• Cylindrus obtusus ;
  • phylogeny;
  • phylogeography;
  • Alpine land snails;
  • shell morphology;
  • histone genes;
  • Cytochrome c oxidase subunit 1 ;
  • rRNA


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The phylogenetic relationships among genera of the subfamily Ariantinae (Pulmonata, Helicidae), especially the sister-group relationship of Cylindrus obtusus, were investigated with three mitochondrial (12S rRNA, 16S rRNA, Cytochrome c oxidase subunit I) and two nuclear marker genes (Histone H4 and H3). Within Ariantinae, C. obtusus stands out because of its aberrant cylindrical shell shape. Here, we present phylogenetic trees based on these five marker sequences and discuss the position of C. obtusus and phylogeographical scenarios in comparison with previously published results. Our results provide strong support for the sister-group relationship between Cylindrus and Arianta confirming previous studies and imply that the split between the two genera is quite old. The tree reveals a phylogeographical pattern of Ariantinae with a well-supported clade comprising the Balkan taxa which is the sister group to a clade with individuals from Alpine localities. Additional lineages representing samples from southern Alpine localities as well as from Slovakia split from more basal nodes, but their relationships are not clearly resolved. To achieve more definitive conclusions concerning the geographical origin of Ariantinae, still more sequence data are needed to obtain a tree with better resolution of basal nodes. The genetic data also provided new insights concerning the genus Cepaea, which was used as one of the outgroup taxa. Cepaea vindobonensis is only distantly related to Cepaea nemoralis and Cepaea hortensis, the latter two being more closely related to Eobania vermiculata. Thus, in our tree, the genus Cepaea is paraphyletic.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The phylogenetic relationships of and within the land snail subfamily Ariantinae (Pulmonata, Helicidae), especially the closest relative of Cylindrus obtusus1, have been discussed for a long time. The taxon Ariantidae was introduced first by Mörch (1864) in the rank of a family of Pulmonata. Schileyko (1991) defined Ariantinae as a subfamily of Helicidae and characterized it mainly by shell morphology and genital anatomy. According to Schileyko (2006), the shells of the Ariantinae are mainly globular to flat, and the mucus glands are tubular, simple or biramous. Steinke et al. (2004) classified the Ariantinae as sister group of the Helicinae based on sequence data from two mitochondrial (mt) genes (COI, 16S rRNA) and two nuclear (nc) genes (ITS-1, 18S rRNA). C. obtusus is one of the most conspicuous species within the Ariantinae. It is an endemic of the Eastern Alps in Austria (Klemm 1974) and is mainly restricted to high elevations (1600–2500 m – comp. Bisenberger et al. 1999). The species displays several interesting particulars regarding its biology and ecology (Kühnelt 1937; Sattmann et al. 1995; Freitag and Desch 1996; Bisenberger et al. 1999; Duda et al. 2010), anatomy of reproductive organs (Martens 1895; Sturany and Wagner 1915; Schileyko 1996; Schileyko et al. 1997; Zopp 2012), population genetics and reproduction biology (Luise Kruckenhauser, Michael Duda, Daniela Bartel et al., unpublished). The systematic position of C. obtusus was discussed controversially in the past. It was originally placed in the genus Pupa (as P. obtusa) by Draparnaud (1805) because of its cylindrical shell. The formerly broadly defined genus Pupa includes various members of the superfamily Pupillioidea according to modern classifications (Bouchet et al. 2005). Martens (1895) gave a first anatomical description and placed it ‘closer to Helix […] than Buliminus or Pupa’ (translated after Martens 1895). Sturany and Wagner (1915) published an enhanced drawing and a short description of the anatomy and consequently allocated C. obtusus within the subfamily Campylaeinae. Boettger and Wenz (1921) also placed C. obtusus in the subfamily Campylaeinae, which comprises most of the genera of current Ariantinae sensu Schileyko (2006). Consequently, Schileyko (2006) assigned C. obtusus to the newly erected monotypic tribus Cylindruini within the subfamily Ariantinae. The current systematic assignment is mainly based on the anatomy of reproductive organs. However, C. obtusus shows a conspicuously aberrant shell form within this group – all other representatives display globular or more or less depressed shells (Kerney et al. 1983). In fact, C. obtusus might be considered as one of the striking cases in which a member of a family has ‘crossed the adaptive valley’ between the two height/width categories (i.e. long versus globular) (see also Gittenberger 2010). The exceptional shell morphology of C. obtusus raised questions about the evolutionary history of the species and its relationships within Ariantinae. Schileyko (2012) summarized the hitherto discussion and hypothesized that the species was very young and originated within its present range at the end of the Würm glaciation splitting from a representative of the Ariantinae due to a mutation. This opinion is largely in accordance with former hypotheses by Boettger (1949) and contrary to Adensamer (1937, 1962), who considered C. obtusus as a representative of an old lineage, which survived the ice age in high alpine habitats. If Schileyko (2012) was right, Cylindrus would exemplify an extremely fast speciation event and would most probably render Arianta (which displays rather high intraspecific genetic distances; Haase et al. 2003; Haase and Misof 2009) paraphyletic.

Although a number of molecular genetic analyses dealing with different taxonomic levels within Helicoidea have been published (Wade et al. 2001, 2007; Steinke et al. 2004; Koene and Schulenburg 2005; Manganelli et al. 2005; Fiorentino et al. 2008, 2010; Groenenberg et al. 2011; Gómez-Moliner et al. 2013), the subfamily Ariantinae has not been investigated extensively so far. An exception to this is a number of studies on Arianta arbustorum (Haase et al. 2003; Gittenberger et al. 2004; Haase and Misof 2009), a widespread species frequently occurring in the Alps and often found syntopically with C. obtusus. A first molecular phylogenetic analysis of the Ariantinae based on fragments of the mt genes Cytochrome b (cytb), Cytochrome c oxidase subunit 1 (COI) and 16S rRNA (16S), as well as of the nc gene for Histone 3 (H3), was performed by Groenenberg et al. (2012). In that study, the combined tree of all marker sequences displayed Cylindrus as the sister group of Arianta. The phylogenetic relationships among other genera of Ariantinae were, however, not well resolved. In a parallel study, our group analysed the phylogenetic relationships of Cylindrus within the Ariantinae using two additional marker sequences, the mt 12S rRNA gene (12S) and the nc Histone H4 gene (H4) besides 16S, COI and H3. Here, we present phylogenetic trees based on these five marker sequences and discuss the position of C. obtusus and phylogeographical scenarios in comparison with the results of Groenenberg et al. (2012).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information


Samples included both fresh specimens recently collected for the analyses (2005–2011) and old, ethanol-preserved material of the mollusc collection of the Natural History Museum of Vienna (NHMW), collected between 1920 and 1991. The present work is not issued for purposes of zoological nomenclature and is not published within the meaning of the ICZN (see Art 8.2). In general, we used the nomenclature of Bank (2012); however, in cases of some taxa, we followed the same practice as Groenenberg et al. (2012), and we used provisional names referring to a taxonomic revision of the Balkan Ariantinae by Péter Subai (in prep.). A total of 48 specimens belonging to 29 land snail species (21 genera) were included in this study (Table 1). In addition to the representatives of Ariantinae, we included species of the subfamily Helicinae (Cepaea nemoralis, Cepaea hortensis, Ceapea vindobonensis, Eobania vermiculata), as well as representatives of two other families, which were used as outgroup: Euomphalia strigella, Monacha cantiana (Hygromiidae) and Helicodonta obvoluta (Helicodontidae; Table 1). Freshly collected animals were prepared according to Kruckenhauser et al. (2011) and subsequently stored in 80% ethanol. For DNA extraction, a tiny piece was cut off the snail's foot without damaging the shell or the rest of the body. Voucher specimens are stored in the mollusc collection of the NHMW; DNA and tissue samples are stored in the tissue collection of the NHMW (Table 1).

Table 1. Specimens used for the phylogenetic study and the sequences analysed. Voucher numbers of the mollusc collection of the NHMW are provided. Numbers in the first column (individual) refer to numbers in the trees. * = from these individuals the COI sequence could not be obtained
Species/IndividualGeographical originVoucher number
Arianta arbustorum (Linnaeus, 1758) 1AT, Lower Austria, Maria SchutzNHM109000/AL/00500/2098
Arianta arbustorum/2AT, Upper Austria, Höllengebirge MtsNHM109000/AL/00499/2102
Arianta chamaeleon (L. Pfeiffer, 1868) 1*SI, Kamniško-Savinjske AlpsNHMW90396
Arianta schmidti (Rossmässler, 1836) 1SI, Kamniško-Savinjske AlpsNHMW90401
Campylaea illyrica (Stabile, 1864) 1SI, Kamniško-Savinjske AlpsNHMW90785
Cattania cf. faueri (Subai, 1990) 1GR, Kentriki Makedonia, Agio PnevmaNHM109000/AL/00501/6998
Cattania cf. faueri/2GR, Kentriki Makedonia, Agio PnevmaNHM109000/AL/00501/6999
Cattania haberhaueri (Sturany, 1897) 1RS, Pčinjski okrug, Bosilegrad, Božička RekaNHM109000/AL/00502/7013
Cattania haberhaueri 2RS, Pčinjski okrug, Bosilegrad, Božička RekaNHM109000/AL/00502/7014
Causa holosericea (Studer, 1820) 1AT, Lower Austria, DürrensteinNHM109000/AL/00504/2106
Causa holosericea 2AT, Upper Austria, Dachstein MtNHM109000/AL/00503/2107
Chilostoma achates cingulina (Deshayes, 1839) 1AT, Styria, Leobener MtNHM109000/AL/00507/2100
Chilostoma achates cingulina 2AT, Salzburg, Hochtor MtsNMHW90823
Chilostoma achates cingulina 3AT, Lower Austria, Sonnwendstein MtNHM109000/AL/00506/2103
Chilostoma cingulatum baldense (Rossmässler, 1839) 1*GE, Bayern, Bad StaffelsteinNMHW87065
Cylindrus obtusus (Draparnaud, 1805) 1AT, Styria, Totes Gebirge, ElmNHM109000/AL/00008/87
Cylindrus obtusus 2AT, Lower Austria, Dürrenstein MtNHM109000/AL/00352/164
Dinarica pouzolzii (Deshayes, 1830) 1ME, Herceg Novi, RoseNHM109000/AL/00508/6557
Faustina faustina (Rossmässler, 1835) 1SK, Žilinský kraj, Malá Fatra MtsNHM109000/AL/00510/2110
Faustina faustina 2*CZ, Moravskolezský Kraj, ZimroviceNMHW74422
Helicigona lapicida (Linnaeus, 1758) 1AT, Lower Austria, Gars am KampNHM109000/AL/00511/2104
Isognomostoma isognomostomos (Schröter, 1784) 1AT, Lower Austria, Maria SchutzNHM109000/AL/00515/2099
Isognomostoma isognomostomos 2AT, Salzburg, SalzbergNHM109000/AL/00514/2105
Josephinella byshekensis (Knipper, 1941) 1AL, Gjirokastër, PacoriNHM109000/AL/00516/7015
Josephinella byshekensis/2AL, Gjirokastër, PacoriNHM109000/AL/00516/7018
Josephinella hemonica (Thiesse, 1884) 1GR, Kentriki Makedonia, VeroiaNHM109000/AL/00517/6568
Josephinella hemonica 2*GR, EpirusNMHW102948
Kosicia intermedia (Férussac, 1832) 1IT, Friuli-Venezia Giulia, Passo di Monte Croce CarnicoNHM109000/AL/00520/2108
Kosicia intermedia 2AT, Carinthia, BärentalNHM109000/AL/00519/2114
Kosicia intermedia 3AT, Carinthia, TscheppaschluchtNHM109000/AL/00518/2115
Kosicia ziegleri (Rossmässler, 1836) 1SI, Kamniško-Savinjske AlpsNHMW90400
Liburnica cf. dunjana (Knipper, 1941) 1AL, Dibër, ShkopetNHM109000/AL/00521/6565
Liburnica setosa setosa (Férussac, 1832) 1HR, Splitsko-dalmatinska županija, Blato na CetiniNHM109000/AL/00522/6566
Thiessea cf. sphaeriostoma (Bourguignat, 1857) 1GR, Sterea Ellada, Euboea, ParalimniNHM109000/AL/00526/6560
Thiessea cf. sphaeriostoma 2GR, Sterea Ellada, Euboea, MantoudiNHM109000/AL/00525/6562
Vidovicia coerulans (C. Pfeiffer, 1828) 1HR, Ličko-senjska županija, Velebit MtsNHM109000/AL/00525/6562
Vidovicia coerulans 2HR, Ličko-senjska županija, Velebit MtsNHM109000/AL/00527/6458
Vidovicia coerulans 3HR, Ličko-senjska županija, Velebit MtsNHM109000/AL/00527/6459
Outgroup species
 Cepaea hortensis (O.F. Müller, 1774) 1ES, Spanish PyreneesNHMW87004
 Cepaea nemoralis (Linnaeus, 1758) 1ES, AsturiasNHMW87010
 Cepaea vindobonensis (C. Pfeiffer, 1828) 1SK, Trenčiansky kraj, Strážovské vrchyNHM109000/AL/00505/2109
 Cepaea vindobonensis 2AT, Lower Austria, PerchtoldsdorfNHMW51761
 Eobania vermiculata (O.F. Müller, 1774) 1GR, Notio Egeo, Makri islandNHMW88046
 Euomphalia strigella (Draparnaud, 1801) 1AT, Lower Austria, Hundsheimerkogel MtNHM109000/AL/00509/2096
 Helicodonta obvoluta (O.F. Müller, 1774) 1AT, Lower Austria, Maria SchutzNHM109000/AL/00513/2097
 Helicodonta obvoluta 2AT, Lower Austria, HeiligenkreuzNHM109000/AL/00512/2101
 Monacha cantiana (Montagu, 1803) 1IT, Friuli-Venezia Giulia, Passo di Monte Croce CarnicoNHM109000/AL/00524/2112
 Monacha cantiana 2SI, Bovec, Soča valleyNHM109000/AL/00523/2113

DNA extraction, molecular markers, amplification and sequencing

Genomic DNA was extracted using the First-DNA all tissue kit (Gen-ial, Troisdorf, Germany) following the manufacturer's protocol. A negative control was used for each set of samples extracted. The final elution volume was 30 μl. A total of five molecular markers were used: partial sequences of three mt genes, COI, 12S and 16S as well as of two nc genes, H3 and H4. These genomic regions were amplified and sequenced with the primers listed in Table 2. Annealing temperatures and resulting amplified fragment sizes are also listed in Table 2. Polymerase chain reactions (PCRs) were performed on a Mastercycler gradient thermal cycler (Eppendorf, Hamburg, Germany) in a final volume of 25 μl, containing 2.5 μl of PCR buffer, 3 mM of Mg2+, 0.2 mM of each nucleotide, 0.5 U of Roche Taq Polymerase and 0.5 μM of each primer. Optimal amounts of template DNA were determined empirically (2–5 μl of the DNA solution). PCR profiles comprised an initial heating step at 94°C for 3 min followed by 35 cycles (40 cycles in some old samples): 30 s at 94°C, 30 s at annealing temperature (Table 2) and 60 s at 72°C. After the last cycle, a final extension of 7 min at 72°C was performed. Control reactions to detect potential contaminations were carried out with: (1) control ‘extractions’ (without sample) instead of the template and (2) with distilled water instead of the template. PCR products were purified using the QIAquick PCR Purification kit (Qiagen, Hilden, Germany) and sequenced in both directions at LGC Genomics (Berlin, Germany). In a few cases, cloning was necessary, for which PCR products were generated with Phusion High Fidelity DNA Polymerase (Finnzymes), extracted from agarose gels with the QIAquick Gel Extraction Kit (Qiagen, Inc.) and subsequently cloned using the TOPO TA Cloning Kit (Invitrogen, Carlsbad, CA, USA). Sequencing of the clones (both directions) was also performed at LGC Genomics.

Table 2. Primers used to amplify and sequence three mt and two nc genes. For H3, an alternative reverse primer was used in some cases. Alignment lengths of 12S and 16S are given for the two options used to trim ambiguous regions (‘no gap’/‘strict’) (see text)
Region sequenced and primer namesPrimer sequences (5′–3′)Ta (°C)Fragment (and alignment) length (bp)
  1. Ta, annealing temperatures.






55628–653 (441/398)






55332–356 (283/250)






48/50654 (654)









316 (316)







57258 (258)

Data analysis

Editing of sequences was performed using BioEdit v. 7.0.1 (Hall 1999). Sequences were aligned using ClustalX (Larkin et al. 2007) with default settings and inspected by eye using Either BioEdit v. 7.0.1 or (as this software was already mentioned before) only BioEdit. The 12S and 16S alignments contained long and varied gaps among the different taxa, which made the alignments in these regions difficult and unreliable. To remove the poorly aligned stretches, we performed an automated trimming procedure with the software TrimAl 1.3 (Capella-Gutiérrez et al. 2009), implemented in the Phylemon 2.0 web tool (Sánchez et al. 2011). Lengths of alignments are provided in Table 2. We tested two different settings the software provides: ‘no gap’ and ‘strict’. The ‘no gap’ method deletes all columns in the alignment with at least one gap in them. The ‘strict’ method involves a first trimming step based on the distribution of gaps in the alignment, followed by a second trimming step based on the similarity of the remaining aligned sequence (see Capella-Gutiérrez et al. 2009 for details). The ‘strict’ option yielded in general the best trees (i.e. with the highest node support). In the (few) cases where the ‘no gap’ option yielded better support, we provide the corresponding data (support values, mt tree, respectively; see 'Results'). Also, for the COI data set, we tested two options, one with the complete sequences and another one where the third codon positions were coded according to the IUPAC nucleotide ambiguity codes as purine (R) or pyrimidine (Y) residuals, respectively (‘COI3rdRY’ coding), to reduce the saturation problem, because saturation affects mostly 3rd codon positions.

It has to be mentioned that the H3 sequences of C. vindobonensis are highly diverged from the other taxa. We repeated the amplification and sequencing of both individuals and obtained the same sequences. As there are no frameshift mutations and no peculiar amino acid substitutions, we assume that this is the authentic H3 sequence and no pseudogene. Furthermore, we computed the trees both including and excluding C. vindobonensis sequences, but the resulting topologies were the same.

The substitution models for the Bayesian inference (BI) of phylogenetic relationships, selected with jModelTest 0.1.1. (Posada 2008) and applying the Akaike information criterion, were as follows: 12S (both no gap and strict): nst = 6, rates = gamma; 16S (no gap): nst = 2, rates = gamma; 16S (strict): nst = 6, rates = gamma; COI (both for complete sequence and ‘COI3rdRY’): nst = 6, rates = gamma; H3: nst = 2, rates = gamma; H4: nst = 6, rates = equal. BI trees were calculated using MrBayes 3.1.2 (Huelsenbeck and Ronquist 2001) using two runs (each with three heated chains and one cold chain) and 5 million MCMC generations (sample frequency 100). Convergence of runs was assessed by visual inspection of plotted log-likelihood values using the software Tracer 1.4 (Rambaut and Drummond 2007). In a conservative approach, 25% of the sampled trees were discarded as burn-in, although the stationary phase in the two runs was reached before. We calculated BI trees using the various options for 12S, 16S and COI as well as combining the three mt and the two nc markers and finally all five marker sequences into one comprehensive alignment. A maximum likelihood (ML) tree was calculated with Garli 2.0 (Zwickl 2006) using the same alignments as for the combined BI tree (H3, H4, 12S ‘strict’, 16S ‘strict’ and COI3rd position RY coded). The substitution models obtained from jModelTest were applied to each of the five data partitions, and a bootstrap search was performed with 500 replicates. The starting trees were calculated using the stepwise algorithm. The trees were rooted using Euomphalia strigella, Monacha cantiana (Hygromiidae) and Helicodonta obvoluta (Helicodontidae). The COI sequence could not be obtained for four individuals. Therefore, they were not used for the concatenated alignment, which, accordingly, comprised 44 sequences. The 12S, 16S H3 and H4 sequences of the remaining four individuals cluster close with their conspecifics (data not shown). The sequences determined in the course of the present study are registered under the following GenBank accession numbers: COI: KF596870KF596913, 12S: KF596774KF596821, 16S: KF596822KF596869, H3: KF596914KF596961, H4: KF596962KF597009.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

The BI tree based on the five concatenated marker sequences using the ‘strict trimming option for 12S and 16S and the ‘COI3rdRY coding for COI provided in general the highest node support values. This tree comprising 44 individuals is presented in Fig. 1. Using other options resulted in the same overall topology, albeit with less support. There were only two nodes that provided higher support with the ‘no gap’ option for 12S and 16S. The corresponding values are also indicated in Fig. 1. In addition, performing a ML analysis with the partitioned data set did not result in better resolved nodes, either (Fig 1). Regarding the analysis of the mt data set, the tree with the highest support values was based on the alignments of the entire COI and the 12S and 16S with all gaps excluded (‘no gap’ option; Figure S1; Supporting information). The tree based on concatenated nc sequences is shown in Figure S2.


Figure 1. Bayesian tree showing the phylogenetic relationships within Ariantinae and related taxa. The tree was calculated with the concatenated sequences of 12S-16S-COI-H3-H4. Support values correspond to the analysis where the ‘strict’ option was used to trim the 12S and 16S sequences, and the ‘COI3rdRY’ option was used for COI (see text). Support values of the maximum likelihood (ML) analyses with the partitioned data set are also depicted (nodes that were absent in the ML analysis are indicated by ‘-’). Furthermore, two nodes obtained higher support in the analysis with the ‘no gaps’ option (indicated in brackets). The general geographical locations where the specimens were collected are provided

Download figure to PowerPoint

The comprehensive tree based on all concatenated sequences (Fig. 1) is divided into three clades, two of them containing species not belonging to Ariantinae: the representatives of the Hygromiidae and Helicodontidae, which were used to root the tree (clade A, black in Fig. 1), and the species of the subfamily Helicinae (B; green), which appear as the sister group of Ariantinae. Within clade B, C. vindobonensis and the sister taxa C. nemoralis + C. hortensis are separated by surprisingly high distances. Eobania vermiculata is more closely related to the latter two rendering Cepaea paraphyletic. This result is highly supported and also observed in both the concatenated mt and nc genes trees, as well as in most of the single gene trees (except for 16S and COI; data not shown). The species of the subfamily Ariantinae (C; red, orange and blue in Fig. 1) are divided into three main clades: (1) Two species of Kosicia (intermedia and ziegleri) from the Southern Alps and Faustina faustina (from the Northern Carpathians). This clade (orange) obtained high support only in the tree where the ‘no gap’ option was used for the mt 12S and 16S genes. It is also found in the concatenated mt and nc gene trees (highly supported in the latter). (2) The second group (blue) is highly supported and comprises taxa distributed in the Balkan region: Josephinella, Thiessea, Cattania, Dinarica, Liburnica and Vidovicia. This clade is also found in the concatenated mt tree, but only when the ‘no gap’ option was used. Within this Balkan clade, most nodes are highly supported, for example the sister-group relationship between Josephinella and Thiessea and their clustering with Cattania as well as the clustering of Liburnica and Dinarica. (3) The remaining taxa are combined in a clade (red) with very low support, which is also found in the concatenated mt tree. It contains specimens from Alpine localities: Chilostoma achates, Helicigona lapicida, Causa holosericea, Isognomostoma isognomostomos, Campylea illyrica and the highly supported sister genera Arianta and Cylindrus. Within this ‘Alpine’ clade, most internal nodes obtained maximum support, and there are three well-supported lineages: a subclade combining Arianta + Cylindrus, another one comprising C. achates, H. lapicida, C. holosericea, and I. isognomostomos, and finally C. illyrica as a separate lineage.

In the tree based on the concatenated nc sequences (Figure S2), some relationships are very well supported, such as the Kosicia + Faustina clade and the differentiation between Helicinae and Ariantinae. However, other groups within the Ariantinae appear disrupted and obtained only low support.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Our comprehensive tree based on five marker sequences provides strong support for the sister-group relationship between Cylindrus and Arianta confirming the results of Groenenberg et al. (2012). In addition, the tree presented here reveals a phylogeographical pattern within Ariantinae with some well-supported clades each combining either western (mainly Alpine and Carpathian) or eastern (Balkan) taxa, respectively. The distances between lineages imply several quite old radiations. However, there is still some ambiguity in the clustering of deeper branches. The hypothesis of Groenenberg et al. (2012) that the origin of Ariantinae, which they dated about 30 million years ago (mya), lies in the Balkan Peninsula is not supported in our analysis, as the species with a distribution in the Balkan region are united within one clade that does not split from the base of the Ariantinae. Yet, the even lower support values in the trees of Groenenberg et al. (2012) do not lend credence to the assumption of a Balkan origin either. In spite of the high amount of sequence data in both investigations (Groenenberg et al. 2012; this study), it seems that additional sequences will be necessary to resolve the deeper nodes within Ariantinae and to test any presumed phylogeographical pattern.

Concerning the lineages of Cylindrus and Arianta, the high distance value implies that this split is quite old (e.g. mean p-distance for COI between Cylindrus and Arianta is 20%, corresponding to a corrected distance of 24% – Tamura-Nei correction), which seems to be in contrast to the hypothesis of Schileyko (2012), who assumed a very young speciation of Cylindrus. Schileyko's argumentation was mainly based on the lack of fossils of C. obtusus and the fact that no other Pleistocene fossils resembling C. obtusus in shell morphology (which could be considered as a possible ancestor or sister taxon) are known. However, it needs to be emphasized that the environmental dynamics in these Alpine regions, which are characterized by changing conditions (temperature, humidity, wind, earthmoving, physical and chemical conditions of the environment, etc.), offers bad opportunities for shell fossilization. Especially, the black soil of the preferred habitat of Cylindrus is acidic and not appropriate for the fossilization of calcite shells. Thus, the lack of fossils in the Alpine region (which is also an argument against using this species for clock calibration; see below) is not a good argument to postulate a recent origin of Cylindrus. In any case, the supposed conflicting assumptions about old versus young origin of Cylindrus are not really contradicting, as the tree of extant species provides information about the split of lineages, but not about speciation or origin of the phenotype of C. obtusus. The latter could only be inferred from the fossil record.

Groenenberg et al. (2012) propose 17.5 my for the split between Arianta and Cylindrus. Yet, one has to keep in mind that a well-resolved tree is a prerequisite for dating splits. Although the distances between lineages within the subfamily Ariantinae are quite high, it remains open whether the datings of Groenenberg et al. (2012) are reasonable estimates. The fossils used for calibration and the low resolution of trees (probably partly due to sequence saturation in some genes) render their molecular clock approach still speculative. Using fossils of C. obtusus for calibration is not advisable, as almost no fossils of this species at all have been found up to now. There are only some rare and mostly young findings within the extant distribution range in the Eastern Alps (Frank 2006). Frank (2006) named five localities with fossil Cylindrus that she dated as late glacial (‘Spätglazial’), during the Würm glaciation (‘Würmeiszeitlich’) or young Holocene (‘Jüngstholozän’). However, no scientific dating is proved in Frank (2006) for these C. obtusus findings, and it is most probable that the existing fossils are not older than Würm. As the other datings used by Groenenberg et al. (2012) cover a very wide time range, those fossils are probably a too weak basis to reliably calculate the age of the split between Cylindrus and Arianta.

A common practice in the absence of reliable fossil records and/or well-defined biogeographical events to calibrate a molecular clock is to apply fixed substitution rates (Wilke et al. 2009). However, the variety of rates reported in the literature is very high, ranging from the very conservative 1% ‘standard mitochondrial clock rate’ (e.g. Avise 2000; substitution rate for protein coding mt genes) to rates that are an order of magnitude higher (e.g. Pons et al. 2010; Kotsakiozi et al. 2012). Considering the 24% COI sequence divergence between Arianta and Cylindrus, the inferred divergence time would – depending on the rate applied – range between 1.2-12 mya, which is not quite informative.

A problem concerning the systematics of Ariantinae is that the taxonomic assignment of species and subspecies is often based mainly on geographical information (sample localities and presumed distribution ranges) rather than on diagnostic traits. Our results imply that additional molecular data (i.e. additional markers) as well as samples from a broader geographical range, complementary to this and previous studies, are needed. This would allow to establish a phylogenetic scaffold to search for further diagnostic traits, to assign clades to described taxa and finally to assess the distribution ranges of the genetic groups. This will subsequently help to disentangle the complicated phylogeny of Ariantinae. Embedded in a sound phylogeny of Helicidae and combined with a higher number of reliable fossil data such a comprehensive analysis would allow to attempt a conclusive molecular clock analysis of this group.

The genetic data presented here revealed new insights concerning the outgroup taxa investigated. The distinct position of C. vindobonensis and the sister-group relationship of C. nemoralis + C. hortensis with E. vermiculata, rendering the genus Cepaea paraphyletic, require broader geographic sampling for a reasonable taxonomic solution. Thus, a comprehensive phylogenetic analysis of the genera Cepaea, Eobania and other possibly related taxa is intended to resolve their relationships.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Financial support for this study was given by the Austrian Science Fund (FWF): project number P 19592 B17 and by the Synthesys Project (AT-TAF-3166), which is financed by the European Community Research Infrastructure Action under the FP7 ‘Capacities’ Program, as well as by the ‘Friends of the Natural History Museum Vienna’. For assistance in the lab, we thank Barbara Däubl. We are grateful for collecting help to Helmut Baminger, Agnes Bisenberger, Barbara Däubl, Doris Kleewein, Ira Richling and Sabine Zwierschitz. Many thanks to Peter Reischütz and Alexander Reischütz who provided samples from the Balkans and to Anita Eschner who supported this study by making available the samples from the mollusc collection of the NHMW. For fruitful discussions, we are indebted to Anatoly Schileyko, Hans Kothbauer, Renate Kothbauer and especially Wilhelm Pinsker, who also critically overhauled the manuscript.

  1. 1

    Schileyko (2012) suggests using Cochlopupa obtusa (Draparnaud, 1805) instead of Cylindrus obtusus, because he claims that (1) Cylindrus Fitzinger, 1833 is the junior homonym of Cylindrus Bartsch, 1789 (see also Dubois and Bour 2010 and, and (2) Cylindrus Fitzinger, 1833 is a junior objective synonym of Cochlopupa Jan, 1830. However, according to Edmund Gittenberger (pers. comm.) the assignment of the name Cylindrus Batsch, 1789 is still disputed. Until this question is settled, we stick to the well-established name, C. obtusus.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Adensamer W (1937) Cylindrus obtusus (Draparnaud 1805), seine relikthafte Verbreitung und geringe Variabilität, sowie zoogeographisch-phylogenetische Betrachtungen über alpine Gastropoden überhaupt. Arch Molluskenkunde 69:66115.
  • Adensamer W (1962) Weitere Betrachtungen über Cylindrus obtusus mit seinem auffallend einheitlichen Artbild. Arch Molluskenkunde 91:6770.
  • Avise JC (2000) Phylogeography. Harvard University Press, Cambridge, MA.
  • Bank R (2012) Gastropoda. Fauna Europaea version 2.5. Available from
  • Bisenberger A, Baumgartner G, Kleewein D, Sattmann H (1999) Untersuchungen zur Populationsökologie von Cylindrus obtusus (DRAPARNAUD, 1805) (Pulmonata, Helicidae). Ann Naturhist Mus Wien 101B:453464.
  • Boettger CR (1949) Die Entstehung der Schalengestalt der ostalpinen Landschnecke Cylindrus obtusus (Draparnaud). Arch Molluskenkunde 78:157158.
  • Boettger CR, Wenz W (1921) Zur Systematik der zu den Helicidensubfamilien Campylaeinae und Helicinae gehörigen tertiären Landschnecken. Arch Moll 53:655.
  • Bouchet P, Rocroi J-P, Frýda J, Hausdorf B, Ponder W, Valdés Á, Warén A (2005) Classification and nomenclator of gastropod families. Malacologia 47:1397.
  • Capella-Gutiérrez S, Silla-Martínez JM, Gabaldón T (2009) trimAl: a tool for automated alignment trimming in large-scale phylogenetic analyses. Bioinformatics 25:19721973.
  • Dubois A, Bour R (2010) The distinction between family-series and class-series nomina in zoological nomenclature, with emphasis on the nomina created by Batsch (1788, 1789) and on the higher nomenclature of turtles. Bonn Zool Bull 57:149171.
  • Duda M, Kruckenhauser L, Haring E, Sattmann H (2010) Habitat requirements of the pulmonate land snails Trochulus oreinos oreinos and Cylindrus obtusus endemic to the Northern Calcareous Alps, Austria. eco. mont-Journal on Protected Mountain Areas Research 2:512.
  • Fiorentino V, Salomone N, Manganelli G, Giusti F (2008) Phylogeography and morphological variability in land snails: the Sicilian Marmorana (Pulmonata, Helicidae). Biol J Linnean Soc 94:809823.
  • Fiorentino V, Salomone N, Manganelli G, Giusti F (2010) Historical biogeography of Tyrrhenian land snails: the Marmorana–Tyrrheniberus radiation (Pulmonata, Helicidae). Mol Phylogenet Evol 55:637.
  • Frank C (2006) Plio-pleistozäne und holozäne Mollusken Österreichs. Teil 2. Vienna. Verlag der Österreichischen Akademie der Wissenschaften, Wien, pp 604610.
  • Freitag B, Desch W (1996) Cylindrus obtusus (Draparnaud 1805) – Futterpflanzen, neue Fundorte und Assoziationen mit anderen Gehäuseschnecken, sowie Vergleiche mit früheren Gehäusevermessungen. (Gastropoda: Helicidae). Mitt dtsch malakozool Ges 58:116.
  • Gittenberger E (2010) Changes in shell shape. In: Bakels C, Fennema K, Out WA, Vermeeren C (eds), Of Plants and Snails: A Collection of Papers Presented to Wim Kuijper in Gratitude for Forty Years of Teaching and Identifying. Sidestone Press, Leiden, pp 6369.
  • Gittenberger E, Piel WH, Groenenberg DSJ (2004) The Pleistocene glaciations and the evolutionary history of the polytypic snail species Arianta arbustorum (Gastropoda, Pulmonata, Helicidae). Mol Phylogenet Evol 30:6473.
  • Gómez-Moliner BJ, Elejalde AM, Arrébola JR, Puente AI, Martínez-Ortí A, Ruiz A, Madeira MJ (2013) Molecular phylogeny of the Helicodontidae and Trissexodontidae (Gastropoda). Zool Scr 42:170181.
  • Groenenberg DSJ, Neubert E, Gittenberger E (2011) Reapraisal of the “Molecular phylogeny of Western Palaearctic Helicidae s.l. (Gastropoda: Stylommatophora)”: When poor science meets GenBank. Mol Phylogenet Evol 61:914923.
  • Groenenberg DSJ, Subai P, Gittenberger E (2012) Molecular phylogeny of the Ariantinae (Gastropoda: Pulmonata: Helicidae), a new approach to an old problem. In: Groenenberg DSJ Molecular taxonomy and natural history collections. Doctoral Thesis, Leiden University, pp 117169.
  • Haase M, Misof B (2009) Dynamic gastropods: stable shell polymorphism despite gene flow in the land snail Arianta arbustorum. J Zool Syst Evol Res 47:105114.
  • Haase M, Misof B, Wirth T, Baminger H, Baur B (2003) Mitochondrial differentiation in a polymorphic land snail: evidence for Pleistocene survival within boundaries of permafrost. J Evol Biol 16:415428.
  • Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp Ser 41:9598.
  • Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17:54755.
  • Kerney MP, Cameron RAD, Jungbluth JH (1983) Die Landschnecken Nord-und Mitteleuropas. Paul Parey, Hamburg und Berlin 384 pp.
  • Klemm W (1974) Die Verbreitung der rezenten Landgehäuseschnecken in Österreich. Denkschriften der Österreichischen Akademie der Wissenschaften, mathematisch- naturwissenschaftliche Klasse 117:1503.
  • Koene JM, Schulenburg H (2005) Shooting darts: co-evolution and counter-adaptation in hermaphroditic snails. BMC Evol Biol 5:25.
  • Kotsakiozi P, Parmakelis A, Giokas S, Papanikolaou I, Valakos ED (2012) Mitochondrial phylogeny and biogeographic history of the Greek endemic land-snail genus Codringtonia Kobelt 1898 (Gastropoda, Pulmonata, Helicidae). Mol Phylogenet Evol 62:681692.
  • Kruckenhauser L, Harl J, Sattmann H (2011) Optimized drowning procedures of pulmonate landsnails allowing subsequent DNA analysis and anatomical dissections. Ann Naturhist Mus Wien 112B:173175.
  • Kühnelt W (1937) Biologische Beobachtungen an Cylindrus obtusus. Arch Molluskenkunde 69:5256.
  • Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23:29472948.
  • Manganelli G, Salomone N, Giusti F (2005) A molecular approach to the phylogenetic relationships of the western palaearctic Helicoidea (Gastropoda: Stylommatophora). Biol J Linnean Soc 85:501512.
  • Martens EV (1895) Die Gattung Cylindrus FITZ. Arch Naturgeschichte 59:103108.
  • Mörch OAL (1864) Fortegnelse over de i Danmark forekommende land- og ferskvandsblöddyr. Videnskabelige Meddelelser fra den Naturhistoriske Forening i Kjöbenhavn 1863: 265367.
  • Pons J, Ribera I, Bertranpetit J, Balke M (2010) Nucleotide substitution rates for the full set of mitochondrial protein-coding genes in Coleoptera. Mol Phylogenet Evol 56:796807.
  • Posada D (2008) jModelTest: phylogenetic model averaging. Mol Biol Evol 25:12531256.
  • Rambaut A, Drummond AJ (2007) Tracer v1.4, Available from
  • Sánchez R, Serra F, Tárraga J, Medina I, Carbonell J, Pulido L, de María A, Capella-Gutíerrez S, Huerta-Cepas J, Gabaldón T, Dopazo J, Dopazo H (2011) Phylemon 2.0: a suite of web-tools for molecular evolution, phylogenetics, phylogenomics and hypotheses testing. Nucleic Acids Res 39:470474.
  • Sattmann H, Kleewein D, Baminger H (1995) Aufenthaltsorte und Aktivität von Arianta arbustorum und Cylindrus obtusus im Warscheneckgebiet. In: Sattmann H, Bisenberger A, Kothbauer H (eds), Arianta, Verlag des Naturhistorischen Museums Wien, Wien, pp 4548.
  • Schileyko AA (1991) Taxonomic status, phylogenetic relations and system of the Helicoidea sensu lato (Pulmonata). Arch Molluskenkunde 120:187236.
  • Schileyko AA (1996) On the peculiar features of the reproductive anatomy of Cylindrus obtusus (DRAPARNAUD, 1821) (Pulmonata, Helicidae). In: Sattmann H, Bisenberger A, Kothbauer H (eds). Arianta II:3738.
  • Schileyko AA (2006) Treatise on Recent terrestrial pulmonate molluscs. Pt. 13. Helicidae, Pleurodontidae, Polygyridae, Ammonitellidae, Oreohelicidae, Thysanophorida. Ruthenica, Supplement 2:17651906.
  • Schileyko AA (2012) On the origin of Cochlopupa (= Cylindrus auct.) obtusa (Gastropoda, Pulmonata, Helicidae). Ruthenica 22:93100.
  • Schileyko AA, Baminger H, Sattmann H (1997) On the variability of the distal genital tract of Cylindrus obtusus (Draparnaud, 1805) (Gastropoda: Helicidae). Ann Naturhist Mus Wien 99B:535538.
  • Steinke D, Albrecht C, Pfenninger M (2004) Molecular phylogeny and character evolution in the Western Palaearctic Helicidae s.l. (Gastropoda: Stylommatophora). Mol Phylogenet Evol 32:724734.
  • Sturany R, Wagner AJ (1915) Über schalentragende Landmollusken aus Albanien und Nachbargebieten. Denkschriften der kaiserlichen Akademie der Wissenschaften, mathematisch-naturwissenschaftliche Klasse. 91:19138.
  • Wade CM, Mordan PB, Clarke B (2001) A phylogeny of the land snails (Gastropoda: Pulmonata). Proc R Soc B 268:413422.
  • Wade CM, Hudelot C, Davison A, Naggs F, Mordan PB (2007) Molecular phylogeny of the helicoid land snails (Pulmonata: Stylommatophora: Helicoidea), with special emphasis on the Camaenidae. J Moll Stud 73:411415.
  • Wilke T, Schultheiss R, Albrecht C (2009) As time goes by: a simple fool's guide to molecular clock approaches in invertebrates. Am Malacol Bull 27:2545.
  • Zopp L (2012) Geographic Variation in Histology and Anatomy of the Genital Tract of Cylindrus obtusus (Gastropoda: Pulmonata: Helicidae). University of Vienna, Master thesis.
  • Zwickl DJ (2006) Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion, PhD dissertation. The University of Texas at Austin.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1. BI tree based on concatenated mt sequences: 12S and 16S (‘no gap’ option), COI (all positions) (44 sequences).

Figure S2. BI tree based on concatenated nc sequences (H3 and H4) (48 sequences).

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.