmonate terrestriality has occurred multiple times

(Nordsieck, 1992; Barker, 2001). Morton (1955a) and Barker(2001) suppose that there has been repeated development of ter-restriality even within Ellobiidae. Multiple origins of terrestrialityare supported by the present analyses, if reversions to aquaticenvironments are not hypothesized, and the monophyly of familiesoccurring in multiple habitat types is assumed where representa-tives are scored from only a subset of habitats. Independent originsare suggested for Stylommatophora, Vaginulidae, the terrestrialOnchidiidae, and the terrestrial Ellobiidae. If there were reversionsto other habitat types, a single origin of terrestriality is possible un-der the assumption that reversions to the marine habitat occurredin Smeagolidae, Ellobiidae (multiple times) and Onchidiidae (prob-ably multiple times, depending on the yet to be determined posi-tion of the terrestrial forms within the family’s phylogeny), andto freshwater by the two Onchidiidae found in this environment(Dayrat, 2009). Habitat reversal to this extent appears improbable.Moreover, both Melampinae and Onchidiidae, at least, have free-living veliger larvae (Barker, 2001) and whilst this form is probablyadvantageous, its repeated homoplasious evolution appears unli-kely, arguing against habitat reversion. It remains possible thatEupulmonata has had multiple origins of terrestriality and alsosome habitat reversals, but the latter at intra-familial scales.

Most freshwater snails in the present analyses belong to themonophyletic Hygrophila, but the results strongly support the sug-gestion that the occupation of this habitat by Glacidorbidae repre-sents an independent colonization. Onchidiidae exhibit at least oneother independent colonization of freshwater by pulmonates. Atleast two colonizations have occurred in Acochlidia (Strong et al.,2008) suggested to be pulmonates, as mentioned above, byKlussmann-Kolb et al. (2008). Although only one freshwater colo-nization (by Hygrophila) has led to a major radiation, the numberof such events in pulmonates, suggests that this group as well asother gastropods (Lydeard et al., 2002; Strong et al., 2008), has sig-nificant potential, at least over evolutionary time, for such habitattransitions.

Whether there were fresh or brackish water stages in the evolu-tion of the major terrestrial lineages or terrestrial stages in the evo-lution of the Hygrophila remains controversial (Mordan and Wade,2008). The present analyses do not discount either possibilityalthough if they did occur they were merely transient, as neitheris represented by extant species. Our analyses do suggest, however,that habitat transitions in these two large clades were indepen-dent. If this were not the case, the marine habit of Siphonariidaewould be implied to be a reversion from either a freshwater or aterrestrial condition.

The phylogenetic approach based on extant taxa has clarifiedpatterns of habitat transition in Pulmonata to some extent butscope undoubtedly remains for the investigation of physiologyand biochemistry regarding the possibility that there have beentransient habitat phases in Hygrophila or Stylommatophora. Forexample, although there is not a strong correlation of uric acid withthe degree of terrestriality in tested species (Russell Hunter, 1964),Andrews (1988), suggests that the excretion of this metabolite bysome Hygrophila implies a terrestrial phase during their evolution.If this not a phylogenetically-recent adaptation for the limited ter-restrial exposure of adult Hygrophila, it may well an informativephysiological relic.

Acknowledgments

We gratefully acknowledge support from the U.S. National Sci-ence Foundation to CL, the Department of Biological Sciences, Uni-versity of Alabama to CL and the Australian Museum to DC. Wethank Michael Shea and Rosemary Golding for identifying someAM specimens. Comments from reviewers on previous versionsof the manuscript have been very helpful.

W.E. Holznagel et al. / Molecular Phylogenetics and Evolution 57 (2010) 1017–1025 1025

References

Albrecht, C.A., Kuhn, K., Streit, B., 2007. A molecular phylogeny of Planorboidea(Gastropoda, Pulmonata): insights from enhanced taxon sampling. Zool. Scripta36, 27–39.

Andrews, E.B., 1988. Excretory systems of molluscs. In: Wilbur, K.M. (Ed.), TheMollusca, vol. II. San Diego, Academic Press, pp. 381–448.

Bandel, K., 1997. Higher classification and pattern of evolution of the Gastropoda. Asynthesis of biological and paleontological data. Cour. Forschungsinst.Senckenb. 201, 57–81.

Barker, G.M., 2001. Gastropods on land: phylogeny, diversity and adaptivemorphology. In: Barker, G.M. (Ed.), The Biology of Terrestrial Mollusca.Oxford, CABI Publishing;, pp. 1–146.

Bouchet, P., Rocroi, P. (with Fryda, J., Hausdorf, B., Ponder, W., Valdes, A., Warén, A.)2005. Classification and nomenclator of gastropod families. Malacologia 47, 1–397.

Castresana, J., 2000. Selection of conserved blocks from multiple alignments fortheir use in phylogenetic analysis. Mol. Biol. Evol. 17, 540–552.

Colgan, D.J., Ponder, W.F., Eggler, P.E., 2000. Gastropod evolutionary rates andphylogenetic relationships assessed using partial 28S rDNA and histone H3sequences. Zool. Scripta 29, 29–65.

Colgan, D.J., Ponder, W.F., Beacham, E., Macaranas, J.M., 2003. Molecularphylogenetic studies of Gastropoda based on six gene segments representingcoding or non-coding and mitochondrial or nuclear DNA. Molluscan Res. 23,159–178.

Dayrat, B., 2009. Review of the current knowledge of the systematics of Onchidiidae(Mollusca: Gastropoda: Pulmonata) with a checklist of nominal species.Zootaxa 2068, 1–26.

Dayrat, B., Tillier, A., Lecointre, G., Tillier, S., 2001. New clades of euthyneurangastropods (Mollusca) from 28S rRNA sequences. Mol. Phylogen. Evol. 19, 225–235.

Dayrat, B., Tillier, S., 2002. Evolutionary relationships of euthyneuran gastropods(Mollusca): a cladistic re-evaluation of morphological characters. Zool. J. Linn.Soc. 135, 403–470.

Dayrat, B., Tillier, S., 2003. Goals and limits of phylogenetics – the euthyneurangastropods. In: Lydeard, C., Lindberg, D. (Eds.), Molecular Systematics andPhyleogeography of Mollusks. Smithsonian Books, Washington, pp. 161–184.

Dinapoli, A., Klussmann-Kolb, A., 2010. The long way to diversity – phylogeny andevolution of the Heterobranchia (Mollusca: Gastropoda). Mol. Phylogen. Evol.55, 60–76.

Giribet, G., Okusu, A., Lindgren, A.R., Huff, S.W., Schrodl, M., Nishiguchi, M.K., 2006.Evidence for a clade composed of molluscs with serially repeated structures:Monoplacophorans are related to chitons. Proc. Natl. Acad. Sci. USA 103, 7723–7728.

Golding, R.E., Ponder, W.F., Byrne, M., 2007. Taxonomy and anatomy ofAmphiboloidea (Gastropoda: Heterobranchia: Archaeopulmonata). Zootaxa1476, 1–50.

Grande, C., Templado, J., Cervera, J.L., Zardoya, R., 2004. Molecular phylogeny ofEuthyneura (Mollusca: Gastropoda). Mol. Biol. Evol. 21, 303–313.

Grande, C., Templado, J., Zardoya, R., 2008. Evolution of gastropod mitochondrialgenome arrangements. BMC Evol. Biol. 8, 61.

Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor andanalysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98.

Haszprunar, G., Huber, G., 1990. On the central nervous system of Smeagolidae andRhodopidae, two families questionably allied to the Gymnomorpha(Gastropoda, Euthyneura). J. Zool. 220, 185–199.

Haszprunar, G., 1985. Heterobranchia – a new concept of the phylogeny of thehigher Gastropoda. Z. Zool. Syst. Evolutionsforsch 23, 15–37.

Haszprunar, G., 1988. On the origin and evolution of major gastropod groups, withspecial reference to the Streptoneura. J. Molluscan Stud. 54, 367–441.

Healy, J.M., 1996. Molluscan sperm ultrastructure: correlation with taxonomic unitswithin the Gastropoda, Cephalopoda and Bivalvia. In: Taylor, J.D. (Ed.), Originand Evolutionary Radiation of the Mollusca. Oxford Press, Oxford, pp. 99–113.

Hubendick, B., 1945. Phylogenie und Tiergeographie der Siphonariidae.zur Kenntnisder Phylogenie in der Ordnung Basommatophoraund des Ursprungs derPulmonatengruppe. Zool. Bid. Upp. 24, 1–216.

Hubendick, B., 1978. Systematics and comparative morphology of theBasommatophora. In: Fretter, V., Peake, J. (Eds.), Pulmonates. SystematicsEvolution and Ecology. Academic Press, New York, vol. 2, pp. 1–47.

Huelsenbeck, J.P., Ronquist, F., 2001. MrBAYES; Bayesian inference for phylogeny.Bioinformatics 17, 754–755.

Klussmann-Kolb, A., Dinapoli, A., Kuhn, K., Streit, B., Albrecht, C., 2008. From sea toland and beyond – new insights into the evolution of Euthyneuran Gastropoda(Mollusca). BMC Evol. Biol. 8, 1–16.

Lydeard, C., Holznagel, W.E., Glaubrecht, M., Ponder, W.F., 2002. Molecularphylogeny and evidence for multiple origins of freshwater gastropods of thecircum-global, diverse superfamily Cerithioidea (Mollusca: Caenogastropoda).Mol. Phylogenet. Evol. 22, 399–406.

Medina, M., Collins, A.G., Silberman, J.D., Sogin, M.L., 2001. Evaluating hypotheses ofbasal animal phylogeny using complete sequences of large and small subunitrRNA. Proc. Natl. Acad. Sci. USA 98, 9707–9712.

Mikkelsen, P.M., 1996.The evolutionary relationships of Cephalaspideas. l.(Gastropoda: Opisthobranchia): a phylogenetic analysis. Malacologia 37, 375–442.

Mikkelsen, P.M., 2002. Shelled opisthobranchs. Adv. Mar. Biol. 2, 67–136.

Misof, B., Misof, K., 2009. A Monte Carlo approach successfully identifiesrandomness in multiple sequence alignments: a more objective means ofdata exclusion. Syst. Biol. 58, 21–34.

Mordan, P.B., Wade, C.M., 2008. Heterobranch phylogeny 2: the pulmonates. In:Ponder, W.F., Lindberg, D. (Eds.), Phylogeny and Evolution of the Mollusca.University of California Press, San Francisco, pp. 409–426.

Moretti, S., Armougom, F., Wallace, I.M., Higgins, D.G., Jongeneel, C.V., Notredame,C., 2007. The M-Coffee web server: a meta-method for computing multiplesequence alignments by combining alternative alignment methods. NucliecAcids Res. 35 (Web Server issue), W645-8.

Morton, J.E., 1955a. The evolution of the Ellobiidae with a discussion on the origin ofthe Pulmonata. Proc. Zool. Soc. Lond. 25, 127–168.

Morton, J.E., 1955b. The functional morphology of Otina otis, a primitive marinepulmonate. J. Mar. Biol. Ass. UK 34, 113–150.

Neusser, T.P., Heß, M., Haszprunar, G., Schrödl, M., 2006. Computer-based three-dimensional reconstruction of the anatomy of Microhedyle remanei (Marcus,1953), an interstitial acochlidian gastropod from Bermuda. J. Morph. 267, 231–247.

Nordsieck, H., 1992. Phylogeny and system of the Pulmonata. Arch. Mollusken 121,31–52.

Passamaneck, Y.J., Schander, C., Halanych, K.M., 2004. Investigation of molluscanphylogeny using large-subunit and small-subunit nuclear rRNA sequences. Mol.Phylogenet. Evol. 32, 25–38.

Ponder, W.F., 1986. Glacidorbidae (Glacidorbiacea: Basommatophora), a new familyand superfamily of operculate freshwater gastropods. J. Linn. Soc. Lond. 87, 53–83.

Ponder, W.F., Avern, G.J., 2000. The Glacidorbidae (Mollusca: Gastropoda:Heterobranchia) of Australia. Rec. Aust. Mus. 52, 307–353.

Ponder, W.F., Lindberg, D.L., 1997. Towards a phylogeny of gastropod molluscs: ananalysis using morphological characters. Zool. J. Linn. Soc. 119, 83–265.

Raes, J., Van de Peer, Y., 1999. ForCon: a software tool for the conversion of sequencealignments. EMBnet.news 6(1). Available at http://vulcan.rug.ac.be/~jerae/ForCon/index.html.

Rambaut, A., Drummond, A.J., 2004. Tracer 1.3, Available at http://beast.bio.ed.ac.uk/Tracer.

Remigio, E., Hebert, P.D.N., 2001. Testing the utility of partial COI sequences forphylogenetic estimates of gastropod relationships. Mol. Phylogenet. Evol. 29,641–647.

Russell Hunter, W., 1964. Physiological aspects of ecology in nonmarine molluscs.In: Wilbur, K.M., Yonge, C.M. (Eds.), Physiology of Mollusca, vol. I. New York,Academic Press, pp. 83–126.

Saghai-Maroof, M.A., Soliman, K.M., Jorgensen, R.A., Allard, R.W., 1984.Ribosomal DNA spacer length variation in barley: Mendelian inheritance,chromosomal location and population dynamics. Proc. Natl. Acad. Sci. USA81, 8018–8021.

Sommerfeldt, N., Schrödl, M., 2005. Microanatomy of Hedylopsis ballantinei, a newinterstitial acochlidian gastropod from the Red Sea, and its significance forphylogeny. J. Molluscan Stud. 71, 153–165.

Soltis, P.S., Soltis, D.E., 2003. Applying the bootstrap in phylogeny reconstruction.Stat. Sci. 18, 256–267.

Stamatakis, A., 2006. RAxML-VI-HPC: Maximum likelihood-based phylogeneticanalyses with thousands of taxa and mixed models. Bioinformatics 22, 2688–2690.

Stamatakis, A., Hoover, P., Rougemont, J., 2008. A rapid bootstrap algorithm for theRAxML web-servers. Syst. Biol. 75, 758–771.

Strong, E.E., Gargominy, O., Ponder, W.F., Bouchet, P., 2008. Global diversity ofgastropods (Gastropoda; Mollusca) in freshwater. Hydrobiologia 595, 149–166.

Swofford, D.L., 2001. PAUP*: Phylogenetic Analysis Using Parsimony (and othermethods). Version 4_10b. Sinauer Associates. Sunderland, Massachusetts.

Talavera, G., Castresana, J., 2007. Improvement of phylogenies after removingdivergent and ambiguously aligned blocks from protein sequence alignments.Syst. Biol. 56, 564–577.

Taylor, D.J., Piel, W.H., 2004. An assessment of accuracy, error, and conflict withsupport values from genome-scale phylogenetic data. Mol. Biol. Evol. 21, 1534–1537.

Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmougin, F., Higgins, D.G., 1997. TheClustalX windows interface. flexible strategies for multiple sequence alignmentaided by quality analysis tools. Nucleic Acids Res. 24, 4876–4882.

Van Mol, J.-J., 1967. Étude morphologique et phylogénétique du ganglion cérébroïdedes gastéropodes pulmonés (mollusques). Mém. Acad. R. Belgique, Cl. Sci. 37, 1–168.

Xia, X., Xie, Z., Salemi, M., Chen, L., Wang, Y., 2003. An index of substitutionsaturation and its application. Mol. Phylogenet. Evol. 26, 1–7.

Xia, X., Lemey, P., 2009. Assessing substitution saturation with DAMBE. In: Lemey,P., Salemi, M., Vandamme, A.-M. (Eds.), The Phylogenetic Handbook: A PracticalApproach to DNA and Protein Phylogeny, 2nd ed. Cambridge, CambridgeUniversity Press, pp. 615–630.

Wade, C.M., Mordan, P.B., Clarke, B., 2001. A phylogeny of the land snails(Gastropoda: Pulmonata). Proc. R. Soc. Lond. B 268, 413–422.

Wade, C.M., Mordan, P.B., Naggs, F., 2006. Evolutionary relationships among thepulmonate land snails and slugs (Pulmonata, Stylommatophora). Biol. J. Linn.Soc. 87, 593–610.

Wägele, H., Klussmann-Kolb, A., Vonnemann, V., Medina, M., 2008. HeterobranchiaI: the Opisthobranchia. In: Ponder, W.F., Lindberg, D. (Eds.), Phylogeny andEvolution of the Mollusca. University of California Press, San Francisco, pp. 383–406.