Phylogenetic position of Antarctic Scalpelliformes (Crustacea: Cirripedia: Thoracica)

Katrin Linse; Chester Sands; John Buckeridge

The Balanomorpha are the largest group of barnacles and rank among the most diverse, commonly encountered and ecologically important marine crustaceans in the world. Paradoxically, despite their relevance and extensive study for over 150years, their evolutionary relationships are still unresolved. Classical morphological systematics was often based on non-cladistic approaches, while modern phylogenetic studies suffer from severe undersampling of taxa and characters (both molecular and morphological). Here we present a phylogenetic analysis of the familial relationships within the Balanomorpha. We estimate divergence times and examine morphological diversity based on five genes, 156 specimens, 10 fossil calibrations, and six key morphological characters. Two balanomorphan superfamilies, eight families and twelve genera were identified as polyphyletic. Chthamaloids, chionelasmatoid and pachylasmatoids split first from the pedunculated ancestors followed by a clade of tetraclitoids and...

Shell structure is a crucial aspect of barnacle systematics. Within Tetraclitidae, the diametric and monometric growth patterns and number of rows of parietal tubes in the shells are key characteristics used to infer evolutionary trends. We used molecular analysis based on seven genes (mitochondrial COI, 16S and 12S rRNA, and nuclear EF1, RPII, H3, and 18S rRNA) to test two traditional phylogenetic hypothesis: (1) Tetraclitid barnacles are divided into two major lineages, which are distinguished according to monometric and diametric shell growth patterns, and (2) the evolutionary trend in shell parietal development began with a solid shell, which developed into a single tubiferous shell, which then developed into multitubiferous shells. The results indicated that Tetraclitinae and Newmanellinae are not monophyletic, but that Austrobalaninae and Tetraclitellinae are. The phylogram based on the genetic data suggested that Bathylasmatidae is nested within the Tetraclitidae, forming a s...

The taxonomy of pedunculate cirripedes belonging to the genus Pollicipes has essentially remained unchanged since Charles Darwin described them in his exhaustive work on the Cirripedia. This genus includes three species of stalked barnacles: Pollicipes pollicipes in the north-eastern Atlantic, P. polymerus in the north-eastern Pacific and P. elegans in the central-eastern Pacific. However, a population genetics analysis of P. pollicipes suggested the presence of a putative cryptic species collected from the Cape Verde Islands in the central-eastern Atlantic. This study examines the morphology of these genetically divergent specimens and compares them with that of representative Atlantic samples of the biogeographically closely related P. pollicipes and with the poorly described P. elegans. Molecular data, including mitochondrial COX1 and nuclear ribosomal interspaces sequences, were obtained for all species of the genus Pollicipes. Morphological distinctiveness, diagnostic characters, congruent divergence level and monophyletic clustering, at both nuclear and mitochondrial loci support the taxonomic status of this new species, Pollicipes darwini.(Received June 01 2010)(Accepted December 21 2010)(Online publication March 04 2011)

Deep-Sea Research I 73 (2013) 99–116 Contents lists available at SciVerse ScienceDirect Deep-Sea Research I journal homepage: www.elsevier.com/locate/dsri Phylogenetic position of Antarctic Scalpelliformes (Crustacea: Cirripedia: Thoracica) Katrin Linse a,n, Jennifer A. Jackson a, Elaine Fitzcharles a, Chester J. Sands a, John S. Buckeridge b a b British Antarctic Survey, Natural Environmental Research Council, High Cross, Madingley Road, Cambridge CB3 0ET, UK RMIT University, Melbourne, VIC 3001, Australia a r t i c l e i n f o abstract Article history: Received 16 May 2012 Received in revised form 15 November 2012 Accepted 17 November 2012 Available online 24 November 2012 The phylogenetic relationships of seven Antarctic barnacle species, one verrucomorph and six scalpelliforms from the Scotia, Weddell and Ross seas were investigated using DNA sequences from two nuclear genes (18 S and 28 S) and one mitochondrial gene (COI), with a combined total length of 3,151 base pairs. Analyses of these new sequences, together with those of previously published ibliform, lepadiform, scalpelliform, balanomorph and verrucomorph species, conﬁrm that the Scalpelliformes are not monophyletic. Bayesian and maximum likelihood analyses consistently recovered a monophyletic group which comprised Ornatoscalpellum stroemii (Sars) and the Southern Ocean scalpellomorphs; Arcoscalpellum sp. from the Weddell Sea, Arcoscalpellum africanum from Elephant Island, A. bouveti from Bouvet Island, the circum-Antarctic Litoscalpellum discoveryi, Litoscalpellum sp. from Shag Rocks and Scalpellum sp. from the Falkland Trough. We also used multiple fossil constraints in a relaxed clock Bayesian framework to estimate divergence times for the 18 Sþ 28 S phylogeny. Our results indicate a mid Cretaceous divergence for the Weddell Sea Arcoscalpellum sp, followed by a late Cretaceous divergence from the North Atlantic O. stroemii. Subsequent to this, the Antarctic scalpellomorphs began to radiate at the Cretaceous-Tertiary boundary. Monophyly within the scalpellid genera Arcoscalpellum, Litoscalpellum and Scalpellum was strongly rejected by all loci. Our results show incongruence between taxonomy and molecular systematics and highlight the need for more species to be sequenced as well as taxonomic revisions to resolve uncertainties in the phylogenetic relationships of the stalked barnacles. Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved. Keywords: Cirripedia Scalpellidae Barcoding 28S 18S CO1 Scotia Sea 1. Introduction Barnacles (Crustacea: Cirripedia) have been of interest to scientists since Darwin’s monographs (1851–1855) (Darwin 1852, 1855). Based on fossil, morphological and molecular datasets several hypotheses for the evolution of Cirripedia have been proposed (e.g. Newman, 1987, 1996; Anderson, 1994, Glenner et al., 1995; Pérez-Losada et al., 2002, 2004, 2008; Buckeridge and Newman, 2006; Glenner and Hebsgaard, 2006). Today over 1000 species are known, belonging to three superorders, the burrowing barnacles, Acrothoracica (Gruvel, 1905), the parasitic Rhizocephala (Müller, 1962 and the stalked or non-pedunculate Thoracica (Darwin, 1854) (Newman, 1996). The Thoracica are the most diverse with 700 species (Spears et al., 2007). They comprise four recent orders: Ibliformes, Lepadiformes, Scalpelliformes and Sessilia (Buckeridge and Newman, 2006) and have a long fossil record going back to the Cambrian (e.g. Newman et al., 1969; n Corresponding author. Tel.: þ44 0 1223 221631; fax: þ 44 0 1223 221597. E-mail address: kl@bas.ac.uk (K. Linse). Buckeridge and Newman, 2006). The two most current phylogenetic studies of the Thoracica, one based on morphology and fossil data (Buckeridge and Newman, 2006), the other based on three genes (18 S, 28 S and H3) and morphological data (Pérez-Losada et al., 2008), named the order Ibliformes to be the most primitive of the orders. But this is where systematic agreement ends within the Thoracica. While Buckeridge and Newman (2006) erected the new orders Ibliformes, Scalpelliformes and Lepadiformes (including the Heteralepadomorpha in the latter), Pérez-Losada et al. (2008) retain the suborders Iblomorpha, Scalpellomorpha, Lepadomorpha and Heteralepadomorpha in their analysis. The phylogenetic trees based on molecular data indicate non-monophyletic status for the suborders Scalpellomorpha (Scalpelliformes) and Verrucomorpha (Sessilia) (Pérez-Losada et al., 2008). These divergences in the taxonomic nomenclature require further studies to clarify the systematic relationships within the Thoracica and to elucidate their evolutionary history. Within the Thoracica, the ancestral forms are heteralepadomorphs, stalked or pedunculate barnacles, with recent species living epizoic on brachyurans. Pedunculate barnacles of the Lepadiformes and Scalpellomorpha are recorded to live attached to ﬂoating 0967-0637/$ - see front matter Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.dsr.2012.11.006 100 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 objects like wood, ship hulls or plastic debris or epizoic on whales, seals, seaspiders or octocorals (e.g. Foster, 1987; Barnes et al., 2004). The pedunculate barnacles occur worldwide but are not as abundant or well-known as their non-pedunculate relatives. The more recently evolved acorn barnacles are one of the most successful and abundant animal groups as they have colonised the world’s oceans from the deep sea to the shores, where they often dominate the intertidal fauna (e.g. Newman, 1982; Foster, 1987). An exception to this distributional success is the Southern Ocean, deﬁned as the waters within the Polar Front (Clarke and Johnston, 2003), and the coastline of Antarctica. Representatives from the Sessilia are rarely found in the Antarctic shallow waters and the only extant acorn species, Bathylasma corolliforme from the Bathylasmatidae, is known from the shallow near-shore to the upper slope (20–1500 m) (Newman and Ross, 1971, Dayton et al., 1982). Dayton et al. (1982) and Foster (1989) discovered planktonic larvae of this species in vertical hauls in the upper water column of the Ross Sea and Scheltema et al., 2010 reported nauplii in oblique plankton tows to around 150 m off the Antarctic Peninsula. Fossil records of shallow-water balanomorph barnacles exist from the rich Eocene La Meseta Formation on Seymour Island in the western Weddell Sea (Zullo et al., 1988). Another Eocene record is Austrobalanus antarcticus, described from erratic rocks from the Mount Discovery area, Ross Sea (Buckeridge, 2000). Buckeridge (2000) inferred that A. antarcticus most likely lived in shallow, warm water conditions. Pedunculate Antarctic cirripede fossils comprise the Aptian Cretiscalpellum aptiensis and Pycnolepas articulata (Taylor, 1965; Collins, 1980), the Late Cretaceous and Palaeocene Cretiscalpellum praescientis from James Ross Basin, (Feldmann et al., 1993) and the Upper Pliocene Fosterella hennigi from Cockburn Island (Jonkers, 1998). From the fossil record to the present, more stalked than acorn barnacles are identiﬁed from Antarctica (Newman and Ross, 1971). Descriptive work on Southern Ocean cirripedes started with Darwin (1851) but the major taxonomic and biogeographic studies on the Antarctic fauna were done by Hoek (1883), Gruvel (1906, 1907a, 1907b), Nilsson-Cantell (1939), Newman and Ross (1971) and especially Zevina (e.g. 1974, 1975, 1978, 1990, 1993). Recently Young (2002) revised and redescribed the Scalpellidae studied by Gruvel including the Antarctic species collected during Gauss and Discovery expeditions. Ecological studies on Southern Ocean scalpellids are rare and mostly describe the epizoic relationship between the stalked barnacle and their carrier (Barnes et al., 2004; Setsaas and Bester, 2006; Reisinger et al. 2010). Prior to this study, one balanomorph (B. corolliforme), one verrucomorph (Verruca gibbosa), two lepadiforms (Conchoderma auritum, Lepas australis) and 31 scalpelliforms were the only extant thoracicans known from the Southern Ocean. In this study, we morphologically identify verrucomorph and scalpelliform barnacles from the Southern Ocean and use molecular tools to examine their phylogenetic position within the Thoracica and combine these with fossil constraints to estimate the divergence times of these taxa in the Southern Ocean. In this regard, we have analysed partial sequences of the ribosomal genes 18 S and 28 S rDNA as well as of the mitochondrial ‘barcoding’ gene Cytochrome c oxidase sub-unit 1 (COI) to reconstruct the phylogeny and historical divergence times of the Southern Ocean thoracicans. 2. Material and methods 2.1. Specimen collection Thirty scalpelliforms and two verrucomorph specimens were collected by Agassiz trawl during ANDEEP II & III, BENDEX and BIOPEARL I research cruises in the Scotia and northern Weddell seas and at Bouvet Island from March 2002 to April 2006 (Table 1, Fig. 1). When specimens were found as epibionts on other fauna, that taxon was recorded. The specimens were preserved in 96% ethanol for DNA extraction and stored in the marine collection of Table 1 Station locations and vial identiﬁcation numbers. Abbreviations: BI – Bouvet Island, EI – Elephant Island, FT – Falkland Trough, LI – Livingston Island, N – Number of specimens, PB – Powell Basin, RS – Ross Sea, SR – Shag Rocks, SSI – South Sandwich Islands, WS – Weddell Sea. Taxon Verrucidae Altiverruca sp. Scalpellidae Arcoscalpellum africanum Arcoscalpellum bouveti Arcoscalpellum sp. Litoscalpellum discoveryi Litoscalpellum sp. Scalpellum sp. Vial ID N Location Station Latitude (S) Longitude 05-622 2 PB PS67/121-7 631 34.920 06-436 06-441 03-064 03-073 05-711 02-353-1 02-353-2-1 02-365-1 06-198 06-206 NIWA 28464 NIWA 28466 NIWA 28468 NIWA 28469 NIWA 28471 NIWA 28474 NIWA 28476 NIWA 28478 NIWA 28480 NIWA 28482 NIWA 28484 06-874 06-034 06-126 3 5 2 1 3 2 1 2 1 1 1 3 1 1 1 1 1 1 1 2 1 6 1 2 EI EI BI BI WS SSI SSI SSI LI LI RS RS RS RS RS RS RS RS RS RS RS SR FT FT EI-AGT-2 EI-AGT-2 PS65/29-1 PS65/29-1 PS67/102-11 PS65/140-7 PS65/140-7 PS65/141-9 LI-AGT-4 LI-AGT-4 TAN0402/26 TAN0402/186 TAN0402/26 TAN0402/91 TAN0402/197 TAN0402/65 TAN0402/186 TAN0402/19 TAN0402/67 TAN0402/101 TAN0402/140 SR-AGT-4 FT-AGT-1b FT-AGT-1b 611 34.52 611 34.520 541 31,590 541 31,590 651 35.400 581 15.110 581 15.110 581 25.680 621 31.500 621 31.500 711 46.410 711 30.430 711 46.410 721 16.360 711 37.140 721 20.060 711 30.430 711 44.060 721 19.150 711 12.040 721 01.480 531 37.780 5410 18.870 5410 18.870 0 Depth (m) Date Epibiont on 0501 41.970 W 2616-2617 14.03.2005 polychaete tube 0551 15.38 W 0551 15.380 W 0031 13,050 E 0031 13,050 E 0361 29.000 W 0241 52.600 W 0241 52.600 W 0251 01.470 W 0611 49.640 W 0611 49.640 W 1701 59.550 E 1711 25.300 E 1701 59.550 E 1711 26.560 E 1701 51.590 E 170130.020 E 1711 25.300 E 1711 44.000 E 1701 28.300 E 1701 56.260 E 1701 46.280 E 0401 54.140 W 0561 40.750 W 0561 40.750 W 990-976 990-976 376-364 376-364 4794-4797 2962-2940 2962-2940 2314-2331 192-190 192-190 230-219 390-389 230-219 414-409 198-211 328-318 390-389 429-454 272-286 565-571 231-240 226-224 292-289 292-289 12.03.2006 12.03.2006 25.11.2003 25.11.2003 06.03.2005 21.03.2002 21.03.2002 23.03.2002 04.03.2006 04.03.2006 09.02.2004 27.02.2004 09.02.2004 14.02.2004 28.02.2004 13.02.2004 27.02.2004 05.02.2004 13.02.2004 18.02.2004 26.02.2004 05.04.2006 27.02.2006 27.02.2006 cidaroid spine cidaroid spine polychaete tube octocoral 0 C. megalonyx C. tortipalpis C. tortipalpis C. australis N. australe A. carolinensis N. australe D. australis P. vanhoffeni P. rhinoceros C. notialis P. vanhoffeni P. patagonica P. patagonica octocoral soft coral 101 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Table 2 Individual codes and GenBank accession numbers. Underlined are accession numbers downloaded from GenBank for this study. Taxon Specimen ID 18S Gen 28S Thoracica Sessilia Verrucomorpha Neoverrucidae Neoverruca brachylepadoformis Neoverruca sp. Neoverruca sp. Verrucidae Verruca spengleri Verruca stroemia Altiverruca sp. Altiverruca sp. Verruca laevigata Metaverruca recta Rostratoverruca sp. Rostratoverruca krugeri Balanomorpha Balanoidea Austromegabalanus psittacus Balanus balanus Balanus perforatus Balanus glandula Balanus glandula Balanus crenatus Balanus eburneus Balanus nubilis Elminius kingie Austrominius modestus Megabalanus californicus Megabalanus californicus Megabalanus tintinnabulum Megabalanus spinosus Megabalanus stultus Menesiniella Aquila Tamiosoma Aquila Semibalanus balanoides Semibalanus balanoides Semibalanus balanoides Semibalanus balanoides Semibalanus cariosus Wanella milleporae Armatobalanus allium Ceratoconcha domingensis Cantellius pallidus Megatrema anglicum Creusia indica Hiroa stubbingsi Darwiniella conjugatum Pyrgoma cancellata Savignium crenatum Pyrgopsella youngi Trevathana dentate Neotrevathana elongatum Hoekia sp. Nobia grandus Chthamaloidea Chthamalus bisinuatus Chthamalus montagui Chthamalus challengeri Chthamalus stellatus Chthamalus fragilis Chamaesipho tasmanica Catomerus polymerus Jehlius cirratus Notochthamalus scabrosus Coronuloidea Chelonibia patula Tetraclitoidea Tetraclita japonica Tetraclita japonica Tetraclita squamosa Tetraclitella divisa Tetraclitella purpurascens EU082398 EU082397 EU082396 05-622-2 AF022230 AY520649 EU489856 EU082381 EU082377 EU082378 EU082379 EU082380 AY520634 AY520628 AY520629 AF201663 AY520625 AY520624 BLNERRNAA AF201665 AY520636 AY520635 AY520632 AY520631 AY520633 AM497924 AY520630 AF201664 EU370426 AY520626 AM497882 DQ777622 AY520627 AM497906 AM497876 AM497885 AM497879 AM497888 AM497891 AM497894 AM497900 AM497903 AM497909 AM497918 AM497912 AM497915 AM497921 AM497897 CO1 EU082317 EU082316 EU082315 AY520615 EU439973 EU082300 EU082296 EU082297 EU082298 EU082299 AY520600 AY520594 AY520595 AY520591 AY520590 AY520602 AY520601 AY859588 AY520598 AY520597 AY520599 AY520596 EU370440 AY520592 AY520593 AY520644 AY520642 AY520643 AY520641 CMHERRNAA AY520647 AY520648 AY520645 AY520646 AY520610 AY520608 AY520609 AY520607 CBIERRNAA EU083395 AY520640 DQ531768 AY520639 AY520637 AY520638 AY520606 AY520613 AY520614 AY520611 AY520612 AY529695 AY520603 AY520604 102 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Table 2 (continued ) Taxon Specimen ID Pedunculata Scalpelliformes Scalpellidae Arcoscalpellum africanum Arcoscalpellum bouveti Arcoscalpellum sp. Calantica villosa Calantica spinosa Capitullum mitella Litoscalpellum discoveryi Litoscalpellum regina Litoscalpellum sp. Ornatoscalpellum stroemi Scalpellum scalpellum Scalpellum sp. Trianguloscalpellum regium Calanticidae Calantica sp. Calantica spinosa Calantica villosa Smilium peronii Eolepadidae Ashinkailepas seepiophila Leucolepas longa Neolepas rapanuii Neolepas zevinae Volcanolepas sp. Volcanolepas osheai Lithotryidae Lithotrya sp. Lithotrya valentiana Pollicipedidae Capitulum mitella Pollicipes polymerus Pollicipes pollicipes Pollicipes pollicipes Lepadiformes Lepadidae Conchoderma virgatum Conchoderma auritum Lepas anatifera Lepas anserifera Lepas australis 06-436-3 06-441-1 06-441-2 06-441-3 06-441-4 06-441-5 03-064-2 03-073-1 05-711-1 05-711-3 18S EU489861 EU489862 EU489863 EU489864 EU489855 EU489857 EU489858 L26513 Gen 28S CO1 EU489831 EU489832 EU489833 EU489834 EU489835 EU414499 EU414500 EU414501 EU414502 EU414503 EU414504 EU489826 EU489827 EU489828 EU489829 EU414494 EU414495 AY428047 02-353-1-1 02-353-1-2 02-353-2-1 02-365-1-1 02-365-1-2 06-206-1 NIWA 28464 NIWA 28466-1 NIWA 28466-2 NIWA 28466-3 NIWA 28468 NIWA 28471 NIWA 28474 NIWA 28476 NIWA 28478 NIWA 28480 NIWA 28482-1 NIWA 28482-2 NIWA 28484 06-874-1 06-874-2 06-874-3 06-874-4 06-874-5 06-874-6 06-034-1 06-126-1 AY520652 EU489852 EU489853 EU489854 AY520618 EU489823 EU489824 EU489825 EU489879 EU489871 EU489842 EU489872 EU489873 EU489843 EU489844 EU489874 EU489875 EU489845 EU489846 EU489847 EU489848 EU489849 EU489850 EU489851 EU489876 EU489877 EU489878 AY520653 EU489865 EU489866 EU489867 EU489868 EU489869 EU489870 EU082387 EU082388 EU489859 EU489860 EU082389 AY520619 EU489836 EU489837 EU489838 EU489839 EU489840 EU489841 EU082307 EU082308 EU489830 EU082309 EU082385 EU082384 CTIERRNAA EU082386 EU082304 EU082303 EU082395 EU082382 EU082390 EU082391 EU082393 EU082394 EU082314 EU082311 EU082309 EU082310 EU082312 EU082313 EU082383 EU082382 EU082302 EU082301 AY520652 AY520651 EU370427 AY520650 AY520618 AY520617 EU370441 AY520616 EU082402 EU082401 L26516 EU082404 EU082405 EU082321 EU082320 EU082305 EU082323 EU082324 EU414490 EU414491 EU414492 EU414493 EU414498 EU414511 EU41412 EU414513 EU414514 EU414515 EU414516 EU414517 EU414518 EU414519 EU414505 EU414506 EU414507 EU414508 EU414509 EU414510 EU414496 EU414497 103 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Table 2 (continued ) Taxon Specimen ID 18S Gen 28S Lepas testudinata Lepas pectinata Oxynaspididae Oxynaspis celata Poecilasmatidae Megalasma striatum Octolasmis sp. Octolasmis cor Octolasmis warwickii Octolasmis lowei Poecilasma inaequilaterale Poecilasma kaempferi Heteralepadidae Paralepas dannevigi Heteralepadomorpha sp. Ibliformes Iblidae Ibla cumingi Ibla quadrivalvia Rhizocephala Peltogastridae Peltogasterella sulcata Sacculinidae Sacculina carcini the British Antarctic Survey. Fourteen scalpelliform specimens attached to pycnogonids were collected during TAN0402 in the Ross Sea in February 2004 (Table 1, Fig. 1A). The specimens were preserved in 96% ethanol and stored in the marine collection of the New Zealand National Institute of Water and Atmospheric Research (NIWA), Wellington. Species were identiﬁed using the identiﬁcation keys for Antarctic Cirripedia by Newman and Ross (1971) and morphological species descriptions therein. In combination with these samples, DNA sequences (18 S, 28 S and CO1) of scalpelliform, verrucomorph (sessilian), lepadomorph, heterolepadomorph, balanomorph and ibliform (outgroup) species were obtained from GenBank (www.ncbi.nlm.nih.gov) and included in the analysis. 2.2. DNA extraction, PCR amplification and DNA sequencing Genomic DNA was isolated from muscle tissue of the peduncle. DNA was extracted with the DNeasy Tissue Extraction Kit (Qiagen, Crawley, West Sussex, United Kingdom) as directed by the manufacturer. PCR ampliﬁcations were performed in 40 mL volumes containing ﬁnal concentrations of 1x PCR buffer (Bioline), 5% bovine serum albumin 10 mg/mL (Sigma), 200 mM each dNTP, 0.5 mM each primer (CO1: LCO 1490 and HCO 2198 (Folmer et al., 1994), 18 S: SSU 1F and SSU 82R (Medlin et al., 1988), 28 S: LSU 5 and LSU 3 (Littlewood, 1994)), 0.5 units of Taq DNA Polymerase (Bioline), and 1 mL template DNA. Magnesium chloride concentrations varied for each gene region: 18 S, 3 mM; 28 S, 3.5 mM and CO1, 2 mM. Cycling conditions were at 94 1C for 2 min followed by 35 cycles of 94 1C for 1 min, 60 1C (18 S) or 45 1C (28 S and CO1) for 30 sec (annealing temperatures) and 72 1C for 1 min. This was followed by a 4 min extension at 72 1C. In order to sequence the entire 18 S gene the primer SSU 22F (Dorris et al., 2002) was used as an internal sequencing primer. Sequencing was carried out at Macrogen Inc (Seoul, Korea). The sequences were proofread in CodonCode Aligner Version 1.6.3 (CodonCode Coporation 2006). EU082406 EU082403 EU082325 EU082322 EU082412 EU082331 EU082411 EU082408 EU082407 EU082330 EU082327 EU082326 EU082328 OCLERRNAA AY520654 EU082410 AY520620 EU082329 EU082399 EU082400 EU082318 EU082319 EU89493 AY52065 EU082332 AY520621 EU082336 DQ826572 AY520622 AY520656 CO1 Coding genes were checked for open reading frames and blast searched (tblastx) to assess gene homology. 2.3. Datasets Four datasets were constructed. In order to resolve the phylogenetic relationships among the Antarctic scalpellomorphs and verrucomorph, and to assess congruence among independent loci, individual datasets of 18 S (1630 bp, 128 sequences) and 28 S (865 bp, 87 sequences) were constructed, which included multiple non-Antarctic representatives within the wider Thoracica and two rhizocephalan outgroups (Peltogasterella sulcata and Sacculina carcini). We also constructed a CO1 dataset (657 bp, 32 sequences), including two non Antarctic species (sessilian Megatrema anglicum FJ71301 and pedunculate Pollicipes mitella AY51402) in order to assess the phylogenetic relationships among the Southern Ocean taxa. After initial exploration of the data we eventually chose a very limited outgroup set in this case, as (1) taxonomic representation across closely related thoracican groups is currently limited for CO1, and (2) as a rapidly evolving mitochondrial gene this CO1 dataset is unlikely to be informative regarding deeper (pre-Tertiary) divergences in the thoracican tree due to sequence saturation, i.e., the cumulative occurrence of reverse and homoplasious mutations along branches stemming from deep nodes in the tree. Such mutations generate ‘noise’, i.e. DNA sequence variation which is not phylogenetically informative (although it may look as if it is) and so can bias estimates of branch lengths at deeper nodes in the tree. In order to estimate the phylogenetic position and timing of divergence of Antarctic scalpellomorphs and verrucomorph within the Thoracia, a combined 18 Sþ28 S dataset (80 sequences, 2494 aligned bp) was also constructed. Initial alignments of rDNA datasets were performed using Clustal X with default parameter settings (Larkin et al., 2007). All alignments were then manually adjusted by eye using Se-Al v2.0a11 (Rambaut 2007). Ambiguous regions of the rDNA alignments were excluded from the ﬁnal datasets; for 18 S and 28 S this represented an aligned total of 173 and 311 bp respectively. The CO1 dataset was aligned by eye with the aid of DNA-protein translation in Se-Al. Unless otherwise indicated, all 104 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Fig. 1. Map of the collection sites. A. Antarctica, B. detailed view of Scotia Sea. Abbreviations: black line (A.) black dotted line (B.) – Polar Front, BI – Bouvet Island, EI – Elephant Island, FT – Falkland Trough, LI – Livingston Island, PB- Powell Basin, SR – Shag Rocks, SSI – South Sandwich Islands, RS – Ross Sea, WS – Weddell Sea. datasets were subject to all phylogenetic analyses detailed in the following section. 2.4. Phylogenetic analysis All datasets were analysed using jModelTest 0.1.1 (Posada 1998; Guindon and Gascuel 2003) in order to determine the best ﬁtting evolutionary model for that locus. The best ﬁtting model was chosen according to taxon- and variable-sites-corrected Akaike Information Criterion (AICc) scores (Supplementary Table 1). We also performed a w2 test for base composition stationarity in PAUP 4b10 for each gene and for each codon position of the CO1 dataset. Maximum likelihood (ML) bootstrapping (100 resamples) of each locus and the combined 105 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 18 Sþ28 S loci was performed using GARLI (Zwickl 2006) for 18 S and 28 S loci, implementing the jModelTest-speciﬁed evolutionary model. Two replicate likelihood analyses were performed for each bootstrap resampling event (i.e. two analyses of the same dataset). The replicate with the best likelihood score was then included in the overall bootstrap summary of percentage support for each clade. Each bootstrap resample was initiated with a stepwise addition tree and was terminated when an improved tree topology was not found after 70,000 generations of searching. An additional maximum likelihood analysis was carried out for each locus using topological constraints in order to estimate likelihood support for a monophyletic scalpellomorph clade. Constraint trees were constructed in MACCLADE and the maximum likelihood analysis was performed as described above in GARLI with ten replicate analyses, and the best tree chosen from among these replicates. To see if the constrained topology is a signiﬁcantly worse ﬁt to the data than the unconstrained (ML) tree, the Shimodaira-Hasegawa (SH) test (Shimodaira and Hasegawa, 1999) was applied to the maximum likelihood constrained and unconstrained trees in PAUP, using the resampling estimated log-likelihood method (Kishino et al. 1990). In order to assess phylogenetic congruence between the 18 S and 28 S loci, crossed SH tests were also performed, to test whether the 18 S ML topology was a signiﬁcantly worse ﬁt to the 28 S dataset than the 28 S ML topology, and vice versa. Constraint trees were constructed using maximum likelihood analysis in GARLI with ten replicate analyses as described above. SH testing was then performed on the constrained and unconstrained trees in PAUP. Bayesian analyses were conducted in Mr Bayes v3.1.2 (Huelsenbeck and Ronquist, 2001; Ronquist and Huelsenbeck, 2003). For 18 S and 28 S we implemented a 6 parameter model (GTRþIþG), with two replicate runs each, sampling every 1000 generations. The 18 Sþ28 S dataset was partitioned by gene, with separate GTRþI þG parameters and branch lengths estimated for each partition. The CO1 dataset strongly rejected base compositional heterogeneity at the ﬁrst and third codon positions (w2 ¼835, p 40.01), and nearly every site at the third codon position was parsimony informative (217/219 sites), suggesting substantial saturation as well as potentially erroneous phylogenetic signal due to the highly heterogeneous base composition (G ¼10%, C ¼14%, T ¼34%, A¼42%). We therefore estimated the Bayesian CO1 phylogeny in four different ways: (1) evolutionary rate parameters estimated separately (‘partitioned’) for codon positions (1þ2) and 3 (each partition GTRþI þG), in order that the phylogenetic signal from the highly variable 3rd codon positions is weighted equally with the less variable 1st and 2nd codons (Shapiro et al., 2006); (2) removal of the 3rd codon partition; (3) translating 3rd codon position DNA into purines (A and G ¼R) and pyrimidines (C and T¼Y; ‘RY’ coding); (4) ‘RY’ coding of 1st and 3rd codons. All analyses were run for 10–20 million generations, and the initial 10% of each analysis was discarded as burn-in. Convergence was assessed using TRACER v1.5 (Rambaut 2007) and all split frequencies were observed to be o0.01 before analyses were terminated. We also conducted phylogenetic reconstruction using a Bayesian ‘species tree’ approach as implemented in nBEAST (Heled and Drummond, 2010). This approach directly models intra-species polymorphism and incomplete lineage sorting in the 18 S and 28 S gene trees and embeds these trees within a shared species tree, accounting also for effective population size of extant and ancestral species. This analysis was conducted to investigate whether the species tree approach yielded results similar to the concatenated tree obtained using 18 S and 28 S, since SH testing of the two datasets indicated conﬂict between them in terms of best ﬁtting topologies. Constant past population sizes were assumed for both topologies and the birth-death model of speciation was imposed. The analysis was run for 100 million generations and monitored in TRACER in order that all parameters fully converged. Evolutionary rate parameters were estimated independently for each gene as implemented for the Bayesian concatenated analysis. Divergence time analysis of the 18 Sþ28 S Thoracica dataset was carried out using BEAST v1.6.1 (Drummond and Rambaut 2007), using a 6 parameter model run for 50 million generations (10% burnin) and partitioned by gene so that GTRþGþ I and branch lengths were estimated separately for each gene. The Bayesian phylogeny for this dataset was used as the initial tree topology in BEAST and a birth-death model was used to parameterise the branch splitting rates. Fossil constraints were imposed on the nodes listed in Table 3, numbered following Pérez-Losada et al. (2008) and distributed as exponential priors, with the lower bound implemented as the lower end of the epoch containing the ﬁrst fossil occurrence. The size of the exponential distribution varied, with a large range imposed on the C1 node (size¼60) to incorporate uncertainty in the split date at the base of the tree. A prior size of 25 was imposed on all other nodes. Numbered nodes C2-C13 matched those of Pérez-Losada et al. (2008) both in terms of location and age. We also added one further constraint; C14 (Austrobalanus antarcticus as ﬁrst occurrence of the Austrobalanidae in the Middle Eocene, Buckeridge and Newman, 2010). For node C1 (representing the split of Iblomorpha from other thoracicans), we implemented the constraint of Pérez-Losada et al. (2008) and used the Upper Carboniferous Praelepas jaworskii (Newman et al., 1969) as the earliest constraint on the node (minimum age 306.5 MYA). We also explored an alternate scenario for C1 based on the ﬁrst occurrence of the thoracican Cyprilepas holmi in the Upper Silurian at 420–445 MYA (Wills, 1962). In this scenario we constrained the node C1 to a minimum age of 420 MYA to explore the possibility that Cyprilepas is a crown rather than stem group thoracican. The root height of the tree (divergence of thoracicans from rhizocephalans) was uniformly distributed from 450–635 MYA (lower Table 3 Fossils used to constrain the Thoracican phylogeny. Nodes are numbered after Pérez-Losada et al. (2008). ‘Min’ refers to minimum age of occurrence. In these cases, constraint date is taken from the most recent end of the range. Species Age of ﬁrst appearance Reference Praelepas jaworskii Cyprilepas holmi Scillaelepas ginginensis Verruca tasmanica Metaverruca recta Tetraclitella judiciae Palaeobalanus lindsayi Austromegabalanus victoriensis Austrobalanus antarcticus U. Carboniferous (Pennsylvanian) U. Silurian (Pridoll) U. Cretaceous (Santonian) U. Cretaceous (Santonian-Campanian) L.Miocene (Aquitanian) L.Miocene (Aquitanian) M. Eocene M-L. Miocene M. Eocene Newman et Wills, 1962 Buckeridge, Buckeridge, Buckeridge, Buckeridge, Buckeridge, Buckeridge, Buckeridge, al., 1969 1983 1983 1983 1983; Buckeridge, 2008 1983 1983 1983; Buckeridge and Newman, 2010 Node Lower boundary constraint C1 C1 (rev) C2 C7 C8 C11 C12 C13 C14 306.5 415 83.5 70.6 20.4 20.4 37.2 11.6 40 106 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 bound basal Ediacaran). We tested the ﬁt of the phylogeny to a strict molecular clock by ﬁrst running the analysis with a relaxed lognormal clock with no fossil constraints, and inspected the posterior distribution of the standard deviation of rates for each gene. In both cases the values were signiﬁcantly greater than zero, indicating that a strict molecular clock was not supported. In order to accommodate possible variation in rates across the Thoracica we therefore applied a relaxed lognormal clock model as a baseline and compared this with an exponential clock model. We also explored the effect of implementing a Yule prior as the branch splitting rate estimator under the relaxed lognormal clock scheme. Each clock and branch splitting model difference was compared with the baseline model (1) using Bayes Factors calculated from the marginal likelihood in BEAST (Suchard et al., 2001). All parameters achieved full stationarity according to TRACER, with effective sample size values 4200. 3. Results The current study collected 46 specimens comprising one verrucomorph and six scalpelliform species from the Scotia, Weddell and Ross Seas (Table 1). In total, DNA for molecular analysis was successfully extracted from one verrucomorph and 24 scalpelliform specimens from the Scotia and Weddell Seas and from 13 scalpelliform specimens from the Ross Sea (Table 2, Fig. 2). 3.1. Species identifications and biogeography The specimens were identiﬁed to species using the keys and descriptions in Newman and Ross (1971). Where our data extend the depth, and/or geographic distribution ranges of cirripede species or their host species we show this information in bold. The size measurements relate to the largest specimen in our collection. Order Sessilia Lamarck and de Monet, 1818 Suborder Verrucomorpha Pilsbry, 1916 Family Verrucidae Darwin, 1854 Altiverruca sp. Distribution: northern Powell Basin (2117–2616 m). Depth: Size: Remarks: 2117–2616 m. 5.9 mm height, 3.9 mm width (05–662–1). The two collected specimens of Altiverruca sp. were attached to a polychaete tube. Until now Altiverruca gibbosa (Hoek, 1883) had been the only verrucomorph recorded in the Southern Ocean – it is known from 500 to 3000 m depth (Nilsson-Cantell, 1929, Newman and Ross, 1971, Zevina, 1974). Newman and Ross (1971) suggested a review of the species’ taxonomy as Nilsson-Cantell (1939) had combined several wide-ranging species into this taxon. Order Scalpelliformes Buckeridge and Newman, 2006 Family Scalpellidae Pilsbry, 1907 Arcoscalpellum africanum (Hoek, 1883) (Fig. 2A) Distribution: Antarctic Peninsula: off Adelaide Island (2782– 2912 m), Elephant Island (976–990 m); Tristan da Cunha Archipelago (80–183 m). Fig. 2. Species collected for this study. A. 02–365–1–2L. discoveryi, B. 06–874–4 Litoscalpellum sp., C. 06–126–2 Scalpellum sp., D. 05–436–1A. africanum, E. 05–436–1A. africanum, F. 05–711–3 Arcoscalpellum sp. Scale bars in mm. K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Depth: Size: Remarks: 80–2912 m. 10.5 mm total height, 7.8 mm capitular height, 3.8 mm capitular width (06–436). The current specimens were attached to spines of the cidaroid urchin Aporocidaris sp. cf. A. eltaniana Mooi et al., 2000. The number of attached scalpellids per spine varied from three to 16 and the barnacles consisted of three size classes. 107 Distribution: Shag Rocks (224–226 m). Depth: 224–226 m. Size: 7.6 mm total height, 6.4 mm capitular height, 3.3 mm capitular width (06–874–6). Remarks: The specimens were attached to a primnoid octocoral and had colonised the stalk and side branches with more than 50 specimens of various size and age classes. Some of the adult females were carrying bright orange eggs within their shells. Arcoscalpellum bouveti (Nilsson-Cantell, 1939) (Fig. 2B) Distribution: Bouvet Island (40–45 m, 364–376 m). Depth 40–376 m Size 10.5 mm total height, 9.2 mm capitular height, 4.5 mm capitular width (03–064–1) Remarks Two of the current specimens were attached to a polychaete tube, the third was attached to a primnoid stalk. Scalpellum sp. (Fig. 2F) Distribution: Falkland Islands: Falkland Trough (289–292 m) Depth: 289–292 m. Size: 19.6 mm total height, 12.1 mm capitular height, 7.6 mm capitular width (06–034–1). Remarks: The three specimens have been collected. Arcoscalpellum sp. (Fig. 2C) 3.2. Phylogenetic analysis Distribution: Weddell Sea Abyssal Plain (4794–4797 m). Depth: 4794–4797 m. Size: 14.6 mm total height, 11.2 mm capitular height, 5.2 mm capitular width (05–711–3). Remarks: The three collected specimens can be separated from the scalpellid species described in Newman and Ross (1971) by the distinct hairlike bristles on the capitular shells. Litoscalpellum discoveryi (Gruvel, 1906) (Fig. 2D) Distribution: South Sandwich Island: off Sanders Island (2331–2962 m); Antarctic Peninsula: off Brabant Island (128–165 m), Clarence Island (342 m), Livingston Island (190–192 m); Adelie Land (46 m); Ross Sea (92–571 m); Balleny Islands (219–414 m); South Georgia (155–178 m) Depth: 92–2962 m. Size: 23.1 mm total height, 18.8 mm capitular height, 9.3 mm capitular width (02–365–1–2, SSI) 23.6 mm total height, 13.6 mm capitular height, 6.8 mm capitular width (NIWA28474, RS) Remarks: Newman and Ross (1971) state that the distribution of Litoscalpellum discoveryi may depend on the presence of large-sized pycnogonid species. In the current study, the specimens from the South Sandwich Islands had been attached to the pycnogonids Collosendeis megalonyx and C. tortipalpis. The specimens from the Ross Sea were found attached to Collosendeis notialis, Decolopoda australis, Nymphon australe, Pallenopsis patagonica, P. vanhoffeni and Pycnognum rhinoceros and the specimens from the Balleny Islands were attached to the pycnogonids Ammothea carolinensis, Colossendei australis and Nymphon australe. Litoscalpellum sp. (Fig. 2E) Partial sequences of the 18 S rDNA, 28 S rDNA and COI mtDNA genes were generated to examine the phylogenetic relationships and molecular structure of Southern Ocean scalpellid species. The 94 verrucomorph and scalpellid sequences and three outgroup sequences used for this study are summarized in Table 2. The new sequences were deposited in NCBI GenBank (Accession numbers provided in Table 2). We restrict the interpretation and discussion of the results to the Antarctic Thoracica as the wider phylogeny has been covered and discussed before by Pérez-Losada et al. (2008) with the same non-Antarctic sequence dataset on the genes we have used. Here we focus on the positions of the Antarctic taxa in the phylogeny of the Thoracia. 3.2.1. 18S analysis The aligned 18 S rDNA dataset comprised 1630 characters of which 387 were variable and 204 of those were parsimony informative (Supplementary Table 1, Supplementary online ﬁle 1). Bayesian and maximum likelihood analysis produced similar topologies, of which only the Bayesian analysis is shown, with maximum likelihood bootstrap support given on branches (Fig. 3). Basal groupings within the Thoracica were only weakly supported, but nevertheless indicated extensive polyphyly of the Scalpellomorpha (at least ﬁve distinct clades within the Thoracica). SH testing of ML trees with and without the scalpellomorph monophyly constraint (log-likelihoods¼ 7628 and 7565 respectively) indicated that scalpellomorph monophyly is a signiﬁcantly worse ﬁt to the dataset at p o0.001 (likelihood difference¼ 63). The Southern >Ocean scalpellomorphs fell into a single, weakly supported clade (88% Bayesian posterior probability; PP, 450% ML bootstrap support; MLBS). A sister taxon relationship with Ornatoscalpellum stroemi was more strongly supported (100% PP, 75% MLBS), and the rest of this scalpellomorph clade consisted of a single basal cluster comprising Litoscalpellum regina, Scalpellum scalpellum and Trianguloscalpellum regium. Support for the whole scalpellomorph clade was fairly weak (83% PP, 450% MLBS). Within the Southern Ocean clade both Arcoscalpellum and Litoscalpellum were polyphyletic, with Litoscalpellum discoveryi clustered with Arcoscalpellum bouveti (87% PP, 57% MLBS) and Arcoscalpellum africanum clustered with Litoscalpellum sp. (100% PP, 89% MLBS). Both 108 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 IBLOMORPHA VERRUCOMORPHA SCALPELLOMORPHA LEPADOMORPHA & HETERALEPADOMORPHA SCALPELLOMORPHA VERRUCOMORPHA SCALPELLOMORPHA BALANOMORPHA Fig. 3. 18 S Thoracica phylogeny derived from Bayesian analysis. Bayesian posterior probabilities (left-hand side) and maximum likelihood bootstrap support (100 replicates, right-hand side) shown on branches. Scalpellum sp. (06–126) and Arcoscalpellum sp. (05–711–1) were placed basal to these groups. The Southern Ocean Altiverruca specimen clustered within a Rostratoverruca/ Metaverruca/ Altiverruca/ Verruca clade with strong support (100% PP, 92% MLBS). This clade is distant from the second verrucomorph clade (comprising Neoverruca), which is currently restricted to chemosynthetic habitat associated species; these results therefore give support to a revision of higher verrucomorph taxonomy and systematics, a project currently underway by John Buckeridge. 3.2.2. 28S analysis The aligned 28 S rDNA dataset comprised 865 characters of which 333 were variable and 229 of those were parsimony informative (Supplementary Table 1, Supplementary online ﬁle 2). Bayesian and maximum likelihood analysis produced similar topologies, of which only the Bayesian analysis is shown, with maximum likelihood bootstrap support given on branches (Fig. 4). As with 18 S, multiple scalpellomorph clades were found. SH testing of ML trees with and without the scalpellomorph monophyly constraint (log-likelihoods¼-6810 and -6669 respectively) indicated that scalpellomorph monophyly is a signiﬁcantly worse ﬁt to the dataset at po0.001 (likelihood difference¼141). The Southern Ocean scalpellomorphs fell within one of these four clades. As with 28 S, the genera Litoscalpellum, Scalpellum and Arcoscalpellum were all polyphyletic. In contrast to 18 S one Southern Ocean species (Arcoscalpellum sp. 05–711–1) was placed basal to a non-Southern Ocean taxon (O. stroemi) with strong Bayesian posterior support. Support for Southern Ocean scalpellomorphs plus O. stroemi was reasonably high (99% PP, 63% MLBS), as was support for these plus the sister taxon cluster of L. regina, S. scalpellum and T. regium (96% PP, 56% MLBS). The L. discoveryi group (represented by only one specimen in 28 S) was placed basal to the remaining Southern Ocean clade, and did not cluster with A. bouveti (in contrast to 18 S). Scalpellum sp. (06–126) was placed proximal to the A. africanumþLitoscalpellum sp. clade, but support for this placement was extremely weak (51% PP, o50% MLBS). In general Bayesian and ML bootstrap support for nodes within the Southern Ocean scalpellomorph clade were slightly higher for 28 S than for 18 S, 109 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 IBLOMORPHA VERRUCOMORPHA SCALPELLOMORPHA LEPADOMORPHA & HETERALEPADOMORPHA SCALPELLOMORPHA VERRUCOMORPHA SCALPELLOMORPHA BALANOMORPHA Peltogasterella sulcata EU082336 Fig. 4. 28 S Thoracica phylogeny derived from Bayesian analysis. Bayesian posterior probabilities (left-hand side)) and maximum likelihood bootstrap support (100 replicates, right-hand side) shown on branches. which may reﬂect the greater number of parsimony informative sites, spread over a smaller taxon set, in the former. 3.2.3. 18Sþ 28S analysis Bayesian and maximum likelihood analysis of the concatenated genes produced similar topologies, of which only the Bayesian analysis is shown, with maximum likelihood bootstrap support given on branches (Fig. 5). As with 28 S alone, the combined phylogeny also placed O. stroemi within the Southern Ocean scalpellomorph clade, and support for this monophyletic group was strong (100% PP, 93% MLBS). As with the single gene analyses, this clade fell as a sister group to the L. regina/ S. scalpellum/ T. regium cluster and the whole grouping was also strongly supported (100% PP, 89% MLBS). A sister group relationship with the Neolepas/ Neoverruca clade was also found (100% PP, 92% MLBS), consistent with the previous phylogeny of PérezLosada et al. (2008) using the same loci. Relationships within the Southern Ocean scalpellomorph clade were the same as those reconstructed for 28 S alone. When compared with the species tree constructed in nBEAST (Supplementary Fig. 2), the only topological difference within the Southern Ocean scalpellomorph clade was the sister group relationship between A. bouveti and L. discoveryi (81% support) which was also supported by the 18 S gene alone. 3.2.4. Cytochrome oxidase I analysis The aligned CO1 mtDNA dataset comprised 657 characters. Among codons, the 3rd positions were particularly variable, with 203 parsimony informative sites compared to 96 across the other codon positions. Base composition is signiﬁcantly skewed at the 1st and 3rd codon positions for this gene. Skewed base frequencies may contribute an erroneous signal of phylogenetic relationships for this dataset, i.e., taxa may group due to base compositional similarity rather than true shared ancestry, particularly for highly variable loci for which multiple DNA substitutions may have taken place per base over the evolutionary timeframe of the phylogeny. However, taxa with similar base compositions can also reﬂect shared ancestry too, 110 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 IBLOMORPHA SCALPELLOMORPHA VERRUCOMORPHA LEPADOMORPHA & HETERALEPADOMORPHA SCALPELLOMORPHA VERRUCOMORPHA SCALPELLOMORPHA BALANOMORPHA Peltogasterella sulcata Sacculina carcini Fig. 5. 18 S and 28 S combined Thoracica phylogeny derived from Bayesian analysis. Bayesian posterior probabilities (left-hand side)) and maximum likelihood bootstrap support (100 replicates, right-hand side) shown on branches. particularly when the loci are less variable over the phylogeny timeframe. The four treatments of the CO1 dataset all yielded the same clade groupings and indicated substantial polyphyly between Arcoscalpellum and possibly also Litoscalpellum (Fig. 6). However, the relationships between these clades varied by treatment. The untreated CO1 dataset (all codon positions included) yielded a sister group relationship between L. discoveryi and L. sp. (96% BPP), which was not found in any other realisation of the dataset, suggesting that the signal was derived most strongly from 3rd codon position transitional changes. When 1st, and 1st and 3rd codon positions are RY coded, support for a sister group relationship between L. sp. and A. africanum is increased. Although this support is marginal (62% BPP), it is also consistent with independent results from both 18 S and 28 S. The Scalpellum sp. clade is placed basal to these three taxon groups in all analyses except the one where 1st and 3rd codon positions are RY coded. In that analysis it is basal to A. africanum and L. sp, though again support is weak ( 4 50% BPP). Similarly, the combined 18 S þ28 S also places Scalpellum sp. in this position, with marginal support (86% BPP). Most striking is the association of Arcoscalpellum sp. with the pedunculate outgroup Pollicipes mitella in all analyses except where 1st and 3rd codon positions are RY coded. The Arcoscalpellum sp group is placed basal to the other Southern Ocean pedunculates in the larger 18 S þ28 S phylogeny, but are very distant from Pollicipes both in terms of phylogenetic distance and morphology. Given the likely age of divergence of the thoracican tree and the high level of substitution of this mitochondrial locus, this spurious grouping is likely a function of long branch attraction, exacerbated by similar base composition in these taxa. Removal of transitions from the 1 st and 3rd codon positions through RY coding greatly reduces this effect, and likely provides a more realistic hypothesis of relationships than the other three, although intra-clade resolution is reduced by the loss of most transitional changes. Within Arcoscalpellum africanum (support for monophyly, 100% BPP) four haplotypes forming two sister groups were identiﬁed. Four haplotypes were found in Litoscalpellum sp., while Litoscalpellum discoveryi (with 14 specimens analysed from Livingston Island, the South Sandwich Islands and the Ross Sea) K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 111 Fig. 6. CO1 Thoracica phylogeny derived from Bayesian analysis. Bayesian posterior probabilities (left-hand side)) and maximum likelihood bootstrap support (100 replicates, right-hand side) shown on branches. Bayesian phylogenetic analysis of cytochrome oxidase I. (A) Gene partitioned into codon positions (1þ 2) and 3 (B) 3rd codon positions removed from the analysis (C) 3rd codon positions RY coded (D) 1st and 3rd codon positions RY coded. formed ﬁve haplogroups with eleven haplotypes. The deep-water South Sandwich Island haplogroup was resolved as the sister group to shelf specimens from Livingston Island and the Ross Sea. 3.2.5. Divergence time analysis Bayes factors calculated across the clock and branch-splitting models examined indicated minimal (o1) difference in support between the two branch splitting schemes (Yule vs Birth-Death). Estimated divergence dates were nearly identical for the Southern Ocean thoracicans under both models. The exponential clock model was less strongly supported than the lognormal model (Bayes Factor 8.0) and shifted divergence dates towards the present for the taxa under examination. Inspection of the posterior distributions under constraint revealed that (1) the root height was strongly skewed towards older values, (2) divergences at constraint nodes C1 (under either calibration scheme) and C2 were very close to the exponential prior lower boundary constraint implemented at these nodes. This was also reﬂected in the divergence time estimates when the tree was modelled with only the C1 constraint imposed: C2 was estimated at median 62MYA (95% PP 23–104MYA) under this scenario (fossil constraint¼83MY, Table 3). Divergence estimates under the lognormal relaxed clock scenario are shown in Fig. 7. Southern Ocean thoracican divergence commenced close to the KT boundary at 64MYA (95% PP 40–95MYA) with the divergence of L. discoveryi. A midCretaceous split was estimated for the deep-water Southern Ocean Arcoscalpellum sp., prior to divergence of the north Paciﬁc O. stroemi from the rest of the Southern Ocean clade (110MYA; 95%PP 72–154MYA). When the analysis was performed with only C1 (Upper Carboniferous fossil) constrained, estimated divergences were very similar (data not shown). The main Southern Ocean radiation commenced at the Cretaceous-Tertiary boundary (64MYA; 95% PP 32–99MYA) and spanned the early Tertiary, while the earlier Arcoscalpellum sp. divergence was in the early Cretaceous (109MYA; 95%PP 64–164MYA). Constraining the minimum age of the crown group radiation to the Upper Silurian at 415 MYA (Wills, 1962) rather than the Upper Carboniferous at 306.5 MYA (Newman et al., 1969) shifted most branches in the tree to older dates, but only increased median divergence times within the Antarctic scalpelliforms by 10MY (Supplementary Fig. 3). Under this constraint the main radiation commenced at 75MYA (95% PP 45–110 MY) in the late Cretaceous and continued through the Tertiary, while the earlier Arcoscalpellum sp./ O. stroemi split was estimated to occur in the early Cretaceous at 128 MYA (95% PP 81–179 MYA). In this study we have focussed on the radiation of thoracicans in the Antarctic. However, it is also instructive to compare our general divergence time estimates with those produced in an earlier study using many of the same set of fossil time constraints 112 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Fig. 7. Bayesian chronogram for the Thoracica.The 95% posterior percentiles are shown either side of each node. Fossil constraints C1-C14 are detailed in Table 3. (Pérez-Losada et al., 2008). One major difference between this study and ours is that Pérez-Losada et al. (2008) included two scalpellomorph fossil constraints in a key clade to which we have added a number of Southern Ocean taxa, and it is this clade that is particularly pertinent to our analysis. We did not impose these fossil constraints (Cretiscalpellum glabrum and Arcoscalpellum fossula) due to uncertainty as to their phylogenetic placement within the clade. The estimated radiation of this clade in our analysis predates the occurrence of both fossils and this is what was expected. Further, with the additional taxon sampling in this scalpellomorph clade, the phylogenetic relationships were found to differ from the smaller taxon set used in Pérez-Losada et al. (2008), and this prevented our placement of the constraints in the same way as had been done in that study. One contrast between the present study and that of Pérez-Losada et al. (2008) is the much earlier divergence date inferred for the large scalpellomorphþNeoverruca clade, with a c.240 MY origin time compared to c.170 MY estimated in this study. The divergence of Neoverruca from its scalpellomorph sister taxa is also much older, with an origin time of c.130 MYA as opposed to c.30MY in this study. The pattern of divergence of the other scalpellomorph clades in the phylogeny is otherwise similar between the two studies, although the current study tends to have more recent median date estimates than those estimated in Pérez-Losada et al. (2008). The contrasts described above may be due to four underlying effects: (A) wider taxon sampling of the scalpellomorphs falling in the scalpellomorphþNeoverruca clade in this current study; (B) fossil constraints imposed in Pérez-Losada et al. (2008) and not used in the current study, which may ‘relax’ estimated divergence times towards more recent values; (C) differences in the underlying datasets; additional gene H3 was used in Pérez-Losada et al. (2008) and multiple regions of alignment uncertainty were removed from 18 S and 28 S rDNA datasets in the present study, so estimated branch lengths are likely to differ between the studies; (D) different relaxed molecular clock models and constraint priors imposed in the two cases. Factors C and D differ over the whole phylogeny while A and B are likely to be the main drivers of the difference within the Southern Ocean scalpellomorph clade of interest. 4. Discussion 4.1. Species richness and biogeography Thoracican cirripedes are one of the less well-known and less diverse groups in Antarctica. At present geo-referenced records of K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 one balanomorph (Bathylasma corolliforme), two verrucomorph (Verruca gibbosa, Altiverruca sp.), two lepadiform (Conchoderma auritum, Lepas australis) and 31 scalpelliform barnacles are known from the Southern Ocean (Newman and Ross, 1971, this study). Only 36 species are known today, whereas groups like isopod crustaceans (almost 1000 species, Brandt et al., 2007a), amphipods and prosobranch gastropods (over 500 species each, Brandt et al., 2007b) or pycnogonids (175 species, Clarke and Johnston, 2003) show a high species diversity in the Southern Ocean. These 36 barnacle species represent less than 5% of the global cirripede species (Newman and Ross, 1971) while, for example, the Antarctic pycnogonids represent 17% of global pycnogonid species diversity (Clarke and Johnston, 2003). As few studies on Antarctic cirripedes have been published (e.g. Nilsson-Cantell, 1929, 1930; Newman and Ross, 1971; Arnaud, 1973; Zevina 1974, 1975, 1993; Foster, 1989; Barnes et al., 2004) most of the geographic or bathymetric records presented in this study are new. Newman and Ross (1971) discussed the contemporary biogeography of the Antarctic cirripedes in detail. CircumAntarctic and circum-subantarctic distributions are reported for the balanomorph and the lepadiforms. Their wide distribution ranges might be linked with their mode of reproduction or settlement substrate. Bathylasma coralliforme has planktrophic larval stages and the two lepadiforms are known as epizoans on mobile megafauna and ﬂoating objects, and have lecithotrophic larvae (Zevina, 1974; Barnes, 1989; Foster, 1989; Hinojosa et al., 2006; Setsaas and Bester, 2006; Reisinger et al., 2010). The geographic ranges of the Antarctic scalpelliforms are restricted, with only four species (Arcoscalpellum africanum, A.vitreum, A. formosum and Gymnoscalpellum klepalae) reported north and south of the Polar Front. Newman and Ross (1971) reported the distribution records of the former three species while Federspiel and Hoffer (1997) described G. klepalae from the Scotia Sea, Ross Sea and off Tasmania. Eighteen species have been recorded from only their type locality or from nearby within the same Antarctic sub-region (as deﬁned in Linse et al., 2006 based on species biogeographic distributions). A further six species are known from two Antarctic sub-regions while only two species, Arcoscalpellum bouvieri and Litoscalpellum discoveryi, are reported from several Antarctic sub-regions (Newman and Ross, 1971). The generally restricted species ranges in Antarctic scalpellids might again be linked with their reproductive mode. The present knowledge is that Antarctic scalpellids brood their young and release advanced, non-feeding cyprids that are expected to settle soon after release, as usually assumed for high latitude and deep-sea species (Newman and Ross, 1971). The recent study by Buhl-Mortensen and Høeg (2006) on three north Atlantic scalpellids showed different settlement strategies in two deep-sea species. While in Ornatoscalpellum stroemi (100–1600 m depth) cyprids hatched and settled soon after release, Arcoscalpellum michelottianum (64–5190 m depth) release nauplii that stay for more than 10 days in the plankton. They linked the reproductive strategies with adaptations to the species’ different habitats, hypothesising that short larval stages are advantageous for high disturbance habitats while longer larval times are beneﬁcial as they provide opportunity for colonization of habitats, like seamounts. To date no studies on larval development in Antarctic scalpellids have been carried out so it is unknown if the Antarctic larval or juvenile stages travel in the water column at all. 4.2. Phylogenetic analysis 4.2.1. Thoracican phylogeny and systematics The topologies obtained from the separate analysis of the three genes with Bayesian and maximum parsimony are generally very 113 similar. All phylogenies agree with the monophyly of the analysed species A. africanum, A. bouveti, Arcoscalpellum sp., L. discoveryi and Litoscalpellum sp. However, the higher relationships among taxa are in contrast with the existing morphological classiﬁcation. The phylogenies based on the 18 S and 28 S rDNA loci, using newly sequenced Southern Ocean taxa together with ibliform, lepadiform, scalpelliform and verrucomorph sequences available in GenBank, are consistent with the phylogeny by Pérez-Losada et al. (2008) and conﬁrm that the order Scalpelliformes is not monophyletic. The present study does not support monoyphyly for the scalpellid genera Arcoscalpellum, Litoscalpellum and Scalpellum while the Southern Ocean Altiverruca sp. forms a wellsupported, monophyletic clade with the other species of this genus. The species of Arcoscalpellum presented here are all from the Southern Ocean. The analysed Acroscalpellum species clustered together with Scalpellum sp. and Litoscalpellum sp. in 18 S and 28 S, indicating that morphological characters for the genus identiﬁcation given in Newman and Ross’ (1971) key are insufﬁcient for supporting phylogenetic relationships. The deep-water Arcoscalpellum sp. is placed basal, and is substantially divergent from other Arcoscalpellum species, suggesting critical additional taxonomic/genetic analysis of this species should be performed. The genus Litoscalpellum shows non-monophyly as well. Litoscalpellum regina from the Gulf of Mexico clusters with Scalpellum scalpellum from Sweden and Trianuloscalpellum regium from the Gulf of Mexico (all sequences from Pérez-Losada et al., 2008) but not with the Southern Ocean Litoscalpellum discoveryi or Litoscalpellum sp. from this study. Litoscalpellum sp. is more closely aligned with A. africanum in 18 S and A. africanum/ A. bouveti in 28 S. The identiﬁed polyphylies in the scalpellid genera might be evidence that the morphological characters for the genus identiﬁcation are not good synapomorphies. In general, a determination of polyphyly can arise due to inadequate phylogenetic inference either because there is not enough resolution in the data or because the gene tree is not representative of the species tree (Funk and Omland, 2003). We consider that this explanation is unlikely to apply here as the data analysed resulted in well-resolved trees from three independently evolving gene regions, and the groups form multiple distantly related clusters within the overall phylogeny. Alternatively, the conﬂict between morphology and phylogenetic relationships observed here could be due to imprecise or limited morphological taxonomy, perhaps confounded by synapomorphies that are in fact homoplasies and indicate convergent evolution (see PérezLosada et al., 2008). Our results show incongruence between taxonomy and molecular systematics and indicate the need for more scalpelliform species to be sequenced to resolve uncertainties in the phylogenetic relationships of the stalked barnacles. An interesting, well supported, higher clade is the one containing the scalpellomorphs linked with hydrothermal vents (Vulcanolepas, Leucolepas, Neolepas), the verrucomorphs linked with hydrothermal vents (Neoverruca) and the scalpellomorph clade comprising the Scalpellidae. All taxa within this clade occur in deeper waters, compared with the more basal clades of scalpellomorphs (Calantica, Smilium, Pollicipes), where species occur from the intertidal to the deeper continental slopes. Nonetheless, these deep-water clades are likely to have originated in shallower waters, subsequently radiating into a deep-water and a refugial, hydrothermal lineage and independently evolving further. The hydrothermal lineage comprises stalked and unstalked forms, currently listed as scalpellomorphs and verrucomorphs; the latter is represented by Neoverruca, a most peculiar form that in early ontogeny has a distinct, bilateral pedunculate stage, before assuming asymmetry along with the very much reduced ‘‘peduncle’’ as an adult (Newman, 1989). The discovery of scalpellomorph dominated hydrothermal vents in the East 114 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Scotia Ridge of the Southern Ocean (Rogers et al., 2012) and the southwest Indian Ridge (Copley, pers. comm.) should enhance our knowledge of the evolutionary history of deep-water scalpellomorphs and the role of Southern Ocean species in this evolution. 4.2.2. Antarctic divergence times The main radiation of Antarctic scalpellomorph species began in the earliest Tertiary and continued through the Eocene; divergence times for this group were fairly robust to the implementation of two very different root nodes on the crown group radiation, and the application of various molecular clock and branch splitting scenarios. One exception is the deep-sea Arcoscalpellum sp., which, based on molecular evidence, is estimated to have diverged in the middle Cretaceous. The Southern Ocean has been identiﬁed as a centre of evolutionary origin for marine species, including shelf, upper slope and deep-sea lineages, during the late Mesozoic and Cenozoic (Zinsmeister and Feldmann, 1984; Brandt, 1992; Held, 2000; Briggs, 2003; Gage, 2004; Brandt et al., 2007a; Rogers, 2000). In the late Cretaceous the Antarctic shallow-water fauna formed part of the cool-water Weddellian Province around southern Gondwana (Crame, 2004) and connected to warmer faunas of central Chile and South Africa. Fragmentation of Gondwana lead to vicariance of the Weddellian province, and diversiﬁcation took place along the coast of Gondwana, forming sibling lineages to the Antarctica ones, like the ﬁsh lineages in South America and southern Australia (Near, 2004). The same historical events might have lead to the radiations in these marine groups. The onset of the continental glaciations and the covering of shoreline and shallow shelf by ice shelves are likely to have resulted in the loss of habitat, and in combination with sea-ice formation in the near shore, made these habitats unavailable for traditional epifaunal intertidal species like mussels and most balanomorphs (Clarke and Crame, 2010). Species living on the deeper shelf or slope were less affected by the glaciations and more affected by the cooling of the sea, which drove warmwater species out of the SO (see Clarke and Crame, 2010 and references therein). Through the Cenozoic the evolutionary history of the bivalve genus Limopsis in the SO showed the disappearance of the shallow-water/ inshore species and the appearance of outer shelf species (Whittle et al., 2011). The morphogroups linked with warm-water habitats in this genus are now absent in the Southern Ocean (Oliver, 1981). The fossil record of Antarctic barnacles gives evidence for the disappearance of warm-water preferring balanomorphs after the Eocene, with only scalpellomorphs appearing in the younger epochs (Zullo et al., 1988; Feldmann et al., 1993; Jonkers, 1998; Buckeridge, 2000). Strugnell et al. (2008) suggested that the Southern Ocean has played a major role in inﬂuencing the diversity of the deep sea globally via the thermohaline circulation and affected the evolutionary origins of many marine deep-water lineages. Links have been made between the arrivals of fauna into the abyss after the Cretaceous and Early Cenozoic anoxia extinction events, the development of the thermohaline circulation, and the presence at abyssal depths of bottom water of Antarctic origin that could have introduced Antarctic species (Rogers, 2000; Jacobs & Lindberg, 1998; Gage, 2004). At present we cannot state if the clade containing the SO Scalpellidae is another example of an Antarctic origin of a global deep-sea lineage as not enough deep-sea scalpellomorphs from non-hydrothermal sites have been sequenced yet. 4.2.3. Antarctic context of results This study presents the ﬁrst molecular sequences (18 S, 28 S, COI) for six species of Antarctic barnacles and one species from the Magellan region: Altiverruca sp. from the Powell Basin, Arcoscalpellum africanum from Elephant Island, A. bouveti from Bouvet Island, Arcoscalpellum sp. from the Weddell Sea, the widely distributed Litoscalpellum discoveryi, Litoscalpellum sp. from Shag Rocks and Scalpellum sp. from the Falkland Trough. All analysed species show strong support for their species groups in the three separate gene topologies. The analysed specimens of Verruca sp., Acroscalpellum bouveti, Acroscalpellum sp. and Scalpellum sp. resulted in single genetic lineages for the ampliﬁed genes. These results probably arise from the fact that only one to two specimens from one locality were successfully sequenced. The analysis of the nuclear genes 18 S and 28 S resulted in single, unique genetic lineages for Acroscalpellum africanum and Litoscalpellum sp. while the analysis of the mitochondrial COI gene revealed the existence of four haplotypes in two sister groups in A. africanum from one site at Elephant Island and two halotypes in Litoscalpellum sp. from one site at Shag Rocks. In the widely-distributed species Litoscalpellum discoveryi, for which sequences were available from the Ross Sea, South Sandwich Islands and Livingston Island, several lineages were found for 18 S, none for 28 S and six haplotypes in ﬁve haplogroups for COI. The topology of the COI gene revealed one well-supported clade of one haplotype for four specimens from the deep waters (2331–2962 m) of the South Sandwich Islands and a sister grouping with the shallow water/shelf specimens from the Ross Sea and Livingston Island (190–571 m). The lineages within the shelf group are structured by a basal cluster with the deeper shelf Ross Sea specimens (570 m) and a subdivided cluster containing the Livingston Island and Ross Sea specimens from 190–389 m collection depth. The separation of COI haplotype groups by depth has been reported for several Antarctic invertebrate species, e.g. the isopod Glyptonotus antarcticus (Held and Wägele, 2005), the crinoids Promachocrinus kerguelensis (Wilson et al., 2007), the bivalve Lissarca notorcadensis (Linse et al., 2007) and octopods of the genus Pareledone (Allcock et al., 2011). These studies have related the strong genetic structure and depth groupings with the existence of cryptic species in the analysed taxa. The presented analysis of L. discoveryi also reveals a haplotype shared between Livingston Island and the Ross Sea, supporting the wide distribution within the Southern Ocean of the species. Similar results have been found in the circum-Antarctic species Nymphon australe (Arango et al., 2011), Pareledone aequipalillae (Allcock et al., 2011) and Sterechinus neumayeri (Dı́az et al., 2011). 5. Conclusions The phylogenetic analysis of the Thoracica gives evidence for a monophyletic clade comprising the SO scalpellomorphs and a Tertiary radiation of the Antarctic species. With deeper bathymetric distributions, the recent SO scalpellomorphs were less affected by the glaciations and cooling events in the SO than the balanomorphs and persisted in the SO until today. This study demonstrates that SO barnacles have been understudied in the last decades and that recent collections in the Atlantic sector of the SO have increased the number of reported species. A wider study, including scalpellomorphs recently collected during the CAML campaigns from the shallow and deep waters from the SO, would enhance our knowledge of the evolutionary history of this taxon as well as of the role the SO has played in forming the global deep water diversity. K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Acknowledgement We are grateful to the cruise leaders, captains, ofﬁcers and crews of RRS James Clarke Ross (JR 144 - BIOPEARL) and of PFS Polarstern (ANT XIX-4, ANT XXII-3) who enabled us to collect the samples for this study. Pete Bucktrout kindly photographed the Antarctic species. We thank A.-N. Lörz (NIWA) for the loan of the Ross Sea specimens which were collected during TAN0402, a biodiversity survey of the western Ross Sea and Balleny Islands undertaken by NIWA and ﬁnanced by the NZ Ministry of Fisheries. We are grateful to A. Gooday, M. Pérez-Losada and two unknown reviewers for their comments improving this publication. This paper is a contribution to British Antarctic Survey core project ‘EVOLHIST’ with ﬁnancial support from NERC, ANDEEP publication no xx and linked with the SCAR ‘EBA’ programme. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.dsr.2012.11.006. References Allcock, A.L., Barratt, I., Eléaume, M., Linse, K., Norman, M.D., Smith, P.J., Steinke, D., Stevens, D.W., Strugnell, J.M., 2011. Cryptic speciation and the circumpolarity debate: a case study on endemic Southern Ocean octopuses using the coxI barcode of life. Deep Sea Res. 58, 242–248, Pt. II. Anderson, D.T., 1994. Barnacles – Structure, function, development and evolution. Chapman and Hall, London. Arango, C.P., Soler Membrives, A., Miller, K.J., 2011. Genetic differentiation in the Circum-Antarctic sea spider Nymphon australe (Pycnogonida; Nymphonidae). Deep Sea Res 58 (Pt. II), 212–219. Arnaud, P.M., 1973. The genus Lepas Linné, 1758, from the French islands in the southern hemisphere and the Antarctic (Cirripedia). Crustaceana 24, 157–162. Barnes, D.K.A., Warren, N.L., Webb, K., Phalan, B., Reid, K., 2004. Polar pedunculate barnacles piggy-back on pycnogona, penguins, pinniped seals and plastic. Mar. Ecol. Prog. Ser. 284, 305–310. Barnes, M., 1989. Egg production in cirripedes. Oceanogr. Mar. Biol. Ann. Rev. 27, 91–166. Brandt, A., 1992. Origin of Antarctic Isopoda (Crustacea, Malacostraca). Mar. Biol. 113, 415–423. Brandt, A., Gooday, A.J., Brandao, S.N., Brix, S., Brökeland, W., Cedhagen, T., Choudhury, M., Cornelius, N., Danis, B., De Mesel, I., Diaz, R.J., Gillan, D.C., Ebbe, B., Howe, J., Janussen, D., Kaiser, S., Linse, K., Malyutina, M., Nunes Pawlowski, J., Raupach, M., Vanreusel, A., 2007a. The Southern Ocean deep sea: ﬁrst insights into biodiversity and biogeography. Nature 447, 307–311. Brandt, A., De Broyer, C., De Mesel, I., Ellingsen, K.E., Gooday, A.J., Hilbig, B., Linse, K., Thomson, M.R.A., Tyler, P.A., 2007b. The biodiversity of the deep Southern Ocean benthos. Phil. Trans. Roy. Soc. Ser. B Biol. Sci. 362, 39–66. Briggs, J.C., 2003. Marine centres of origin as evolutionary engines. J..Biogeogr. 30, 1–18. Buckeridge, J.S., 1983. The fossil barnacles (Cirripedia: Thoracica) of New Zealand and Australia. New Zeal. Geol. Sur., Paleont. Bull. 50, 151. Buckeridge, J.S., 2000. A new species of Austrobalanus (Cirripedia, Thoracica) from Eocene erratics, Mount Discovery, McMurdo Sound, East Antarctica, in: Stilwell, J.D., Feldmann, R.M. (Eds.) Paleobiology and Paleoenvironments of Eocene Rocks: McMurdo Sound, East Antarctica, Ant. Res. Ser. 76, 329–333. Buckeridge, J.S., 2008. Two new species and a new subspecies of Tetraclitella (Cirripedia: Thoracica) from the Cainozoic of Australia and New Zealand and a consideration of the signiﬁcance of tubiferous walls. Zootaxa 1897, 43–52. Buckeridge, J.S., Newman, W.A., 2006. A revision of the Iblidae and the stalked barnacles (Crustacea:Cirripedia:Thoracica), including new ordinal, familial and generic taxa, and two new species from New Zealand and Tasmanian waters. Zootaxa 1136, 1–38. Buckeridge, J.S., Newman, W.A., 2010. A review of the subfamily Elminiinae (Cirripedia: Thoracica: Austrobalanidae), including a new genus, Protelminius nov. from the Oligocene of New Zealand. Zootaxa 2349, 39–54. Buhl-Mortensen, L., Høeg, J.T., 2006. Reproduction and larval development in three scalpellid barnacles (Scalpellum scalpellum (Linnaeus, 1767), Ornatoscalpellum stroemii (M. Sars 1859) and Arcoscalpellum michelottianum (Seguenza 1876), Crustacea: Cirripedia: Thoracica): Implications for reproduction and dispersal in the deep-sea. Mar. Biol. 149, 829–844. Clarke, A., Crame, J.A., 2010. Evolutionary dynamics at high latitudes: speciation and extinction in polar marine faunas. Phil. Trans. Roy. Soc. B 365, 3655–3666. Clarke, A., Johnston, N.M., 2003. Antarctic marine benthic diversity. Oceanogr. Mar. Biol. An Rev. 41, 47–114. 115 Collins, J.S.H., 1980. A new Pycnolepas (Cirripedia, Balanidae) from the (?) Lower Aptian of Alexander Island. Brit. Ant. Sur. Bull. 50, 21–26. Darwin, C., 1851. A monograph on the fossil Lepadidae, or pedunculated Cirripedia of Great Britain. Palaeontographical Society Monographs 88 pp. Darwin, C., 1852. A monograph on the subclass Cirripedia with ﬁgures of all the species. The Lepadidae; or, pedunculated cirripedes. Ray Society Publications, London 400 pp. Darwin, C., 1854. A monograph on the subclass Cirripedia, with ﬁgures of all the species. The Balanidae, Verrucidae etc. Ray. Soc. Publ., London 684 pp. Darwin, C., 1855. A monograph on the fossil Palanidae and Verrucidae of Great Britain. Palaeontographical Society Monographs 44 pp. Dayton, P.K., WA Newman, W.A., J Oliver, J., 1982. The vertical zonation of the deep-sea Antarctic acorn barnacle, Bathylasma corolliforme (Hoek): Experimental transplants from the shelf into shallow water. J. Biogeo. 9, 95–109. Dı́az, A., Féral, J.-P., David, B., Saucede, T., Poulin, E., 2011. Evolutionary pathways among shallow and deep-sea echinoids of the genus Sterechinus in the Southern Ocean. Deep Sea Res. 58 (Pt. II), 205–211. Dorris, M., Viney, M.E., Blaxter, M.L., 2002. Molecular phylogenetic analysis of the genus Strongyloides and related nematodes. Int. J. Parasit 32, 1507–1517. Federspiel, K., Hoffer, T., 1997. Eine neue Art der Gattung Gymnoscalpellum Newman & Ross, 1971 aus der antarktischen und neuseeländischen Region (Crustacea: Cirripedia: Thoracica: Scalpellidae). Ann. Naturhist. Mus.. Wien 99B, 555–563. Feldmann, R.M., Tshudy, D.M., Thomson, M.R.A., 1993. Late Cretaceous and Palaeocene decapod crustaceans from James Ross Basin, Antarctic Peninsula. Paleont. Soc. Mem. 28, 41. Folmer, O., Black, M., Hueh, W., Lutz, R., Vrijenhoek, R., 1994. DNA primers for ampliﬁcation of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol. Marine Biol. Biotech. 3, 294–299. Foster, B.A., 1987. Barnacle ecology and adaptation. In: Southward, A.J. (Ed.), Barnacle Biology, Crustacean Issues 5, A.A. Belkema, Rotterdam, The Netherlands, pp. 113–134. Foster, B.A., 1989. Balanomorph barnacle larvae in the plankton at McMurdo Sound, Antarctica. Pol. Biol. 10, 175–177. Funk, D.J., Omland, K.E., 2003. Species-level paraphyly and polyphyly: Frequency, causes, and consequences, with insights from animal mitochondrial DNA. Annu. Rev. Ecol. Evolution Syst. 34, 397–423. Gage, J.D., 2004. Diversity in deep-sea benthic macrofauna: the importance of local ecology, the larger scale, history and the Antarctic. Deep-Sea Res. 51 (Pt, II), 1689–1708. Glenner, H., Grygier, J.T., Høegg, J.T., Jensen, P.G., Schram, F.R., 1995. Cladistic analysis of the Cirripedia Thoracica (Crustacea:Thecostraca). Zool. J. Linn. Soc. 114, 365–404. Glenner, H., Hebsgaard, M.B., 2006. Phylogeny and ecolution of life history strategies of the parasitic barnacles (Crustacea, Cirripedia, Rhizocephala). Mol. Phylogenet. Evol. 41, 528–538. Gruvel, A., 1905. Monographie des Cirrhipe des ou The cosrace s. Masson et Cie, Paris: 472 pp. Gruvel, A., 1906. Cirrhipe des du Discovery. B. Mus. Hist. Nat. Paris 12, 270–273. Gruvel, A., 1907a. Crustacea. VI. Cirrhipe des. National Antarctic Expedition. Natural History volume III. Zoology and Botany, British Museum (Natural History), London, pp 1-4, 1pl. Gruvel, A., 1907b. Note préliminaire sur les cirrhipe des pe doncule s recueillis par l’expédition antarctique allemande du ‘‘Gauss’’. B. Soc. Zool. Fr. 32, 157–163. Held, C., 2000. Phylogeny and biogeography of serolid isopods (Crustacean, Isopoda, Serolidae) and the use of ribosomal expansion segments in molecular systematics. Mol. Phylogenet. Evol. 15, 165–178. Heled, J., Drummond, A.J., 2010. Bayesian inference of species trees from multilocus data. Mol. Biol. Evol. 27 (3), 570–580. Held, C., Wägele, J.W., 2005. Cryptic speciation in the giant Antarctic isopod Glyptonotus antarcticus (Isopoda: Valvifera: Chaertiliidae). Sci. Mar. 69, 175–181. Hinojosa, I., Boltana, S., Lancelloti, D., Macaya, E., Ugalde, P., Valdivia, N., Vasquez, N., Newman, W.A., Thiel, M., 2006. Geographic distribution and description of four pelagic barnacles along the south east Paciﬁc coast of Chile–a zoogeographical approximation. Rev. Chil. Hist. Nat. 79, 13–27. Hoek, P.P.C., 1883.Report on the Cirripedia collected by H.M.S. Challenger during the years 1873-1876. Report on the scientiﬁc results of the voyage of H.M.S. Challenger during the years 1873-1876, Zoology, 8pt 25, pp 1-169, 3 ﬁgs, pls 1-13. Jonkers, H.A., 1998. The Cockburn Island Formation; Late Pliocene interglacial sedimentation in the James Ross Basin, northern Antarctic Peninsula. Newsl. Stratigr 36, 63–76. Lamarck J.P.B.A. de Monet de , 1818. Histoire naturelle des animaux sans verte bres, présentant les caracte res généraux et particuliers de ces animaux, leur distribution, leurs classes, leurs familles, leurs genres, et la citation des principales espe ces qui s’y rapportent; précédée d’une introduction offrant la détermination des caracte res essentiels de l’animal, sa distinction du végétal et des autres corps naturels, enﬁn, l’exposition des principes fondamentaux de la zoologie. Tome cinquie me. - pp. [1-3], 1-612. Paris. (Deterville, Verdiere). Larkin, M.A., Blackshields, G., Brown, N.P., Chenna, R., McGettigan, P.A., McWilliam, H., Valentin, F., Wallace, I.M., Wilm, A., Lopez, R., Thompson, J.D., Gibson, T.J., Higgins, D.G., 2007. Clustal W and Clustal X version 2.0. Bioinformatics 23, 2947–2948. Linse, K., Grifﬁths, H.J., Barnes, D.K.A., Clarke, A., 2006. Biodiversity and Biogeography of Antarctic and Sub-Antarctic Mollusca. Deep-Sea Res. 53 (Pt. II), 985–1008. 116 K. Linse et al. / Deep-Sea Research I 73 (2013) 99–116 Linse, K., Cope, T., Lörz, A.-N., Sands, C.J., 2007. Is the Scotia Sea a center of Antarctic marine diversiﬁcation? Some evidence of cryptic speciation in the circum-Antarctic bivalve Lissarca notorcadensis (Arcoidea: Philobryidae). Pol. Biol. 30, 1059–1068. Littlewood, D.T.J., 1994. Molecular phylogenetics of cupped oysters based on partial 28S rRNA gene sequences. Mol. Phylogenet. Evol.3, 211-229. Kishino, H., Miyata, T., Hasegawa, M., 1990. Maximum likelihood inference of protein phylogeny and the origin of chloroplasts. J. Mol. Ecol. 30, 151–160. Medlin, L., Elwood, H.J., Stickel, S., Sogin, M.L., 1988. The characterization of enzymatically amplifed eukaryotic 16S-like rRNAcoding regions. Gene 71, 491–499. Müller, F., 1962. Die Rhizocepahlen, eine neue Gruppe schmarotzender Kruster. Archive der Naturgeschichte, 28. Jahrgang, Band 1. Near, T.J., 2004. Estimating divergence times of notothenioid ﬁshes using a fossilcalibrated molecular clock. Ant. Sci. 16, 37–44. Newman, W.A., 1996. Sous-classe des Cirripe des (Cirripedia Burmeister, 1834), superordre des Thoraciques et des Acrothoraciques (Thoracica Darwin, 1854– Acrothoracica Gruvel, 1905). J. Forest, Editor, Traité de Zoologie—Anatomie, systématique, biologie 7: Crustacés 2, 453–540. Newman, W.A., 1989. Juvenile ontogeny and metamorphosis in the most primitive living sessile barnacle, Neoverruca, from abyssal hydrothermal springs. B. Mar. Sci. 42, 467–477. Newman, W.A., 1987. Evolution of cirripedes and their major groups. In: Southward, A.J. (Ed.), Barnacle Biology, Crustacean Issues 5, A.A. Belkema, Rotterdam, The Netherlands, pp. 3–42. Newman, W.A., Zullo, V.A., Withers, T.H., 1969. Cirripedia. In: Moore, R.C. (Ed.), Treatise if invertebrate paleontology, Part R Arthropoda 4, 1. Geological Society of America and University of Kansas Press, Lawrence, pp. 206–295. Newman, W.A., Ross, A., 1971. Antarctic Cirripedia. Ant. Res. Ser. 14, 1–257. Nilsson-Cantell, C.A., 1929. Cirripedien des Genus Verruca der Deutschen TiefseeExpedition auf dem Dampfer Valdivia 1898–1899. Zool. Jahrb. 58, 459–480. Nilsson-Cantell, C.A., 1930. Thoracid cirripedes collected in 1925–27. Discovery Rep. 2, 223–260. Nilsson-Cantell, C.A., 1939. Thoracic cirripedes collected in 1925–1936. Discovery Rep. 18, 1–81 28 ﬁg., 3 pls. Oliver, P.G., 1981. The functional morphology and evolution of recent Limopsidae (Bivalvia, Arcoidea). Malacologia 21, 61–93. Pérez-Losada, M., Høeg, J.T., Kolbasov, G.A., Crandall, K.A., 2002. Reanalysis of the relationships among the Cirripedia and the Ascothoracida and the phylogenetic position of the Facetotecta (Maxillopoda: Thecostraca) using 18S rDNA sequences. J. Crust. Biol. 22, 661–669. Pérez-Losada, M., Høeg, J.T., Crandall, K.A., 2004. Unraveling the evolutionary radiation of the thoraciacan barnacles using molecular and morphological evidence: A comparison of several divergence time estimation approaches. Syst. Biol. 53, 244–264. Pérez-Losada, M., Harp, M., Høeg, J.T., Achituv, Y., Jones, D., Watanabe, H., Crandall, K.A., 2008. The tempo and mode of barnacle evolution. Mol. Phylogenet. Evol. 26 382–346. Pilsbry, H.A., 1907. The barnacles (Cirripedia) contained in the collections of the U. S. National Museum. B. US Nat. Mus. 60, 1–122 11 pls. Pilsbry, H.A., 1916. The Sessile Barnacles (Cirripedia) Contained in the Collections of the U.Sl. National Museum; Including a Monagraph of the American Species. U.S. Nat. Mus. B. 93, 1–366. Reisinger, R.R., McIntyre, T., Bester, M.N., 2010. Goose barnacle hitchhike on satellite-tracked southern elephant seals. Pol. Biol. 33, 561–564. Rogers, A.D., 2000. The role of the oceanic oxygen minima in generating biodiversity in the deep sea. Deep-Sea Res. 47 (Pt. II), 119–148. Rogers, A.D., Tyler, P.A., Connelly, D.P., Copley, J.T., James, R., Larter, R.D., Linse, K., Mills, A., Naveira Garabato, R.A., Pancost, A., Pearce, R.D., Polunin, D.A., German, N.V.C., Shank, C.R., Boersch-Supan, T., Alker, P.H., Aquilina, B.J., Bennett, A., Clarke, S.A., Dinley, A., Graham, R.J.J., Green, A.G.C., Hawkes, D.R.H., Hepburn, J.A., Hilario, L., Huvenne, A., Marsh, V.A.I., Ramirez-Llodra, L., Reid, E., Roterman, W.D.K., Sweeting, C.N., Thatje, C.J., Zwirglmaier, K., S., 2012. The Discovery of New Deep-Sea Hydrothermal Vent Communities in the Southern Ocean and Implications for Biogeography. PLoS Biol. 10 (1), e1001234, http://dx.doi.org/10.1371/journal.pbio.1001234. Scheltema, R.S., Scheltema, A.H., Williams, L.P., Halanych, K.M., 2010. Seasonal occurrence of balanomorph barnacle nauplius larvae in the region of the Antarctic Peninsula. J Exp. Mar. Biol. Ecol. 392, 125–128. Setsaas, T.H., Bester, M.N., 2006. Goose barnacle (Lepas australis) infestation of the Subantarctic fur seal (Arctocephalus tropicalis). Afr. Zool. 41, 305–307. Shapiro, B., Rambaut, A., Drummond, A.J., 2006. Choosing appropriate substitution models for the phylogenetic analysis of protein-coding sequences. Mol. Biol. Evol. 23 (1), 7–9. Shimodaira, H., Hasegawa, M., 1999. Multiple comparisons of log-likelihoods with applications to phylogenetic inference. Mol. Biol. Evol. 16, 1114–1116. Spears, T., Abele, L.G., Applegate, M.A., 2007. Phylogenetic study of cirripedes and selected relatives (Thecostraca) based on 18 S rDNA sequence analysis. J. Crust. Biol., 641–656. Suchard, M.A., Weiss, R.E., Sinsheimer, J.S., 2001. Bayesian selection of continuoustime Markov chain evolutionary models. Molecular Biology and Evolution 18 (6), 1001–1013. Taylor, B.J., 1965. Aptian Cirripedes from Alexander Island. Brit. Ant. Surv. B. 7, 37–42. Wills, L.J., 1962. Cyprilepas holmi, a pedunculate cirripede from the Upper Silurian of Oesel, Esthonia. Palaeontology 6, 161–165. Wilson, N.G., Hunter, R.L., Lockhart, S.J., Halanych, K.M., 2007. Multiple lineages and absence of panmixia in the ‘‘circumpolar’’ crinoid Promachocrinus kerguelensis from the Atlantic sector of Antarctica. Mar. Biol. 152, 895–904. Young, P.S., 2002. Revision of the Scalpellidae (Crustacea, Cirripedia) in the collection of the Muséum national d’Histoire naturelle, France, studied by Abel Gruvel. Zoosystema 24, 309–345. Zevina, G.B., 1974. The Cirripedia Thoracica of the Kerguelen Islands. Crustaceana 27, 209–215. Zevina, G.B., 1975. Barnacles, collected during 11th scientiﬁc voyage of RV ‘‘Akademik Kurchatov’’ in the Atlantic sector of Antarctic. Trans. Shirshov Inst. Oceanol. 103, 183–193, in Russian. Zevina, G.B., 1993. Abyssal species of Scalpellidae (Cirripedia, Thoracica) in the Atlantic region of the Antarctic. Trans. P.P. Shirshov Inst. Oceanol. 127 (122–128), in Russian. Zinsmeister, W.J., Feldmann, R.M., 1984. Cenozoic high latitude heterochroneity of southern hemisphere marine faunas. Science 224, 281–283.

RELATED PAPERS

RELATED TOPICS

Log In

Phylogenetic position of Antarctic Scalpelliformes (Crustacea: Cirripedia: Thoracica)

Phylogenetic position of Antarctic Scalpelliformes (Crustacea: Cirripedia: Thoracica)

Related Papers

RELATED PAPERS

RELATED TOPICS