por_033.fm 126 Polar Research 26 2007 126 – 134 © 2007 The Authors shausen Sea and around Peter I IslandJ. S. Troncoso et al. Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island Jesús S. Troncoso, 1 Cristian Aldea, 2 Patrick Arnaud, 3 Ana Ramos 4 & Francisco García 5 1 Departamento de Ecología y Biología Animal, Facultad de Ciencias del Mar, Campus Lagoas Marcosende, 36310, Universidad de Vigo, Spain 2 Departamento de Ecología y Biología Animal, Universidad de Vigo, Spain, and Fundación Centro de Estudios del Cuaternario de Fuego-Patagonia y Antártica (CEQUA), Avenida Bulnes 01890 – Casilla 113-D, Punta Arenas, Chile 3 Le Jas des Batarins, 04110, Vachères, France 4 IEO Centro Oceanográfico de Vigo, Cabo Estai, Vigo, Spain 5 Departamento de Sistema Fisicos, Qvimicos y Naturales, Facultad de Ciencias Experimentales, Universidad Pablo de Olavide, Sevilla, Spain. Abstract Macrobenthic soft-bottom molluscs were sampled in 30 stations located in the Bellingshausen Sea at depths ranging from 90 to 3304 m. The samples were collected using a quantitative grab box-corer during the cruises BENTART 03, from 24 January to 3 March 2003, and BENTART 06, from 2 January to 16 February 2006. Molluscs represent 1074 specimens belonging to 62 species of Polyplacophora, Gastropoda, Bivalvia and Scaphopoda. The bivalve Cyamiocar- dium denticulatum was the most abundant species (448 specimens). The abun- dance per station varied between 1 and 446 specimens. The Shannon–Wiener diversity index ranged between one specimen and 2.36, the Pielou evenness index ranged between 0.00 and 1 and species richness ranged from 1 to 14 species. Diversity showed great variations at different stations. After multi- variate analysis (cluster analysis and nonmetrical multidimensional scaling) based on Bray–Curtis similarities, we were able to separate two principal clusters. The first cluster groups together species from shallower bottoms near Peter I Island and the Antarctic Peninsula, and the second cluster groups together species from deeper bottoms in the Bellingshausen Sea. The combina- tion of environmental variables with the highest correlations with faunistic data was that of depth and coarse sand at the surface. Keywords Soft-bottom; Molluscs; Bellingshausen Sea; Peter I Island. Information about the benthic macrofauna of the Belling- shausen Sea and Peter I Øy (Peter I Island) was scarce. To date, approximately 895 species of gastropods and 379 species of bivalves are known from the Southern Ocean and adjacent regions (Linse et al. 2006); Clarke et al. (2004) mentioned approximately 530 species of gastro- pods and 110 of bivalves found exclusively in the Southern Ocean. Only three families and 11 species were recorded in the Bellingshausen Sea, and two families and three species were recorded at Peter I Øy. This apparently low number of species is the result of poor sampling of this area: two samples on the continental shelf (1– 1000 m), none on the continental slope (1000–3000 m) and six deeper ( > 3000 m) have been obtained in the Bellingshausen and Amundsen seas out of 1490, 98 and 36 respective samples performed in these three zones in the whole Southern Ocean (Clarke et al. 2004). The BENTART 03 (from 24 January to 3 March 2003) and BENTART 06 (from 2 January to 17 February 2006) research programs were carried out on board the RV Hes- perides in the Bellingshausen Sea and off Peter I Øy. Benthic molluscs have a particularly wide ecological and etho- logical spectrum. So the assessment and analysis of their assemblages is likely to contribute to a better understand- ing of the structure and interactions inside the more com- plex benthic assemblages in which they live and interact. As an integrated study of the benthic ecosystem, the BENTART programme is a good opportunity for analysing malacological assemblages. On the basis of the data obtained during this survey, we try to answer the Correspondence Jesús S. Troncoso, Departamento de Ecología y Biología Animal, Facultad de Ciencias del Mar, Campus Lagoas Marcosende, 36310, Universidad de Vigo, Spain. E-mail: troncoso@uvigo.es. doi:10.1111/j.1751-8369.2007.00033.x J. S. Troncoso et al. Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island Polar Research 26 2007 126 – 134 © 2007 The Authors 127 following questions. Is the molluscan fauna really poor in the Bellingshausen Sea and nearby areas? How many molluscan assemblages are living there? What are their species diversity and dominant species? What are their ecological characteristics or requirements? Material and methods Study area During the research cruises BENTART 03 (January– February 2003) and BENTART 06 (January–February 2006) molluscs were collected, together with other benthic invertebrates, at 40 stations in the Bellingshausen Sea, from the Antarctic Peninsula to Thurston Island on the border of the Amundsen Sea (Fig. 1), at depths rang- ing from 90 to 3304 m (Table 1). Because of the weather conditions (big waves, heavy winds) the box-corer can- not work in all sampling stations, and for this reason only 30 stations were sampled and used in the present quanti- tative studies. Sample collection In each station, the samples were collected using a quan- titative grab box-corer (BC) with a maximum break- through of 60 cm and an effective surface of sampling of 25 × 25 cm. Four samples were taken in each station. The first BC was used immediately after sampling to measure temperature, pH and redox (Eh) at 6 and 25 cm below the surface of the sediment, and was subsampled later in the laboratory to analyse the sediment grain size, level of carbonates and content of organic matter. The three rep- lica BC remaining were used to obtain the fauna. The content of each replica was sieved using three mesh sizes (5, 1 and 0.5 mm). The molluscan groups retained in the two largest sieves were counted to estimate their relative abundance in the total macrobenthic fauna. Afterwards, the material was fixed in formalin and preserved in 70% ethanol for further taxonomic study. Using the superficial layer of sediment, the following percentages of granulometric fractions were recorded: coarse sand, medium sand, fine sand and mud. Carbonate content (%) was measured by treatment of the sample with hydrochloric acid, and the total organic matter con- tent (OM%) was estimated from the weight loss of the samples after 4 h at 450 ° C in an oven. Data analysis Data were organized into station by species matrices. Univariate measures were calculated for each sampling Fig. 1 Study area and benthic sampling stations during the BENTART cruises in 2003 and 2006. 128 Polar Research 26 2007 126 – 134 © 2007 The Authors Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island J. S. Troncoso et al. station: total abundance ( N ), number of species ( S ), the Shannon–Wiener diversity index ( H ′ , log e ) and Pielou’s evenness ( J ′ ). Molluscan assemblages were determined through non-parametric multivariate techniques as described by Field et al. (1982) using the PRIMER v5.0 (Plymouth Routines in Multivariate Ecological Research) software package (Clarke & Warwick 1994). A similarity matrix between sampling stations was constructed by means of the Bray–Curtis similarity coefficient by first applying fourth-root transformation on species abun- dance to downweight the contribution of the most abun- dant species. From this matrix, a classification of the stations was performed by cluster analysis based on the group-average sorting algorithm, as well as an ordination by means of nonmetrical multidimensional scaling (MDS). Possible differences in faunistic composition between the two major groups of stations were tested using a one-way ANOSIM test. The SIMPER program was then used to identify species that greatly contributed to the differentiation of station groups. The BIO-ENV procedure was used to research the pos- sible relationship between molluscan distributions in the Bellingshausen Sea and the measured environmental variables. The following variables were considered in this analysis: depth (m), redox (Eh), OM (%) and all granulometric fractions (%). Carbonates were discarded from the analysis because of the lack of such data from BENTART 03. All variables were previously transformed by log( x + 1). Results Abundance, number of species and diversity A total of 1074 molluscan specimens were collected belonging to 62 species in four classes. The best- represented class, in terms of number of species, was Gastropoda (14 families and 33 species), followed by Bivalvia (15 families and 25 species), Scaphopoda (three families and three species) and Polyplacophora (one Table 1 Location, depth and environmental parameters at the surface of sediments of survey stations: redox potential (mV), organic matter (OM%), carbonates (%), gravel (% > 2 mm), coarse sand (% > 0.5 mm), medium sand (% > 0.25 mm), fine sand (% > 0.0625 mm) and mud (% < 0.0625 mm). Station Latitude S Longitude W Depth (m) Redox OM Carbonates Gravel Coarse sand Medium sand Fine sand Mud MB1 (2003) 70 ° 38.22 ′ 95 ° 15.36 ′ 534 252.2 4.81 n.d. 14.30 7.90 7.50 19.10 51.20 MB2 (2003) 70 ° 29.25 ′ 95 ° 14.83 ′ 780 289.3 5.02 n.d. 81.40 1.80 1.10 4.20 11.50 MB3 (2003) 70 ° 17.58 ′ 95 ° 11.86 ′ 1431 259.8 5.42 n.d. 29.00 4.30 3.40 30.50 32.80 MB4 (2003) 70 ° 52.86 ′ 98 ° 26.12 ′ 425 271.3 4.56 n.d. 31.00 9.60 5.40 16.40 37.60 PI 5 (2003) 68 ° 56.70 ′ 90 ° 35.70 ′ 126 199.3 1.43 n.d. 0.14 0.14 0.32 19.50 79.90 PI 6 (2003) 68 ° 49.61 ′ 90 ° 48.78 ′ 210 122.5 1.35 n.d. 0.00 0.10 0.10 21.00 78.80 PI 7 (2003) 68 ° 42.20 ′ 90 ° 40.80 ′ 410 174.8 1.85 n.d. 0.00 0.20 0.20 6.10 93.50 PI 8 (2003) 68 ° 50.18 ′ 90 ° 51.08 ′ 90 155.8 1.23 n.d. 0.10 0.80 4.90 58.90 35.30 MB9 (2003) 70 ° 14.40 ′ 81 ° 47.03 ′ 532 261.8 5.96 n.d. 3.90 6.10 4.40 12.40 73.20 MB10 (2003) 70 ° 44.31 ′ 81 ° 27.85 ′ 497 260 4.05 n.d. 15.80 5.20 7.90 16.40 54.70 MB11 (2003) 69 ° 27.07 ′ 82 ° 06.76 ′ 1289 266 3.81 n.d. 22.40 8.50 3.70 10.60 54.80 MB12 (2003) 69 ° 24.27 ′ 82 ° 11.88 ′ 2032 261.5 5.29 n.d. 23.00 11.10 5.80 18.06 42.04 MB13 (2003) 69 ° 49.56 ′ 77 ° 43.68 ′ 605 240.5 4.64 n.d. 10.20 3.60 4.10 17.50 64.60 MB14 (2003) 69 ° 21.12 ′ 78 ° 04.91 ′ 498 n.d. 3.68 n.d. 34.70 5.10 3.80 11.70 44.70 MB17 (2003) 68 ° 54.88 ′ 78 ° 14.16 ′ 2044 224.7 1.98 n.d. 64.50 18.60 3.90 2.90 10.10 PA21 (2003) 64 ° 54.01 ′ 63 ° 01.11 ′ 107 133.5 2.49 n.d. 4.40 14.80 17.90 34.80 28.10 PA22 (2003) 64 ° 50.58 ′ 62 ° 57.91 ′ 294 137 6.40 n.d. 0.00 0.00 0.30 3.20 96.50 PA23 (2003) 64 ° 55.95 ′ 63 ° 38.40 ′ 655 272.5 6.75 n.d. 0.00 0.50 0.50 7.10 91.90 PA24 (2003) 64 ° 20.11 ′ 61 ° 58.82 ′ 1056 170.5 8.32 n.d. 0.00 0.24 0.23 1.53 98.00 PA25 (2003) 63 ° 52.85 ′ 61 ° 48.52 ′ 110 n.d. 1.16 n.d. 23.80 22.90 13.50 36.30 3.50 MB26 (2006) 70 ° 14.62 ′ 95 ° 02.20 ′ 1920 178.9 1.99 5.87 1.33 11.22 29.09 49.43 8.94 MB29 (2006) 69 ° 26.08 ′ 88 ° 26.17 ′ 3304 262.1 8.92 1.14 1.54 5.56 2.47 5.25 85.19 MB30 (2006) 69 ° 58.98 ′ 87 ° 31.08 ′ 1814 187.7 7.01 2.97 58.38 1.78 1.02 8.88 29.95 MB31 (2006) 69 ° 56.98 ′ 86 ° 19.27 ′ 1426 207.8 5.31 2.54 0.00 2.22 4.81 20.74 72.22 MB33 (2006) 70 ° 15.90 ′ 84 ° 11.45 ′ 438 290.2 4.02 1.38 20.11 12.99 8.86 26.32 31.72 MB34 (2006) 70 ° 08.20 ′ 84 ° 51.68 ′ 603 326 1.80 1.27 0.00 12.91 14.98 59.89 12.21 MB35 (2006) 69 ° 56.03 ′ 85 °11.30′ 1117 260.7 7.36 2.40 47.65 3.78 1.73 9.13 37.72 MB36 (2006) 69°56.28′ 80°24.55′ 560 289 8.51 0.47 33.15 1.08 1.08 3.96 60.72 MB37 (2006) 69°26.38′ 80°51.62′ 495 244 5.70 0.64 35.37 17.04 10.27 16.15 21.17 MB38 (2006) 69°14.08′ 80°61.20′ 1324 298.2 5.98 0.83 65.69 3.14 1.26 2.72 27.20 MB, Bellingshausen Sea; n.d., not determined; PA, Antarctic Peninsula; PI, Peter I Øy. BENTART 2003, January–February 2003; BENTART 2006, January–February 2006. J. S. Troncoso et al. Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island Polar Research 26 2007 126 – 134 © 2007 The Authors 129 Table 2 Systematic list of species found, indicating stations where these were collected. Class/Family Species Station POLYPLACOPHORA Leptochitonidae Leptochiton kerguelenensis Haddon, 1886 MB11, MB14, MB17, PA23 GASTROPODA Anatomidae Anatoma euglypta (Pelseneer, 1903) MB31 Trochidae Antimargarita sp. MB37 Calliotropis pelseneeri Cernohorsky, 1977 MB4 Calliotropis sp. MB30 Margarella antarctica (Lamy, 1905) PI8 Solariella antarctica Powell, 1958 MB11 Submargarita notalis (Strebel, 1908) PA21 family and one species) (Table 2, Fig. 2). The bivalves Thyasira bongraini and Cyamiocardium denticulatum accounted for 60% of all molluscs. The number of molluscs species per sampling site ranged from 1 to 14 (Table 3). Maximal abundances were found at sites PI5 (446) and PI8 (244), and the lowest were found at sites MB1, MB10, MB12 and MB17, with only one specimen in each. In MB26, MB29 and MB34, no molluscs species were obtained. The highest diversity was recorded at site PA21 (H′ = 2.36) in contrast to that found at MB9 (H′ = 0.27). J′ is usually high in sampling sites, except in stations near Peter I Øy. Molluscan assemblages The cluster analysis (Fig. 3) showed the presence of major groups of sites at a similarity level of 20%: group A (six shallow-water stations, 90–410 m in depth, high % of mud and low values of OM) and group B (16 deep- water stations, 438–2044 m in depth, medium values of mud % and high OM). MDS ordination (Fig. 4) showed similar results to those of the dendrogram, with an acceptable stress value (0.05). Two major groups (A and B) are segregated from right to left, which can be identi- fied as the depth gradient from shallow stations off Peter I Øy and the Antarctic Peninsula to deeper stations in the Bellingshausen Sea. The ANOSIM test showed signifi- cant differences on the faunistic composition between groups A and B according to the depth factor: shallow (group A) vs. deep (group B) stations (global R = 0.571, p = 0.001). Results of the SIMPER analysis for the dissimilarity between groups are shown in Table 4. The bivalves Thya- sira bongraini, Cyamiocardium denticulatum, Cuspidaria infe- lix, Adacnarca nitens, Cyclocardia astartoides, Limopsis lilliei and the scaphopod Siphonodentalium dalli f. antarcticus contributed greatly to the similarity (up to a cumulative 92%) in populations found for shallow-water stations of group A. The deeper-water of group B is mainly deter- mined by Dentalium majorinum. The species that contri- bute most to the dissimilarity between the two groups according to ratio values were T. bongraini, D. majorinum, C. infelix and C. denticulatum. Relationship between biotic and environmental variables Sediments were predominantly mud, with relatively high contents of organic matter (Table 1). The superficial sedi- ments appeared to be oxidized, as shown by redox (Eh) values of greater than 122.5 mV. Carbonate content was low in stations of BENTART 06 (MB26, 29, 30, 31, 33, 34,Fig. 2 Number of species and specimens of each molluscan class. 130 Polar Research 26 2007 126 – 134 © 2007 The Authors Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island J. S. Troncoso et al. Zerotulidae Dickdellia labioflecta (Dell, 1990) PA22 Eatoniellidae Eatoniella glacialis (Smith, 1907) PA21 Rissoidae Onoba gelida (Smith, 1907) PI8, PA21 Onoba kergueleni (Smith, 1875) PI8 Capulidae Torellia planispira (Smith, 1915) PA21 Naticidae Amauropsis anderssoni (Strebel, 1906) PA25 Amauropsis aureolutea (Strebel, 1908) MB11 Falsilunatia delicatula (Smith, 1902) PI5, PI8 Eulimidae Balcis antarctica (Strebel, 1908) MB13 Muricidae Trophon cuspidarioides Powell, 1951 PI7 Trophon drygalskii Thiele, 1912 MB31 Trophon longstaffi Smith, 1907 PI5 Buccinidae Chlanidota signeyana Powell, 1951 PI5, PI7, PA24 Neobuccinum eatoni (Smith, 1875) PI5 Pareuthria regulus (Watson, 1882) PI5 Probuccinum costatum Thiele, 1912 PI7 Prosipho chordatus (Strebel, 1908) PI8 Prosipho hedleyi Powell, 1958 PI8 Volutidae Harpovoluta charcoti (Lamy, 1910) PA23 Conidae Belaturricula gaini (Lamy, 1910) PA23 Turridae Leucosyrinx paratenoceras Powell, 1951 MB36 Lorabela sp. PI5, PI8 Pleurotomella simillina Thiele, 1912 PI7 Typhlodaphne innocentia Dell, 1990 PI5, PI8 Acteonidae Acteon antarcticus Thiele, 1912 MB3 Neactaeonina cf. edentula (Watson, 1883) PI5 BIVALVIA Nuculanidae Propeleda longicaudata (Thiele, 1912) PA21, MB33 Yoldiidae Yoldia eightsi (Couthouy, 1839) PA22 Yoldiella antarctica (Thiele, 1912) PI7 Yoldiella ecaudata (Pelseneer, 1903) MB3, MB36 Yoldiella oblonga (Pelseneer, 1903) MB9, MB14 Yoldiella profundorum (Melvill & Standen, 1912) PA22 Malletiidae Malletia pellucida Thiele, 1912 MB31, MB35 Arcidae Bathyarca sinuata Pelseneer, 1903 MB3, MB13, MB30, MB35, MB38 Limopsidae Limopsis lilliei Smith, 1915 PA21, PA25 Limopsis longipilosa Pelseneer, 1903 MB33, MB37 Philobryidae Philobrya sublaevis (Pelseneer, 1903) PI8 Adacnarca nitens Pelseneer, 1903 PI5, PI8, PA21 Mytilidae Dacrydium albidum Pelseneer, 1903 MB38 Limidae Limatula simillina Thiele, 1912 MB37 Pectinidae Adamussium colbecki (Smith, 1902) MB2, MB3, PI5, PI7, MB30, MB31, MB35 Thyasiridae Thyasira bongraini (Lamy, 1910) PI5, PI6, PI7, PI8, PA21, PA22, PA25 Thyasira dearborni Nicol, 1965 PA22 Lasaeidae Mysella antarctica (Smith, 1907) PA22 Pseudokellya cardiformis (Smith, 1885) PA21 Cyamiidae Cyamiocardium denticulatum (Smith, 1907) MB4, PI5, PI7, PI8, PA21 Cyamiomactra laminifera (Lamy, 1906) PI8 Carditidae Cyclocardia astartoides (Martens, 1878) PA21, PA25 Cyclocardia cf. intermedia Thiele, 1912 MB4 Thraciidae Thracia meridionalis Smith, 1885 PA21 Cuspidariidae Cuspidaria infelix Thiele, 1912 PI5, PI7, PI8, PA21 SCAPHOPODA Dentaliidae Dentalium majorinum Mabille & Rochebrune, 1891 MB1, MB2, PI7, MB9, MB10, MB13, MB14, MB31, MB33, MB35, MB36, MB38 Rhabdidae Rhabdus cf. perceptus (Mabille & Rochebrune, 1891) PI7, MB12 Gadilidae Siphonodentalium dalli f. antarcticus (Odhner, 1931) PI5, PI6, PI7, PA24 Class/Family Species Station Table 2 Continued J. S. Troncoso et al. Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island Polar Research 26 2007 126 – 134 © 2007 The Authors 131 35, 36, 37, 38 and 39). The depths in the sampling sites of the Bellingshausen Sea varied from 90 to 3304 metres. The ANOSIM test also showed differences in faunistic com- position among stations following a gradient of depth. The BIO-ENV procedure (Table 5) showed that the combination of environmental variables with the highest correlations with faunistic data was depth and coarse sand at the surface. Redox (Eh) and fine sand at the surface were not involved in the best combinations. Depth was the variable with the best value when each variable was considered alone (pw = 0.390). Discussion Linse et al. (2006) divided the Southern Ocean into seven subregions based on richness hotspots of gastropods and bivalves: Antarctic Peninsula, Weddell Sea, three East Antarctic subregions (Dronning Maud Land, Enderby Land and Wilkes Land), Ross Sea, and the independent Scotia Arc and sub-Antarctic islands, and excluded the West Antarctic Region (Eights Coast and the Belling- shausen and Amundsen seas), previously defined by Clarke et al. (2004) from this classification because it is insufficiently studied. So, the present study, together with future more comprehensive faunistic studies (Aldea & Troncoso, unpubl. data), represents notable progress in the knowledge of the Bellingshausen Sea. For neighbouring areas, several surveys of semiquanti- tative (e.g. Arnaud et al. 1998) and quantitative analyses of molluscan or macrobenthic fauna of the soft bottom have appeared in recent literature (e.g. Sáiz-Salinas et al. 1997; Sahade et al. 1998; Arnaud et al. 2001) showing distribution patterns and assemblages of several species, mainly from north-west of the Antarctic Peninsula. Like- wise, similar studies about the Ross Sea have previously been published (e.g. Russo & Gambi 1994; Gambi et al. 1997). Only one survey of the benthic marine fauna of Table 3 Faunistic parameters at each station: species richness (R), total abundante (N), Pielou even ness index ( J′) and Shannon–Wiener diversity index calculated with the natural logarithm (H′), MO data (***). Station Richness (S) Abundance (N) Evenness (J′) Diversity(H′) MB1 1 1 *** 0 MB2 2 2 1 0.69 MB3 4 5 0.96 1.33 MB4 3 3 1 1.10 PI5 14 446 0.40 1.06 PI6 2 145 0.72 0.50 PI7 12 67 0.58 1.45 PI8 14 244 0.49 1.29 MB9 2 13 0.39 0.27 MB10 1 1 *** 0 MB11 3 3 1 1.10 MB12 1 2 *** 0 MB13 3 3 1 1.10 MB14 3 7 0.87 0.96 MB17 1 1 *** 0 PA21 13 22 0.92 2.36 PA22 6 55 0.91 1.63 PA23 3 3 1 1.10 PA24 2 2 1 0.69 PA25 4 7 0.92 1.28 MB30 3 8 0.82 0.90 MB31 5 7 0.96 1.55 MB33 3 3 1 1.10 MB35 4 4 1 1.39 MB36 3 7 0.72 0.80 MB37 3 7 0.72 0.80 MB38 3 6 0.92 1.01 Fig. 3 Faunistic assemblages in the study are determined by cluster analysis based on Bray- Curtis Similarity. 132 Polar Research 26 2007 126 – 134 © 2007 The Authors Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island J. S. Troncoso et al. Table 4 Results of SIMPER analysis. Species were ranked according to their average contribution to dissimilarity (Contrib. %) between groups of stations. Average abundance for each group (Av. abund.), average dissimilarity (Av. diss.), ratio value (dissimilarity/standard deviation, Diss./SD) and percentage of cumulative dissimilarity (Cum. %). Groups B & A (Average dissimilarity = 96.79) Group B Group A Species Av. abund. Av. abund. Av. diss. Diss./SD Contrib. % Cum. % Thyasira bongraini 0 2.36 16.46 1.54 17.01 17.01 Dentalium majorinum 1.21 0.31 8.47 1.54 8.75 25.76 Cyamiocardium denticulatum 0 1.64 7.17 1.12 7.41 33.17 Siphonodentalium dalli f. antarcticus 0 0.86 6.72 0.67 6.94 40.11 Cuspidaria infelix 0 1.03 4.74 1.36 4.9 45.01 Cyclocardia astartoides 0 0.4 3.87 0.62 4 49.01 Limopsis lilliei 0 0.33 3.26 0.62 3.37 52.38 Amauropsis anderssoni 0 0.22 3.01 0.44 3.11 55.49 Adacnarca nitens 0 0.65 2.9 0.98 2.99 58.48 Adamussium colbecki 0.29 0.33 2.76 0.72 2.85 61.33 Bathyarca sinuata 0.29 0 2.04 0.53 2.11 63.44 Onoba gelida 0 0.39 1.86 0.7 1.92 65.37 Chlanidota signeyana 0 0.36 1.75 0.67 1.81 67.18 Falsilunatia delicatula 0 0.4 1.62 0.68 1.67 68.85 Lorabela sp. 0 0.39 1.55 0.69 1.6 70.45 Typhlodaphne innocentia 0 0.36 1.47 0.7 1.51 71.97 Propeleda longicaudata 0.09 0.17 1.46 0.49 1.51 73.48 Yoldiella oblonga 0.18 0 1.31 0.41 1.36 74.83 Torellia planispira 0 0.22 1.27 0.44 1.32 76.15 Thracia meridionalis 0 0.22 1.27 0.44 1.32 77.47 Pleurotomella simillina 0 0.22 1.21 0.44 1.25 78.71 Malletia pellucida 0.18 0 1.17 0.42 1.2 79.92 Neactaeonina cf. edentula 0 0.26 1.03 0.44 1.07 80.99 Submargarita notalis 0 0.17 0.97 0.44 1 81.99 Eatoniella glacialis 0 0.17 0.97 0.44 1 82.99 Pseudokellya cardiformis 0 0.17 0.97 0.44 1 83.99 Philobrya sublaevis 0 0.24 0.96 0.44 0.99 84.98 Trophon cuspidarioides 0 0.17 0.92 0.44 0.95 85.92 Probuccinum costatum 0 0.17 0.92 0.44 0.95 86.87 Yoldiella antarctica 0 0.17 0.92 0.44 0.95 87.82 Rhabdus cf. perceptus 0 0.17 0.92 0.44 0.95 88.77 Margarella antarctica 0 0.2 0.81 0.44 0.83 89.6 Prosipho hedleyi 0 0.2 0.81 0.44 0.83 90.44 Fig. 4 MDS ordination of faunistic assemblages. The dotted line represents the separation of two main groups (A, B). The subgroups delimited with solid lines are derivates of cluster analysis. J. S. Troncoso et al. Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island Polar Research 26 2007 126 – 134 © 2007 The Authors 133 the Bellingshausen Sea has been published previously (Richardson & Hedgpeth 1977). The main species recorded in the present survey are the same as those recorded during a similar survey in the South Shetland Islands and Bransfield Strait (Arnaud et al. 2001): the tiny Cyamiocardium denticulatum and Thyasira bongraini, and the large Dentalium majorinum, Siphonodentalium dalli f. antarcticus, Cuspidaria infelix and Yoldia eightsi. Mühlenhardt-Siegel (1989) also found a high presence of T. bongraini. The same situation occurs in the Ross Sea, where Cattaneo-Vietti et al. (2000) found mainly the bivalves Adamussium colbecki, Galeommatidae undet., Y. eightsi and T. dearborni (named as Genaxinus debilis). Molluscan assemblages exhibit a bathymetric pattern from deep-water stations (depth > 438 m) to shallow- water ones (depth < 410 m). Deep-water stations with very low abundances (<13 specimens) mainly belong to group B, which is widely dominated by the large D. majorinum, followed by the bivalves A. colbeki, Bathyarca sinuata and the tiny species Limopsis longipilosa and the chiton Leptochiton kerguelenensis. On the contrary, shallow-water stations with high abundances (22–446 specimens), belonging mainly to group A, and mostly located near the Antarctic Peninsula or Peter I Øy, are dominated by the tiny C. denticulatum and T. bongraini, followed by the larger scaphopod S. dalli f. antarcticus, C. infelix and Y. eightsi. This result agrees with those of studies carried out in adjacent areas at depths generally shallower than 500 m, where Arnaud et al. (2001) have reported the high presence and abundances of the bivalves C. denticulatum and T. bongraini. However, this result disagrees with the survey of Cattaneo-Vietti et al. (2000), who found mainly large bivalves (A. colbecki, Y. eightsi) in depths of less than 380 m, and the tiny infaunal bivalve T. dearborni (under the name G. debilis) at depths of 222–1000 m. This is an inverse pattern, with tiny infaunal molluscs dominating at greater depths. In conclusion, the main molluscan species of the Ross Sea (Schiaparelli et al. 2006), Weddell Sea (Hain 1990) and the South Shetland Islands (Arnaud et al. 2001) are also present in the Bellingshausen Sea, but in lesser abundance. Peter I Øy presents the highest values of abundances and number of species. T. bongraini and C. denticulatum contribute 60% of the total number of the specimens. Depth and percentage of coarse sand could be the environmental variables that explain the distribution of molluscan assemblages in the Belling- shausen Sea. Acknowledgements This research has been supported by the Spanish Govern- ment through the Ministry of Education and Science. The officers and crew of the RV Hesperides and our colleagues from the BENTART cruises in 2003 and 2006 played a prominent part in the success of this project, and we express our gratitude to all of them. We are also grateful to Juan Moreira and two anonymous referees for their helpful comments that improved the manuscript. References Arnaud P. M., López C.M., Olaso I., Ramil F., Ramos-Esplá A.A. & Ramos A. 1998. Semi-quantitative study of macrobenthic fauna in the region of the South Shetland Islands and the Antarctic Peninsula. Polar Biology 19, 160–166. Arnaud P. M., Troncoso J. S. & Ramos A. 2001. Species diver- sity and assemblages of macrobenthic Mollusca from the South Shetland Islands and Bransfield Strait (Antarctica). Polar Biology 24, 105–112. Cattaneo-Vietti R., Chiantore M., Schiaparelli S. & Albertelli G. 2000. Shallow- and deep-water mollusc distribution at Terra Nova Bay (Ross Sea, Antarctica). Polar Biology 23, 173–182. Clarke A., Aronson R.B., Crame J.A., Gili J.M. & Blake D.B. 2004. Evolution and diversity of the benthic fauna of the Southern Ocean continental shelf. Antarctic Science 16, 559– 568. Clarke K.R. & Warwick, R.M. 1994. Changes in marine com- munities: an approach to statistical analysis and interpretation. Plymouth: Natural Environmental Research Council, Plymouth Marine Laboratory. Field J. G., Clarke K.R. & Warwick, R.M. 1982. A practical strategy for analysing multispecies distributions patterns. Marine Ecology Progress Series 8, 37–52. Gambi M.C., Castelli A. & Guizzardi M. 1997. Polychaete populations of the shallow soft bottoms off Terra Nova Bay (Ross Sea, Antarctica): distribution, diversity and biomass. Polar Biology 17, 199–210. Hain S. 1990. Die beschalten Mollusken (Gastropoda und Bivalvia) des Weddellmeeres, Antarktis. (The shell-bearing Table 5 Best combinations of variables obtained through BIO-ENV analy- sis according to the values of the Spearman’s rank correlation (pw) for survey stations: depth (m), organic matter (OM %), gravel (% > 2mm), coarses and (% > 0.5mm), mediums and (% > 0.25 mm) and mud (% < 0.0625 mm). Number of variables Correlation (pw) Best variable combination 2 0.454 depth; coarse sand 3 0.410 depth; OM; coarse sand 1 0.390 Depth 3 0.385 depth; gravel; coarse sand 3 0.382 depth; coarse sand; mud 4 0.378 depth; OM; coarse sand 3 0.378 depth; coarse sand; medium sand 4 0.368 depth; OM; coarse sand; medium sand 4 0.362 depth; OM; coarse sand; mud 3 0.359 depth; OM; gravel 134 Polar Research 26 2007 126 – 134 © 2007 The Authors Quantitative analysis of soft-bottom molluscs in the Bellingshausen Sea and around Peter I Island J. S. Troncoso et al. Mollusca [Gastropoda and Bivalvia], Weddell Sea, Antarctica.) Berichte zur Polarforschung 70, 1–181. Linse K., Griffiths H.J., Barnes D.K.A. & Clarke A. 2006. Biodi- versity and biogeography of Antarctic and sub-Antarctic mollusca. Deep-Sea Research II 53, 985–1008. Mühlenhardt-Siegel U. 1989. Quantitative investigations of Antarctic zoobenthos communities in winter (May/June) 1986 with special reference to the sediment structure. Archiv für Fischereiwissenschaft 39, 123–141. Richardson M.D. & Hedgpeth J.W. 1977. Antarctic soft bottom, macrobenthic community adaptations to a cold, stable, highly productive glacially affected environment. In G.A. Llano (ed.): Adaptations within Antarctic ecosystems. Washing- ton DC: The Smithonian Institution. Russo G.F. & Gambi M.C. 1994. First quantitative data on coastal soft bottoms populations off Terra Nova Bay (Ross Sea, Antarctica): bivalve molluscs. National Scientific Commit- tee, Antarctic Oceanographic Campaign 1989–90, Data Report (1994) II. Genoa: Ministry of University, Scientific and Tech- nological Research. Sahade R.J., Tatian M., Kowalke J., Kühne S. & Esnal G.B. 1998. Benthic faunal associations on soft substrates at Potter Cove, King George Island, Antarctica. Polar Biology 19, 85–91. Sáiz-Salinas J.I., Ramos A., García F.J., Troncoso J.S., San Martin G., Sanz C. & Palacin C. 1997. Quantitative analysis of macrobenthic soft-bottom assemblages in South Shetland waters (Antarctica). Polar Biology 17, 393–400. Schiaparelli S., Lörz A. & Cattaneo-Vietti R. 2006. Diversity and distribution of mollusc assemblages on the Victoria Land coast and the Balleny Islands, Ross Sea, Antarctica. Antarctic Science 18, 615–631.