por_003.fm 48 Polar Research 26 (2007) 48 – 63 © 2007 The Author Abstract This paper presents a survey of the mollusc fauna in Kongsfjorden, an Arctic glacial fjord in Spitsbergen, Svalbard, based on 197 samples collected with van Veen grabs, dredges, scuba-diving collections and baited traps at depths ranging from 5 to 390 m. Eighty-seven mollusc species were recorded. The species distribution accords well with the distribution of the main substrata: barren rock, kelp bed, gravel and soft bottom. For the most common substrate type, the soft bottom, the distribution and diversity of molluscs were analysed in relation to environmental factors. Glacial activity (particularly the inflow of glacial meltwater loaded with mineral solids) is responsible for the main gradients of environmental variables in the fjord. Silt concentration in sedi- ments, the water temperature near the bottom and inorganic suspensions in the surface water best predict the species distribution of the soft bottom. Two faunal associations located in glacial bays and three faunal associations in the central basin of the fjord can be distinguished for the fauna of the soft bottom. Molluscs are much more abundant in glacial bays (200–300 individuals (ind.) 0.1 m 2 ) than in the central basin assemblages (30–40 ind./0.1 m 2 ). Yoldiid ( Yoldiella solidula , Y. lenticula and Yoldia hyperborea ) and thyasirid bivalves ( Thya- sira dunbari , T. gouldi and Axinopsida orbiculata ) cope particularly well with glacial sedimentation and occur in high quantities in glacial bays. Although there is no effect of glacial disturbance on the molluscan sample species richness and species diversity, there are significant clines of evenness and taxonomic distinctness in areas near to the glacier. The patterns of molluscan diversity are not fully consistent with the patterns described for complete macrobenthic communities. Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard): a species list and patterns of distribution and diversity Maria W odarska-Kowalczuk Institute of Oceanology, Polish Academy of Sciences, Sopot, Poland l Keywords Mollusca; glacial fjord; species list; species distribution; diversity; Svalbard. Kongsfjorden is situated on the west coast of Spitsbergen, which is the largest island of the Svalbard archipelago. Despite its high latitude location (79 ° N) the fjord has a sub-Arctic character (Hop et al. 2002). The outer and central basins of Kongsfjorden are influenced by the warm Atlantic waters of the West Spitsbergen Current (Svendsen et al. 2002). Three tidal glaciers terminate in the fjord. Kongsbreen, situated in the innermost part of the fjord, is the most active glacier in the Svalbard archi- pelago (Lefauconnier et al. 1994). There is an interna- tional research centre in the settlement of Ny-Ålesund, which makes Kongsfjorden one of the most intensively studied high latitude fjords. The physical and biological settings of the fjord have been reviewed by Svendsen et al. (2002) and Hop et al. (2002). The Implementation and Networking of Large-scale Long-term Marine Biodi- versity in Europe (BIOMARE) project selected Kongs- fjorden as one of the European Marine Biodiversity Research Sites, which provide reference localities for large-scale European biodiversity studies (Warwick et al. 2003). The biota inventory in selected sites is a crucial starting point for large-scale studies of temporal and spatial changes of coastal biodiversity (Stork et al. 1996). Species richness records are largely dependent on the sampling effort (Magurran 2004). The assessment of species rich- ness of a marine bay, an inlet or an island requires inten- sive sampling using a variety of sampling techniques, and preferably covering several sampling seasons (Bouchet et al. 2002). Correspondence M. W odarska-Kowalczuk, Institute of Oceanology, PAS, Powstancow Warszawy 55, 81-712 Sopot, Poland. Email: maria@iopan.gda.pl doi:10.1111/j.1751-8369.2007.00003.x l M. Wl odarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 49 Molluscs are widely distributed and can be abundant in a variety of marine habitats ranging from rocky shores to abyssal muddy plains. Molluscs can be highly responsive to local human activities (Terlizzi et al. 2005) as well as to either long-term or large-scale climatic changes (Miesz- kowska et al. 2006), and have been proposed as possible surrogates for the study of the distribution and diversity of the whole macrobenthic community (Anderson et al. 2005; Smith 2005). In Arctic fjords molluscs exhibit the highest preservation potential among the marine mac- robenthic biota (Aitken 1990). Between 30 and 60% of the modern Arctic mollusc fauna is represented in Qua- ternary marine Arctic macrofossils, and understanding the environmental factors controlling the distribution of living communities is crucial for reconstructing the palae- oecology of Quaternary glaciations and interglaciations (Gordillo & Aitken 2000). The macrobenthic communities of Kongsfjorden have been described in several publications. Kendall et al. (2003), Wl odarska-Kowalczuk & Pearson (2004), Wl o- darska-Kowalczuk et al. (2005) and Somerfield et al. (2006) studied the distribution and diversity of subtidal communities of the soft bottom at depths ranging from 30 to 300 m. The shallow subtidal macrofauna has been described by Kaczmarek et al. (2005) and Bick & Arlt (2005). Lippert et al. (2001) studied the phytophylous macrofauna at a single location close to the island of Blomstrandhalvøya. Jorgensen & Gulliksen (2001) investigated the fauna of the hard bottom at the Kvade- huken at a depth of 20–30 m using a suction sampler and underwater photography. Rozycki (1991) published a list of molluscs found at 12 stations sampled with grabs and dredges. Those studies were limited in scope to a single location, habitat or depth range within the fjord. The published information on mollusc diversity and distribution in Kongsfjorden is scattered and incomplete. In the present study I bring together extensive distribu- tional data of molluscs collected in several benthic sur- veys carried out in Kongsfjorden by the Institute of Oceanology, Polish Academy of Sciences, in 1996–2000. The material contains 197 samples taken with a variety of sampling gear at a range of depths and habitats, and is likely to give a reliable assessment of the mollusc diversity and distribution patterns in the fjord. The aims of the paper are to (1) present a mollusc species list for Kongsf- jorden; (2) identify the environmental predictors of the mollusc species distribution; (3) identify the species characteristic for fjord habitats/mollusc associations; (4) quantify the response of molluscs at the soft bottom to glacial disturbance and compare it with the perturbation signal in the whole macrobenthic community (as described by Wl odarska-Kowalczuk et al. 2005). Study area Kongsfjorden is an open fjord located on the north-west coast of Spitsbergen in the Svalbard archipelago (12 ° E, 79 ° N; Fig. 1). The fjord is 26 km long, with an area of 231 km 2 . Depths rarely exceed 400 m. Two tidal glaciers (Kongsbreen and Conwaybreen) terminate in the inner basin, whereas a third one, Blomstrandbreen, is situated on the northern mid-fjord coast. The hydrology of Kongsfjorden is shaped by the inter- play among: (1) warm saline Atlantic waters entering the fjord from the shelf [temperature ( T ) > 1 ° C, salinity ( S ) > 34.7]; (2) local fjord waters ( T < 1 ° C, salinity S > 34.4); (3) freshwater input from the glaciers (Svend- sen et al. 2002). In the inner basin there are deep depres- sions in which very cold winter waters ( T < − 0.5 ° C, S > 34.4), which are the remnants of winter cooling and deep convection, can be also observed (Svendsen et al. 2002). Fairly uniform fine sediments cover much of the Kongsfjorden subtidal region, particularly in the inner basin where high sediment deposition rates are accompanied by relatively weak currents (Wl odarska- Kowalczuk & Pearson 2004). Gravel beds occur in some parts of the outer fjord where strong bottom cur- rents winnow out fine sediments. Ice rafted pebbles and stones (drop-stones), typical of Arctic glaciated fjords (Dale et al. 1989), are distributed all over the fjord. At the edge of the fjord there are rocky shelves (Jorgensen & Gulliksen 2001) that gradually change into mixed bottom (hard bottom with macroalgae and bedrock with pockets of soft sediment) in the shallow subtidal region in the middle of the fjord (Kaczmarek et al. 2005). Kelp forests ( Alaria esculenta, Laminaria saccharina and L. digi- tata ) are restricted to sites sheltered from drifting ice- bergs and with limited grazing by sea urchins (Lippert et al. 2001; Hop et al. 2002). The meltwater inflows from the Kongsfjorden glaciers result in a steep gradient in both the concentration of mineral suspensions and the sedimentation rates in the water column (Svendsen et al. 2002). The sediment accu- mulation rate decreases by about one order of magnitude from the Kongsbreen glacial bay (20 000 g m − 2 a − 1 ) to the central part of the fjord (1800–3800 g m − 2 a − 1 ) and, again, by another order of magnitude towards the outer fjord (200 g m − 2 a − 1 ; Svendsen et al. 2002). Intensive sedimen- tation results in the formation of unconsolidated labile sediments (Syvitski et al. 1987). Near-glacier sediments may be scoured by icebergs (of up to 10 m in height) that either circulate or stay anchored in the inner basin (Dowdeswell & Forsberg 1992). The organic matter supply to benthic biota decreases towards the inner basin, where primary productivity is 50 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wl odarska-Kowalczuk limited by high levels of water turbidity and where the available organic matter is diluted by the huge mineral sedimentation (Gorlich et al. 1987). The outer fjord and shelf waters are influenced by Atlantic waters and have an enhanced marine organic matter supply (Winkelman & Knies 2005). The particulate organic carbon (POC) concentration in Kongsfjorden sediments increases along the fjord axis from 0.1–0.2 mg g − 1 in the Kongsbreen gla- cial bay and 0.3–0.5 mg g − 1 in Blomstrandbreen glacial bay to 1.1–2.4 mg g − 1 in the outer fjord and shelf sediments (Fetzer et al. 2002; Wl odarska-Kowalczuk & Pearson 2004). Materials and methods Material was collected during cruises with the RV Oce- ania and land expeditions based at the Norwegian Polar Institute station in Ny-Ålesund during the summer seasons from 1996 to 2000. Samples were collected throughout the fjord (Fig. 1) using a range of techniques (Table 1). Eighty quantitative samples were taken using van Veen grabs, 117 samples were collected using quali- tative methods (dredges, scuba-diving and baited traps). Three replicate samples were collected at van Veen grab and scuba-diving stations (with the exception of a few van Veen grab stations where only one replicate was taken because of navigational difficulties). One replicate was taken at each station sampled using either an epibenthic sledge or baited traps. Either five or six sam- ples from sites situated along the depth transect (5, 10, 15, 20, 30 and 50 m) were taken at six stations using a small rectangular dredge. The small dredge, epibenthic sledge and baited traps were constructed with 1 mm mesh size nets. Samples were sieved on either 0.5 mm mesh (van Veen grabs and samples collected by a diver) or 1 mm mesh (all samples from dredges and baited traps). All animals were sorted, identified to the lowest possible taxonomic level and counted. Some individuals could not be identified to species level as either the shells were damaged or the organisms were juvenile and specific traits were not fully developed. The species names and taxonomic affiliations are consistent with the European Register of Marine Species (available at www.marbef.org/data/erms.php). The number of observed species ( S obs ) was plotted as a function of the sampling effort. A species accumula- tion curve with 95% confidence intervals was com- puted using the formulae of Colwell et al. (2004). Two alternative approaches can be applied to estimate true species richness: the extrapolation of the accumulation curve (to predict an asymptote); non-parametric tech- niques based on the concept that rare and uncommon species carry information about the number of species missing in samples. The Michaelis-Menten ( MM ) func- tion was used to generate an asymptotic curve that fit- ted the species accumulation curve (Magurran 2004). Chao2 , a non-parametric estimator based on species occurrence data ( Chao2 = S obs + Q 1 2 /2 Q 2 , where Q 1 is the number of species that occur in just one sample and Q 2 is the number of species that occur in exactly two sam- ples) was also calculated (Chao 2004). The MM asymp- tote estimation and Chao2 , along with log-linear 95% Figure 1 Location of sampling stations. Kvadehuken Kapp Guissez Blom strandbreen Conwaybreen K o n g s b r e e n Blomstrand- halvøya Greenland Svalbard Norway Kongsfjorden 0o 70oN 80oN van Veen grabs (38-380 m) scuba diving (10 m) epibenthic sledges (30-390 m) baited traps (5-60 m) transects - shallow water dredging (5-50 m) glacier front Ny- Ålesund M. Wl odarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 51 confidence intervals as suggested by Chao (1987), were computed using E STIMATE S (Colwell 2005). Constrained ordination techniques were used to explore the relationship between mollusc species distribution and environmental settings (ter Braak & Smilauer 2002). The data collected using quantitative (van Veen grabs) and qualitative (dredges and scuba-diving) methods were treated separately. The material obtained with baited traps differed much from other samples and was not considered in the analyses. Only presence/absence data were used for qualitative material. The quantitative data (van Veen grabs) were square-root transformed prior to analyses. The set of environmental variables used in the analyses of qualitative mollusc samples included depth, longitude (as a proxy for distance to the glacier) and six nominal variables, namely, presence of rock, mud, sand, gravel, kelp and drop stones. The analyses of the grab data were performed only with data from a 1998 benthic cruise. The following environmental variables were recorded: depth, longitude (as a proxy for distance to the glacier), temper- ature and salinity of the water near the bottom, concen- tration of mineral and organic suspensions in surface waters, occurrence of drop stones, percentage of gravel, sand, silt and clay, mean grain size and the concentration of POC and the POC : particulate organic nitrogen concen- tration in sediment (PON) ratio in sediments. The results of a preliminary de-trended correspondence analysis (DCA) pointed to canonical correspondence analyses (CCA) as the most appropriate for the qualitative data, and to redundancy analyses (RDA) as the best suited for the grab data (ter Braak & Smilauer 2002). The forward selection of environmental variables was used to identify and rank their importance of variables in deter- mining the species distribution (ter Braak & Smilauer 2002). The forward selection method performs the fol- lowing functions:it estimates the fit of each variable sepa- rately (marginal effects); it selects the best variable; and it ranks all the remaining variables on the basis of the fit that each variable gives in conjunction with the selected variable(s) (conditional effects). In the case of grab data, a subset of best fitted variables was selected. The selection of best fitted variables was performed via repeated for- ward selection analyses and the progressive elimination of variables with non-significant conditional effects until a set of variables with only significant effects was attained. The significance of the ordination axis in CCA and in RDA, and the effects of environmental variables, were tested employing Monte Carlo permutation tests using 499 unrestricted permutations. The analyses were carried out using C ANOCO v. 4.5 software (ter Braak & Smilauer 2002). Clustering (using group-average linking) of Bray- Curtis similarities was used to visualize the patterns of mollusc distribution. Again presence/absence qualitative data and square root transformed quantitative data were used. Based on the resulting dendrograms, groups of sam- ples of similar composition (mollusc associations) were distinguished. The frequency, dominance, average abundance and indices of the community fidelity (Salzwedel et al. 1985) of each species were calculated for mollusc associations defined by multivariate analyses. The indices of fidelity included DAS (degree of association regarding stations, i.e. the number of stations within the association at which the species concerned occurred as the percentage of the total number of stations at which this species occurred) and DAI (degree of association regarding individuals, i.e. the number of individuals of the species concerned within the group as a percentage of the number of indi- viduals of that species found in the whole study area) (Salzwedel et al. 1985). The following criteria have been used to select the species that are the best descriptors of the associations on the soft bottom: typical species (frequency > 75%, dominance > 2%) and characteristic species (typical + either DAI or DAS > 60%). In the case of qualitative data only frequency and DAS was considered, and species of frequency > 25% and DAS > 60% were regarded as characteristic species. Table 1 Sampling effort and basic information on samples used in the Kongsfjorden mollusc study. The table includes references to papers with details of sampling methods and the locations of stations. Sampling year Sampling gear Number of stations Number of samples Depths (m) Reference 1997 van Veen grab, catching area of 0.1 m 2 14 33 38–380 Wl odarska-Kowalczuk & Pearson 2004 1998 16 47 40–355 1996 Epibenthic sledge 9 9 30–300 — 1997 8 8 35–390 1999 12 12 35–300 2000 5 5 140–330 1999 Small rectangular dredge (80 × 30 cm) 6 32 5–50 Kaczmarek et al. 2005 2001 Collection by diver, frame 0.25 m 2 9 27 10 Kuklinski & Porter 2004 1998–2000 Baited traps 24 24 5–60 Legezynska 2001, 2002 52 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wl odarska-Kowalczuk The differences in N (number of individuals per sample) between associations distinguished based on grab samples were tested using the non-parametric Kruskal-Wallis test, as even after transformation variances could not be homogenized. Post-hoc testing of differences between pairs of associations was performed using pair-wise Mann-Whitney U tests. Species richness defined as the total number of species in a sample ( SP ), species diversity measured using the Shannon-Wiener index ( H ) and the evenness (equitability) of distribution of individuals between species as estimated by the Pielou index ( J ) were calculated for quantitative samples. The differences in SP , H and J between the associations were tested using a one- way ANOVA . The Fisher’s LSD (least significant difference) tests were used for post-hoc multiple comparisons. To explore the taxonomic diversity of samples, two measures of taxonomic distinctness were calculated: avTD (average taxonomic distinctness of presence/absence data) and varTD (variation in taxonomic distinctness) (Clarke & Warwick 2001). The taxonomic diversity mea- sures differ from other metrics of species richness and species diversity, because they include information on the taxonomic position of species. avTD describes the average taxonomic distance (the “path length” between two spe- cies following Linnean taxonomy) of all the species in the association. varTD is defined as the variance of the taxo- nomic distances between all pairs of species in the associ- ation. Five taxonomic levels were used in calculations: species, genus, family, order and class, and equal step levels between successive taxonomic levels were assumed. The differences between soft-bottom associa- tions in AvTD and varTD were identified with the use of a one-way ANOVA. The post-hoc testing was carried out using Fisher’s LSD tests. Results The material contained 35 398 individual molluscs rep- resenting 87 species of 56 genera and 43 families. One species of Caudofoveata, five species of Polyplacophora, 38 species of Gastropoda and 43 species of Bivalvia were recorded. A full taxonomic list is presented in the Appendix. The species accumulation curves tended to stabilize towards an asymptotic value in the cases of both the observed and the estimated numbers of species (Fig. 2). The MM estimator of the asymptotic value of total species richness gave 93 species. Chao2 gave an estimate of 99 species (with 95% confidence intervals from 93 to 117). The shallow waters of the Kongsbreen glacial bay were very poor in the numbers of molluscs: samples collected by scuba-diving at 10 m and in baited traps deployed at depths from 5 to 50 m yielded no mollusc specimens. Samples collected with small dredges at 5–15 m con- tained only a few specimens. Samples from baited traps located along the coast in the central and outer parts of the fjord contained only scavenging buccinid gastropods (Buccinum glaciale, B. undatum, B. scalariforme, B. polare and Colus kroeyeri). Species–environment relationship In the CCA of qualitative data the first two ordination axes explained about 50% of the variance of the species– environment relationship. The first canonical axis was significant at P = 0.002, the second axis was not signifi- cant (P > 0.05). The occurrence of rock and mud was significantly correlated to the first axis (0.85 and −0.80, respectively, Fig. 3). Five variables were selected as signif- icant predictors of species distribution by the forward selection analyses: rock, depth, mud, gravel and sand (Table 2). Two groups of species can be delineated on the CCA diagram: species situated on the left-hand side of the diagram with an occurrence restricted to muddy sedi- ments (Y. solidula, Y. lenticula, Thyasira dunbari, Nuculana pernula, Ciliatocardium ciliatum, Frigidoalvania cruenta, Ennucula tenuis, Arctinula groenlandica and Y. hyperborea); species situated on right-hand side of the plot, with an Figure 2 Species accumulation curves plotted for observed number of species (Sobs) and the true number of species estimated using Chao2 and the Michaelis-Menten (MM) equation. Sobs and Chao2 are plotted with 0.95 confidence intervals.Samples Sobs Chao2 MM N u m b e r o f sp e ci e s 140 120 100 80 60 40 20 20 40 60 80 100 120 140 160 M. Wlodarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 53 affinity to occurrence at the hard bottom (Erginus rubellus, Margarites groenlandicus, Tonicella rubra, T. marmorea, Bucci- num glaciale, Puncturella noachina, Oenopota pyramidalis, Mysella sovaliki and Moelleria costulata). Margarites helicinus was placed close to the kelp variable centroid. Bathyarca glacialis and Frigidoalvania janmayeni were located close to the drop stones centroid. In the RDA of grab data the first two ordination axes were significant (P = 0.002) and explained about 89% of the variance of the species–environment relationship. A third axis was not significant (P > 0.05). The concentra- tion of silt (0.68) and the longitude (0.62) had the highest positive correlation to the first axis (Fig. 4a). The salinity at the bottom was negatively correlated to the first axis (−0.63). The temperature at the bottom was negatively correlated to the second axis (−0.63). Forward selection, applied to the full set of environmental variables consid- ered, discriminated 10 variables with significant condi- tional effects (Table 2). The small conditional effect of longitude, temperature, salinity and sand compared with their large marginal effect results from the high correla- tion of these variables to silt concentration. The extra fit of the variables mentioned above is very small, as a large part of the effect is already explained by a strongly co- varying variable: silt. Five environmental variables were selected in repeated forward selection analyses by pro- gressive elimination of the variables with non-significant conditional effects: silt, temperature, mineral and organic suspensions and POC/PON in sediments. Three groups of species can be distinguished on the RDA diagram: (1) species that correlated positively with silt, mineral sus- pensions and longitude (Y. solidula, Y. lenticula and T. dun- bari); (2) species that correlated negatively with longitude (distance along the fjord) and its co-variables (Lepeta caeca, Margarites costalis and Ischnochiton albus); (3) species not correlated to environmental variables considered in the analyses (Mya truncata, Y. hyperborea and T. gouldi). Molluscan associations Four groups of stations were distinguished on the den- drogram resulting from clustering the presence/absence data from samples taken by qualitative methods (Fig. 5): (1) ROCK, samples collected at shallow stations (5–20 m) situated on rocky shelves at two Kongsfjorden capes Figure 3 Ordination diagram based on the canonical correspondence analyses of mollusc species occurrences with respect to environmental variables in Kongsforden. Circles represent centroids of dummy variables, arrows are linear variables and triangles are species. Only species with a 10% minimum fit to the lower axis are plotted. The species names are coded by the first three letters of the generic and specific name (for species names consult the Appendix). 1 .0 1.0-0.6 -1 .0 Yol sol Yol lenFri cru Cil cil Nuc per Arc groEnn ten Yol hyp Fri jan Bat gla Cre dec Moe cos Mys sov Oen pyr Pun noa Buc gla Erg rub Mus dis Mar gro Hia arc Bor cas Mar hel Ton rub Ton mar Thy dun rock kelp drop- depth gravel mud longitude sand stones Table 2 Marginal and conditional effects obtained from the forward selection of environmental variables. λ1 is a fit with an eigen value with one variable only, λA is an additional fit with an increase in the eigen value, P is the significance level obtained with a Monte Carlo simulation using 499 unrestricted permutations. Variable Marginal effects Conditional effects Pλ1 λA Qualitative data Rock 0.52 0.52 0.002 Depth 0.31 0.26 0.012 Mud 0.49 0.21 0.034 Gravel 0.25 0.21 0.022 Sand 0.25 0.2 0.034 Kelp 0.14 0.14 0.102 Longitude 0.31 0.13 0.08 Drop stones 0.23 0.08 0.574 Grab data Silt 0.29 0.29 0.002 Temperature 0.28 0.08 0.004 Mineral suspensions 0.13 0.13 0.002 Depth 0.16 0.05 0.004 POC/PON 0.02 0.03 0.018 Organic suspensions 0.14 0.04 0.004 Drop stones 0.03 0.02 0.066 Salinity 0.22 0.03 0.018 Gravel 0.10 0.03 0.022 POC 0.19 0.05 0.002 Longitude 0.26 0.03 0.008 Sand 0.26 0.02 0.002 Mean grain size 0.10 0.01 0.04 POC, particulate organic carbon concentration in sediment; PON, particulate organic concentration in sediment. 54 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wlodarska-Kowalczuk (Kvadehuken and Kapp Guissez); (2) KELP, samples collected in the shallow sublittoral on different types of bottom covered with kelp (L. saccharina, L. digitata and A. esculenta); (3) SEDIMENT, samples collected on soft bottom at depths ranging from 5 to 300 m; and (4) GRAVEL, samples collected on gravel/stones at the entrance of the fjord (depths below 150 m). There were six species characteristic for the ROCK asso- ciation: the chitons T. rubra and T. marmorea; the gastro- pods M. costulata, P. noachina and M. groenlandicus; and the bivalves Crenella decussata and M. sovaliki (Table 3). Only the gastropod M. helicinus could be classified as character- istic for the group of samples collected at sites with kelp. The gastropods Lepeta caeca, Margarites costalis and Trophon clathratus and the chiton Ischnochiton albus were character- istic for the GRAVEL association. Thirteen species could be identified as characteristic for the SEDIMENT association (Table 3). Five groups of samples could be distinguished on the dendrogram of Bray-Curtis similarities of square-root transformed quantitative data (Fig. 6): (1) G1, samples collected in the Kongsbreen glacial bay; (2) G2, samples collected in the Blomstrandbreen glacial bay; (3) CB, sam- ples collected in the central basin; (4) CD, a small number of stations in the central basin where drop stones were common; (5) CE, samples from three stations situated at the mouth of the fjord. The molluscs of the Kongsbreen glacial bay (G1) were dominated by the bivalves Y. solidula, Y. lenticula and T. dunbari (Table 4). Both species of Yoldiella were also numerous in the glacial bay of the Blomstranbreen (G2), but the characteristic species of this association included only Y. hyperborea, A. orbiculata and Thyasira gouldi. All the species already mentioned were also present, although much less numerous, in the CB association, where Chaeto- derma nitidulum was the characteristic species. B. glacialis Figure 4 Ordination diagram based on redundancy analyses of square root transformed mollusc species densities in grab samples with respect to environmental variables. Circles represent centroids of dummy vari- ables, small arrowheads indicate species and large arrowheads indicate linear variables. Only species with a 5% minimum fit to the lower axis are plotted. The species names are coded by the first three letters of the generic and specific name (for species names consult the Appendix). POC, particulate organic carbon concentration in sediments; POC/PON, partic- ulate organic carbon/particulate organic nitrogen ratio; Mz, mean grain size in sediments. -1.0 1.0 -1 .0 1. 0 Axi orb Yol hyp Mya tru Thy gou Nuc per Yol sol Yol len Thy dun Arc gro Cus subCyl occ Yol luc longitude silt Mz mineral suspensions organic suspensions POC/PON clay salinity gravel POC depthsand temperature drop-stones Mac cal Fri cru Isc alb Mar cos Lep cae Figure 5 Distribution of associations distin- guished on a dendrogram, plotted using the group average linking of Bray-Curtis similarities of presence/absence data of species occur- rences in samples taken using qualitative meth- ods (dredges and scuba-diving). The simplified dendrogram is presented. sediment kelp rock gravel Bray-Curtis similarity (%) 40 80 0 kelp rock gravel sediment M. Wlodarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 55 Table 3 The most common species in associations distinguished in multivariate analyses of qualitative data. F is frequency (%), NI is number of all individuals. Species with frequency exceeding 25% are presented. Characteristic species [degree of association regarding stations (DAS) exceeding 50%] are in boldface. ROCK KELP GRAVEL SEDIMENT NI F NI F NI F NI F Tonicella rubra 113 78 1 5 1 17 1 3 Crenella decussata 92 33 2 3 Moelleria costulata 48 33 1 3 Mysella sovaliki 51 33 Puncturella noachina 8 33 Margarites groenlandicus 14 78 12 25 32 3 Astarte borealis 34 33 35 15 Cingula castanea 11 33 3 5 3 5 Erginus rubellus 90 89 19 35 2 3 Hiatella arctica 45 89 18 35 29 67 44 28 Tonicella marmorea 39 56 5 5 48 33 Margarites helicinus 11 22 3316 95 15 10 Lepeta caeca 16 83 Astarte montagui 6 50 69 20 Margarites costalis 2 33 Ischnochiton albus 13 22 13 17 Trophon clathratus 2 33 1 3 Yoldiella solidula 1 17 2479 68 Yoldiella lenticula 7333 73 Nuculana pernula 1837 78 Arctinula groenlandica 869 63 Mya truncata 34 33 7 20 351 60 Ennucula tenuis 1 11 1 5 1448 55 Ciliatocardium ciliatum 219 50 Thyasira dunbari 418 43 Frigidoalvania cruenta 6 17 344 40 Axinopsida orbiculata 15 22 1129 38 Bathyarca glacialis 5 17 304 38 Macoma calcarea 2 11 12 17 302 33 Cuspidaria subtorta 381 28 Figure 6 Distribution of associations distin- guished on a dendrogram, plotted using the group average linking of Bray-Curtis similarities of square root transformed densities of species in quantitative samples from the soft bottom. The simplified dendrogram is presented. glacial bays associations central basin associations G1 G2 CD CB CE B ra y- C u rt is s im ila ri ty ( % ) 40 80 0 G1 G2 CDCB CE 56 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wlodarska-Kowalczuk was both the most abundant and characteristic species in the CD association, whereas the gastropod L. caeca was characteristic for the CE association. E. tenuis, N. pernula and Macoma calcarea were not characteristic for any of the associations—they were common in soft sediments all over the fjord. Density and diversity in the quantitative samples from the soft bottom The number of individuals (N) was significantly higher in the glacial bays (214 ind./0.1 m2 on average in G1, 387 ind./0.1 m2 in G2) than in the central basin associa- tions (32 ind./0.1 m2 in CB, 40 ind./0.1 m2 in CD, 27 ind./ 0.1 m2 in CE; Fig. 7). SP was at the highest level and significantly different from most other assemblages in G2 (Fig. 7; Table 5). H was higher in G2, CB and CD than in G1 and CE. J was lower in glacial bays (G1 and G2) than in the central basin assemblages, CB, CD and CE. avTD was significantly lower in glacial bays (G1 and G2) than in the central basin assemblages (CB, CD and CE; Fig. 7; Table 5). There was no significant difference in varTD at P < 0.05 (one-way ANOVA). Discussion The survey of Kongsfjorden molluscs yielded 87 mollusc species. Rozycki (1991) reported 39 mollusc species in 18 samples collected in Kongsfjorden in 1988. His list included nine species (Littorina saxatilis, Nucella lapillus, Table 4 The dominant species in mollusc associations at the soft bottom. D is dominance (%), avN is average density (ind./0.1 m2). The five most abundant species are presented for each association. Characteristic species [either degree of association regarding individuals (DAI) or DAS > 60%] are in boldface. G1 G2 CB CD CE avN D avN D av N D avN D avN D Yoldiella solidula 116.0 54 24.2 6 5.2 16 0.3 1 0.3 1 Yoldiella lenticula 53.0 25 9.0 2 7.7 24 0.5 2 0.1 1 Thyasira dunbari 24.7 12 1.0 0 0.2 1 0.3 1 0.1 1 Ennucula tenuis 8.7 4 29.0 8 2.3 7 1.1 4 Nuculana pernula 3.6 2 15.8 4 2.7 8 0.9 3 Axinopsida orbiculata 1.0 1 177.2 46 0.6 2 0.5 2 Yoldia hyperborea 0.1 0 48.3 13 0.2 1 Thyasira gouldi 48.2 12 0.1 0 2.8 10 Macoma calcarea 0.1 0 8.7 2 0.2 1 8.3 30 Frigidoalvania cruenta 0.5 0 6.3 2 2.0 6 Chaetoderma nitidulum 0.9 0 0.7 0 3.7 11 1.8 6 1.1 4 Bathyarca glacialis 0.1 0 2.3 7 23.3 78 Frigidoalvania janmayeni 1.4 4 2.0 7 Oenopota sp. 0.0 0 0.4 1 1.5 5 Lepeta caeca 0.7 2 8.0 29 Figure 7 Density (N, number of individuals per 0.1 m2) and diversity (SP, number of species per sample; H, Shannon-Wiener index; J, Pielou index; avTD, average taxonomic distinctness; varTD, variance of taxonomic distinctness) in mollusc assemblages of the soft bottom (mean and 95% confidence intervals). S P N ( in d ./ 0 .1 m 2 ) 100 300 500 6 10 14 18 1.2 1.6 2.0 H J 0.6 0.7 0.8 0.9 a vT D va rT D 60 64 68 72 76 160 240 320 G1 G1G2 G2CB CBCD CDCE CE glacial bays glacial bayscentral basin central basin associations associations M. Wlodarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 57 Buccinum hydrophanum, Colus latericus, Turbonilla inter- rupta, Thyasira ferruginea, Macoma loveni, Kellyella miliaris and Poromya subtorta) that were absent in material I have studied here. Combining the data, there have been 96 species recorded in Kongsfjorden to date. This number of species lies within the confidence interval of the Chao2 estimate of the true species richness in the fjord, and is a quarter of all the Arctic species (436 species of molluscs, excluding Nudibranchia and Cephalopoda) listed by Sirenko (2001). It is difficult to compare the assessment of Konsfjorden molluscan species richness to other Arctic localities as there are few studies of comparative sampling effort covering the full range of fjordic habitats and depths. Seventy nine mollusc species were identified in 106 benthic samples collected with the use of grabs and dredges at depths from 2 to 190 m in Jørgen Brønlund Fjord, North Greenland (Schiøtte 1989). Only 26 mollusc species were found in 27 dredge samples in Expedition Fiord, Northwest Territories, Canada (Aitken & Gilbert 1996). Odhner (1915) recorded 130 mollusc species (cephalopods and nudibranchs excluded) in a compre- hensive study of Isfjorden (west Spitsbergen) based on 130 stations sampled with various dredges at depths rang- ing from 2 to 406 m. Rozycki (1993) added another nine bivalve species to Odhner’s list. Isfjorden is about four times longer and much larger in area than Kongsfjorden, so it can host a higher number of species, but its species richness is nevertheless dramatically lower than that found in lower latitude sites of a similar area. A species inventory of a tropical coastal site with an area of 295 km2 off the coast of New Caledonia recorded as many as 2738 species of marine molluscs (Bouchet et al. 2002). The differences between the mollusc species richness in polar and tropical sites are enormous. The huge latitudi- nal cline of diversity, more pronounced in molluscs than in some other taxa, may be related to the increased cost of calcification in low temperatures, as suggested by Clarke (1992). The majority of mollusc species in Kongsfjorden showed strong preferences towards a single substratum type; only a few (e.g. Hiatella arctica and Mya truncata) were abundant both on soft and hard substrata. The major discontinuity in mollusc species distributions sepa- rated the biota of rocky shelves, gravel beds and the soft bottom. Multivariate analyses distinguished the kelp associated fauna as the fourth association, but kelp bed samples differed from the rocky shelf material mostly in the common occurrence and dominance of M. helicinus, a well-known dweller of kelp cauloids and phylloids in the shallow waters of Spitsbergen (Rozycki & Gruszczynski 1986). The substrate dictates the taxonomic and func- tional organization of mollusc associations. The rocky shelf fauna is dominated by grazing chitons (Tonicella) and gastropods (Erginus and Margarites), and suspension feeding sedentary bivalves (Crenella, Astarte and Hiatella), the gravel beds are inhabited by grazers, surface deposit feeders and suspension feeders (Ischnochiton, Lepeta and Astarte), whereas mobile deposit feeding bivalves (Yoldi- ella, Yoldia, Nuculana, Ennucula and Axinopxida) dominate in soft sediments. Such distribution of functional groups agrees well with the model patterns described for Boreal coastal macrobenthic communities by Pearson & Rosenberg (1987). Five separate assemblages distinguished in soft sedi- ments could be regarded as different expressions of the same species pool, as any differences were largely a result of the varying patterns of dominance of the same set of species. The changes in the relative dominance of the infaunal molluscs could be largely explained in terms of the distance that any particular assemblage lies from the head of the fjord and from the most active glacier, Kongs- breen. Although RDA analysis showed that there are many factors that were related to the distribution of fau- nal assemblages, the significant factors were co-variable and the overall influence of the glacier was predominant. The granulometric characteristics, the quantity of organic carbon and the stability of sediments all reflect the glacial sedimentation gradients in an Arctic glacial fjord (Gorlich et al. 1987; Syvitski et al. 1987; Wlodarska-Kowalczuk & Pearson 2004). Similarly the salinity and temperature gradients are produced by the inflow of fresh and cold glacial meltwaters, and are controlled by the glacier activity (Svendsen et al. 2002). High levels of mineral sedimentation and sediment deposition have been shown to be an acute disturbance agent causing a dramatic decrease in benthic densities and diversity (e.g. Airoldi 2003; Anderson et al. 2004; Thrush et al. 2004; Wlodarska-Kowalczuk et al. 2005). Arctic molluscs cope surprisingly well with the glacial Table 5 Results of Kruskal-Wallis test comparing density (N, number of individuals per 0.1 m2) and ANOVA comparing diversity measures (SP is the number of species per sample, H is Shannon-Wiener index, J is Pielou index, avTD is average taxonomic distinctness, varTD is variance of taxo- nomic distinctness) of mollusc associations at the soft bottom. Pairwise contrasts determined by Mann-Whitney U tests for N and post-hoc Fisher’s LSD tests for SP, H, J and avTD. G1 and G2 are glacial bay associa- tions, CB, CE and CD are central basin associations. F or H P Significant contrasts in pairs of assemblages (at P < 0.05) N 47.5 (H) 0.000 G1, G2 > CB,CD,CE; G2 > G1 SP 13.0 (F ) 0.000 G2 > G1,CB,CD,CE; G1,CD > CE; CD > CB H 9.4 (F ) 0.000 G1 < CB,CD,G2; CB < CD; CD,G2 > CE J 13.9 (F ) 0.000 G1, G2 < CB,CD, CE avTD 17.2 (F ) 0.000 G1,G2 < CB,CD, CE varTD 1.1 (F ) 0.382 — 58 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wlodarska-Kowalczuk sedimentation disturbance. In Kongsfjorden, the mollusc numbers are ten times higher in glacial bays than in the stable sediments of the central basin because of the high abundance of a few species of Yoldiella and Thyasira. Pro- tobranch bivalves are the most common dominants in the sediments influenced by either glacial or glaciofluvial inputs in several other Arctic and sub-Arctic sites, whereas suspension feeding bivalves Astarte, Hiatella and Bathyarca seem to prefer the outer basins of fjords of high latitude (Table 6). Several Yoldiella species (Y. intermedia, Y. fraterna and Y. lenticula) were also reported as pioneer species populating defaunated sediments in succession following the onset of deglaciacion in late Pleistocene shelf sediments off the coast of northern Norway (Thom- sen & Vorren 1986). This pioneer community was later replaced by an “established low-Arctic community” com- prising B. glacialis, Astarte crenata, Frigidoalavania janmay- eni, T. gouldi and A. groenlandica (Thomsen & Vorren 1986). The common features of yoldiid and thyasirid bivalves successful in glacial bays include small, simple, smooth and thin shells with no external ornamentation, which facilitate movement and the maintenance of the proper position in unstable quickly accumulating sedi- ments (Rhoads 1974). Yoldiid bivalves possess a large and muscular foot and are known to move into and through a soft substratum (Yonge 1939). They collect detritus particles from the sediment surface with the use of long palp proboscides, and can very efficiently sort the parti- cles (Stasek 1965). The efficient mechanism of elimina- tion of pseudofaeces protects the respiratory organs of protobranch bivalves from being clogged by mineral par- ticles accumulating in the mantle cavity (Rhoads 1974). All of these traits of yoldiid bivalves (small size, high mobility, selective deposit feeding and the efficient elimi- nation of mineral particles) obviously facilitate their sur- vival in glacial bays. What makes thyasirids well-suited to high sedimentation and instable sediments is less clear. Some Thyasira species construct a complicated system of mucus-lined tunnels, host symbiotic chemoautotrophic bacteria and are nutritionally dependent on hydrogen sulphides in sediments (Dando & Southward 1986; Dando & Spiro 1993). The Spitsbergen fjordic sediments are well oxygenated (Jorgensen et al. 2005), and the local thyasirids must depend on energy sources other than energy derived from the oxidation of sulphur com- pounds. The high sedimentation does not favour the construction of stable tunnels. The life habits of the Kongsfjorden thyasirid dominant species must differ from those of the temperate Atlantic species described by Dando & Southward (1986) and Dando & Spiro (1993). The dominants in two Kongsfjorden glacial bays are different, although closely related: Y. solidula and T. dun- bari in Kongsbreen glacial bay (G1) and Y. hyperborea, A. orbiculata and T. gouldi in Blomstrandbreen glacial bay (G2). The Blomstrandbreen glacial bay is located closer to the fjord mouth and has a much higher concentration of organic matter in the sediments than the sediments of Kongsbreen glacial bay (Wlodarska-Kowalczuk & Pear- son 2004). Wlodarska-Kowalczuk et al. (1998) reported the similar separation of protobrach bivalves in a survey of six west Spitsbergen glacial bays—Y. hyperborea domi- nation in two locations situated at either the open coast or close to the fjord mouth and Yoldiella solidula (errone- ously identified as Yoldiella fraterna) in four sites situated in inner fjord basins. Ockelmann (1958) suggested that Y. hyperborea has fairly high energetic demands and there- fore occurs primarily at either the open coast or in the outer parts of the fjords, where the supply of organic matter is larger than in the inner fjord basins. Domination by Y. hyperborea was reported for open shelf waters influ- Table 6 Dominant mollusc taxa in subtidal sediments in inner (either glacial or glaciofluvial sedimentation influenced) basins versus outer parts of Arctic fjords. Location Inner fjord Outer fjord Expedition Fiord, Northwestern Territories, Canada (Aitken & Gilbert 1996) Portlandia arctica, Thyasira gouldi Astarte borealis, Astarte warhami, Hiatella arctica, Mya sp., Trichotropis sp. McBeth and Itirbilung fiords (Syvitski et al. 1989, Aitken& Fournier 1993) Portlandia arctica, Hiatella arctica, Axinopsida orbiculata Axinopsida orbiculata, Bathyarca glacialis, Astarte spp. Cambridge Fiord (Aitken & Fournier 1993) Yoldiella intermedia, Yoldiella lenticula, Axinopsida orbiculata Axinopsida orbiculata, Bathyarca glacialis, Astarte spp. Yoldiabukta, Isfjorden, Spitsbergen (Wlodarska- Kowalczuk et al. 1999) Yoldiella fraterna — Hornsund, unpublished data Portlandia arctica Ennucula tenuis, Ciliatocardium ciliatum van Mijenfjorden, Svalbard (Gulliksen et al. 1985, Renaud et al. 2007) Portlandia arctica, Yoldiella solidula Macoma calcarea Jørgen Brønlund Fjord, North Greenland (Schiøtte 1989) Portlandia arctica Hiatella arctica Scoresby Sund, Greenland (Thorson 1934) Portlandia arctica Astarte crenata, Bathyarca glacialis Young Sound, East Greenland (Sejr et al. 2000) — Astarte spp., Hiatella arctica M. Wlodarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 59 enced by the terrigenous inflows off Svalbard (Wlodarska et al. 1996), as well as off southern Greenland and Ice- land (Peres 1982). Similarly, T. gouldi was reported to occur in large numbers in Chukchi Sea shelf sediments (Feder et al. 1994), and may be less suited to survival in inner basin glacial bay sediments that are poor in organic matter. The yoldiid and thyasirid bivalves occurred at most stations with soft bottoms in Kongsfjorden, but it was only in glacial bays that these species were found in large numbers. In central basins, in stable conditions with low inorganic sedimentation, the bivalves are replaced by numerous tube-building polychaetes (Maldane sarsi and Spiochaetopterus typicus; Wlodarska-Kowalczuk & Pearson 2004). The same pattern is observed in Canadian fjords: Portlandia arctica is present in large numbers only near glaciers and is much less numerous in central and outer fjord basins inhabited by Maldanid association (Syvitski et al. 1989; Aitken & Fournier 1993). The dense popula- tions of tube-building polychaetes can diminish the num- bers of mobile burrowers as a result of the competition for space and food (Wilson 1991). The experimental removal of tube builders results in the dramatic increase in num- bers of mobile detritus feeders (Woodin 1974). The high densities of protobranch and thyasirid bivalves in glacial bays may result from the competitive release, as the tube builders are naturally eliminated by high sedimentation and the deposition of mineral material. The weak com- petitive capabilities of Protobranchia were expressed in the evolutionary history of the group. Protobranchia are numerically dominant and are the most diverse deep sea bivalves known at present (Allen & Sanders 1996). Dur- ing the Mesozoic the Atlantic Protobranchia migrated to the deep sea as a consequence of competitive exclusion by the eulamellibranch bivalves radiating at that time in shallow waters (Allen 1978). The patterns of mollusc distribution at the soft bottom and diversity in Kongsfjorden are not fully consistent with those of the whole macrobenthic community described by Wlodarska-Kowalczuk et al. (2005). The main discontinuity in patterns of both complete macrob- enthos and mollusc species distribution separates the glacial bays and central basin stable sediments, but the dramatic species diversity cline towards the glaciers described for the whole macrofauna was not observed in mollusc assemblages. The high increase of numbers of individuals in assemblages near glaciers was also a pattern unique to molluscs. The taxonomic distinctness showed the contrasting patterns along the glacial sedimentation gradient. avTD drops for molluscs, but increases for the whole macrobenthic community the closer one gets to the glacier. It is noteworthy that the discrepancies in taxonomic distinctness patterns for different taxonomic groups may result from different responses to environ- mental gradients, but also from differences in the hierar- chical taxonomic systems of different phyla (Ellingsen et al. 2005). Smith (2005) observed the high correlation of mollusc diversity and the overall benthic community diversity on rocky shores, and recommended using mol- luscs as surrogates for complete macrobenthic communi- ties in benthic diversity and impact assessment studies. However, the molluscs of the soft bottom seem to react to glacial disturbance in a different way than do other benthic phyla and I would not recommend molluscs as a useful surrogate for the whole macrobenthic community in surveys of sedimentary habitats. Acknowledgements The samples used in this study came from several sam- pling campaigns. I would like to thank Dr J. Lege y ska, Dr M. Zaj czkowski and all colleagues involved in the field work. Dr J. Lege y ska and Dr P. Kukli ski provided samples from baited traps and diving surveys, which is gratefully acknowledged. I wish to thank Dr A. Warèn for the verification of mollusc identifications and for valuable comments. My visits to the Swedish Natural History Museum in Stockholm and the Zoological Museum in Copenhagen were supported by the HIGHLAT project (HPRI-CT-2001-0125) and the COBICE project (HPRI- CT-1999-00021). I wish to thank Dr K. Blachowiak- Samolyk for help with statistical analyses. I am very grateful to Dr M.A. Kendall, Prof. J.M. W slawski and the two anonymous referees whose comments much improved the manuscript. The study has been completed thanks to funds provided by grant PBZ-KBN-108/P04/ 2004 from the Polish National Committee for Scientific Research. The author acknowledges the support by the MarBEF Network of Excellence project Marine Biodiver- sity and Ecosystem Functioning, which is funded by the Sustainable Development, Global Change and Ecosys- tems Programme of the European Community’s Sixth Framework Programme (contract no. GOCE-CT-2003- 505446). This publication is contribution number MPS- 06039 of MarBEF and a contribution to the MarBEF responsive mode programme ArctEco. References Airoldi L. 2003. The effects of sedimentation on rocky coast assemblages. Oceanography and Marine Biology Annual Review 41, 161–236. Aitken A.E. 1990. Fossilization potential of Arctic fjord and continental shelf benthic macrofaunas. In J.A. Dowdeswell & J.D. Scourse (eds.): Glacimarine environments: processes and sediments. Geological Society of London Special Publication No 53. Pp. 155–176. London: Geological Society. ż n¢ a̧ ż n¢ n¢ ȩ 60 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wlodarska-Kowalczuk Aitken A.E. & Fournier J. 1993. Macrobenthos communities of Cambridge, McBeth and Itirbilung fiords, Baffin Island, Northwest Territories, Canada. Arctic 46, 60–71. Aitken A.E. & Gilbert R. 1996. Marine Mollusca from Expedi- tion Fiord, Western Axel Heiberg Island, Northwest Territo- ries. Arctic 49, 29–43. Allen J.A. 1978. Evolution of the deep sea protobranch bivalves. Philosophical Transactions of the Royal Society of London 284, 387–401. Allen J.A. & Sanders H.L. 1996. The zoogeography, diversity and origin of the deep-sea protobranch bivalves of the Atlantic: the epilogue. Progress in Oceanography 38, 95–153. Anderson M.J., Diebel C.E., Blom W.M. & Landers T.J. 2005. Consistency and variation in kelp holdfast assemblages: spa- tial patterns of biodiversity for the major phyla at different taxonomic resolutions. Journal of Experimental Marine Biology and Ecology 320, 35–56. Anderson M.J., Ford R.B., Feary D.A. & Honeywill C. 2004. Quantitative measures of sedimentation in an estuarine sys- tem, and its relationship with intertidal soft-sediment infauna. Marine Ecology Progress Series 272, 33–48. Bick A. & Arlt G. 2005. Intertidal and subtidal soft-bottom macro-and meiofauna of the Kongsfjord (Spitsbergen). Polar Biology 28, 550–557. Bouchet P., Louzet P., Maestrati P. & Heros V. 2002. Assessing the magnitude of species richness in tropical marine envi- ronments: exceptionally high numbers of molluscs at a New Caledonia site. Biological Journal of the Linnean Society 75, 421– 436. Chao A. 1987. Estimating the population size for capture- recapture data with unequal catchability. Biometrics 43, 783– 791. Chao A. 2004. Species richness estimation. In N. Balakrishnan et al. (eds.): Encyclopedia of statistical sciences. Pp. 1–23. New York: Wiley. Clarke A. 1992. Is there a latitudinal diversity cline in the sea? Trends in Ecology and Evolution 7, 286–287. Clarke K.R. & Warwick R.M. 2001. A further biodiversity index applicable to species lists: variation in taxonomic distinctness. Marine Ecology Progress Series 216, 265–278. Colwell R.K. 2005. Estimates: Statistical estimation of species of richness and shared species from samples. Version 7.5 Users guide: http://purl.oclc.org/estimates. Colwell R.K., Mao C.X. & Chang J. 2004. Interpolating, extrap- olating, and comparing incidence-based species accumula- tion curves. Ecology 85, 2717–2727. Dale J.E., Aitken A.E., Gilbert R. & Risk M.J. 1989. Macro- fauna of Canadian Arctic fjords. Marine Geology 85, 331–358. Dando P.R. & Southward A.J. 1986. Chemoautotrophy in bivalve molluscs of the genus Thyasira. Journal of the Marine Biological Association of the United Kingdom 66, 915–929. Dando P.R. & Spiro B. 1993. Varying nutritional dependence of the thyasirid bivalves Thyasira sarsi and T. equalis on chemoautotrophic symbiotic bacteria, demonstrated by iso- tope ratios of tissue carbon and shell carbonate. Marine Ecol- ogy Progress Series 92, 151–158. Dowdeswell J.A. & Forsberg C.F. 1992. The size and frequency of icebergs and bergy bits derived from tidewater glaciers in Kongsfjorden. northwest spitsberger. Polar Res 11, 81–91. Ellingsen K.E., Clarke K.R., Somerfield P.J. & Warwick R.M. 2005. Taxonomic distinctness as a measure of diversity applied over a large scale: the benthos of the Norwegian continental shelf. Journal of Animal Ecology 74, 1069–1079. Feder H.M., Foster N.R., Jewett S.C., Weingartner T.J. & Baxter R. 1994. Mollusks in the northeastern Chukchi Sea. Arctic 47, 145–163. Fetzer I., Lonne O.J. & Pearson T. 2002. The distribution of juvenile benthic invertebrates in an Arctic glacial fjord. Polar Biology 25, 303–315. Gordillo S. & Aitken A.E. 2000. Palaeoenvironmental interpre- tation of late quaternary marine molluscan assemblages, Canadian Arctic archipelago. Geographie Physique et Quater- naire 54, 301–315. Gorlich K., Weslawski J.M. & Zajaczkowski M. 1987. Suspen- sion settling effect on macrobenthos biomass distribution in the Hornsund fjord, Spitsbergen. Polar Research 5, 175–192. Gulliksen B., Holte B. & Jakola K.J. 1985. The soft-bottom fauna in van Mijenfjord and Raudfjord, Svalbard. In J. Gray & M.E. Christiansen (eds.): Marine biology of polar regions and effects of stress on marine organisms. Pp. 199–215. Oslo: John Wiley and Sons. Hop H., Pearson T.H., Hegseth E.N., Kovacs K.M., Wiencke C., Kwasniewski S., Eiane K., Mehlum F., Gulliksen B., Wlodarska-Kowalczuk M., Lydersen C., Weslawski J.M., Cochrane S., Gabrielsen G.W., Leakey R., Lonne O.J., Zajaczkowski M., Falk-Petersen S., Kendall M.A., Wängberg S.-Å., Bischof K., Voronkov A., Kovaltchouk N.A., Wiktor J., Poltermann M., di Prisco G., Papucci C. & Gerland S. 2002. The marine ecosystem of Kongsfjorden, Svalbard. Polar Research 21, 167–208. Jorgensen B.B., Glud R.N. & Holby O. 2005. Oxygen distribu- tion and bioirrigation in Arctic fjord sediments (Svalbard, Barents Sea). Marine Ecology–Progress Series 292, 85–95. Jorgensen L.L. & Gulliksen B. 2001. Rocky bottom fauna in Arctic Kongsfjord (Svalbard) studied by means of suction sampling and photography. Polar Biology 24, 113–121. Kaczmarek H., Wlodarska-Kowalczuk, M., Legezynska J. & Zajaczkowski M. 2005. Shallow sublittoral macrozoobenthos in Kongsfjord, West Spitsbergen, Svalbard. Polish Polar Research 26, 137–155. Kendall M.A., Widdicombe S. & Weslawski J.M. 2003. A multi- scale study of the biodiversity of the benthic infauna of the high-latitude Kongsfjord, Svalbard. Polar Biology 26, 383– 388. Kuklinski P. & Porter J.S. 2004. Alcyonidium disciforme: an exceptional Arctic bryozoan. Journal of the Marine Biological Association of the United Kingdom 84, 267–275. Lefauconnier B., Hagen J.O. & Rudant J.P. 1994. Flow speed and calving rate of Kongsbreen glacier, Svalbard, using SPOT images. Polar Research 13, 59–65. Legezynska J. 2001. Distribution patterns and feeding strate- gies of lyssianassid amphipods in shallow waters of an Arctic fjord. Polish Polar Research 22, 173–186. http://purl.oclc.org/estimates M. Wlodarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 61 Legezynska J. 2002. Ekologia padlinozernych bezkregowcow w Arktycznym fiordzie. (Ecology of scavenging invertebrates in an Arctic fjord.) PhD Thesis, University of Gdansk, Poland. Lippert H., Iken K., Rachor E. & Wiencke C. 2001. Macrofauna associated with macroalgae in the Kongsfjord (Spitsbergen). Polar Biology 24, 512–522. Magurran A.E. 2004. Measuring biological diversity. Malden, MA: Blackwell Publishing. Mieszkowska N., Kendall M.A., Hawkins S.J., Leaper R., Williamson P., Hardman-Mountford N.J. & Southward A.J. 2006. Changes in the range of some common rocky shore species in Britain—a response to climate change? Hydrobiolo- gia 555, 241–251. Ockelmann K.W. 1958. Marine Lamellibranchiata. Meddelelser om Gronland 122. Odhner N.H. 1915. Die Molluskenfauna des Eisfjordes. Kungliga Svenska Vetenskapsakasemiens Handligar 54. Pearson T.H. & Rosenberg R. 1987. Feast and famine: structur- ing factors in marine benthic communities. In J.H.R. Gee & P.S. Giller (eds.): Organization of communities. Past and present. Pp. 373–95. Oxford, UK: Blackwell Science. Peres J.M. 1982. Major benthic assemblages. In O. Kinne (ed.): Marine ecology. Vol. V. Pp. 373–522. New York: John Wiley and Sons. Renaud P.E., Wlodarska-Kowalczuk, M., Trannum H., Holte B., Weslawski J.M., Cochrane S., Dahle S. & Gulliksen B. 2007. Multidecadal stability of benthic community structure in a High-Arctic glacial fjord (van Mijenfjord, Spitsbergen). Polar Biology 30, 295–305. Rhoads D.C. 1974. Organism-sediment relations on the muddy sea floor. Oceanography and Marine Biology Annual Review 12, 263–300. Rozycki O. 1991. Benthic molluscs of Kongsfjorden (Sval- bard). In J. Repielewska-Pekalowa & K. Pekala (eds.): Wyprawy geograficzne na Spitsbergen. (Geographic expeditions to Spitsbergen.) Pp. 289–295. Lublin, Poland: UMCS. Rozycki O. 1993. Shallow-water molluscs of Isfjorden (west Spitsbergen, Svalbard). Polish Polar Research 14, 55–63. Rozycki O. & Gruszczynski M. 1986. Macrofauna associated with laminarians in the coastal waters of west Spitsbergen. Polish Polar Research 7, 337–351. Salzwedel H., Rachor E. & Gerdes D. 1985. Benthic macro- fauna communities in the German Bight. Veroffentlichungen des Instuts fur Meeresforschungin Bremerhaven 20, 199–267. Schiøtte T. 1989. Marine Mollusca from Jorgen Bronlund Fjord, North Greenland, including the description of Diaphana vedelsbyae n. sp. Meddelelser om Gronland 28, 3–24. Sejr M.K., Jensen K.T. & Rysgaard S. 2000. Macrozoobenthic community structure in a High-Arctic East Greenland fjord. Polar Biology 23, 792–801. Sirenko B.I. 2001. List of species of free-living invertebrates of Eur- asian Arctic seas and adjacent deep waters. St Petersburg: Zoo- logical Institute. Smith S.D.A. 2005. Rapid assessment of invertebrate biodiver- sity on rocky shores: where there’s a whelk there’s a way. Biodiversity and Conservation 14, 3565–3576. Somerfield P.J., Cochrane S.J., Dahle S. & Pearson T.H. 2006. Free-living nematodes and macrobenthos in a high-latitude glacial fjord. Journal of Experimental Marine Biology and Ecology 330, 284–296. Stasek C.R. 1965. Feeding and particle-sorting in Yoldia ensifera (Bivalvia: Protobranchia) with notes on other nuculanids. Malacologia 2, 349–366. Stork N.E., Samways M.J. & Eeley H.A.C. 1996. Inventorying and monitoring biodiversity. Trends in Ecology and Evolution 11, 39–40. Svendsen H., Beszczynska-Moller A., Hagen J.O., Lefaucon- nier B., Tverberg V., Gerland S., Orbaek J.B., Bischof K., Papucci C., Zajaczkowski M., Azzolini R., Bruland O., Wiencke C., Winther J.G. & Dallmann W. 2002. The physical environment of Kongsfjorden–Krossfjorden, an Arctic fjord system in Svalbard. Polar Research 21, 133–166. Syvitski J.P.M., Burrell D.C. & Skei J.M. 1987. Fjords. Processes and products. New York: Springer-Verlag. Syvitski J.P.M., Farrow G.E., Atkinson R.J.A., Moore P.G. & Andrews J.T. 1989. Baffin Island fjord macrobenthos: bot- tom communities and environmental significance. Arctic 42, 232–247. ter Braak C.J.T. & Smilauer P. 2002. CANOCO reference manual and CanoDraw for Windows, user’s guide. Wageningen and Ceske Budejovice: Biometris. Terlizzi A., Scuderi D., Fraschetti S. & Anderson M.J. 2005. Quantifying effects of pollution on biodiversity: a case study of highly diverse molluscan assemblages in the Mediterra- nean. Marine Biology 148, 293–305. Thomsen E. & Vorren Y.O. 1986. Macrofaunal palaeoecology and stratigraphy in late quaternary shelf sediments off northern Norway. Palaeogeography, Palaeoclimatology, Palaeo- ecology 56, 103–150. Thorson G. 1934. Contributions to the animal ecology of the Scoresby Sound Fjord complex (East Greenland). Meddelelser om Gronland 100. Thrush S.F., Hewitt J.E., Cummings V.J., Ellis J.I., Hatton C., Lohrer A. & Norkko A. 2004. Muddy waters: elevating sedi- ment input to coastal and estuarine habitats. Frontiers in Ecology and Environment 2, 299–306. Warwick R.M., Emblow C., Feral J.P., Hummel H., van Ave- saath P. & Heip C. 2003. European Marine Biodiversity Research Sites. Report of the European Concerted Action: BIOMARE. Imple- mentation and Networking of large scale, long term marine biodi- versity research in Europe. Yerseke, Netherlands: Netherlands Institute for Ecological research, Centre for Estuarine and Marine Research Ecology. Wilson W.H. 1991. Competition and predation in marine soft-sediment communities. Annual Review of Ecology and Systematics 21, 221–241. Winkelman D. & Knies J. 2005. Recent distribution and accu- mulation of organic carbon on the continental margin west of Spitsbergen. Geochemistry Geophysics Geosystems 6, 1–22. Wlodarska-Kowalczuk M. & Pearson T.H. 2004. Soft-bottom macrobenthic faunal associations and factors affecting species distributions in an Arctic glacial fjord (Kongsfjord, Spitsbergen). Polar Biology 27, 155–167. 62 Polar Research 26 (2007) 48 – 63 © 2007 The Author Molluscs in Kongsfjorden (Spitsbergen, Svalbard) M. Wlodarska-Kowalczuk Wlodarska-Kowalczuk M., Pearson T.H. & Kendall M.A. 2005. Benthic response to chronic natural physical disturbance by glacial sedimentation in an Arctic fjord. Marine Ecology Progress Series 303, 31–41. Wlodarska-Kowalczuk M., Szymelfenig M. & Kotwicki L. 1999. Macro- and meiobenthic fauna of the Yoldiabukta glacial bay (Isfjorden, Spitsbergen). Polish Polar Research 20, 367–386. Wlodarska-Kowalczuk M., Weslawski J.M. & Kotwicki L. 1998. Spitsbergen glacial bays macrobenthos—a compara- tive study. Polar Biology 20, 66–73. Wlodarska M., Weslawski J.M. & Gromisz S. 1996. A compari- son of the macrofaunal community structure and diversity in two Arctic glacial bays—a “cold” one off Franz Josef Land and a “warm” one off Spitsbergen. Oceanologia 38, 251–283. Woodin S.A. 1974. Polychaete abundance patterns in a marine soft-sediment environment: the importance of biological interactions. Ecological Monographs 44, 171–187. Yonge C.M. 1939. The protobranchiate mollusca; a functional interpretation of their structure and evolution. Philosophical Transactions of the Royal Society of London 230, 79–147. Appendix List of mollusc species found in materials collected in Kongsfjorden in 1996–2000. CAUDOFOVEATA CHAETODERMATIDAE Chaetoderma nitidulum Lovén, 1844 POLYPLACOPHORA LEPTOCHITONIDAE Leptochiton arcticus (Sars, G.O. 1878) Leptochiton alvelous (Lovén, 1846) ISCHNOCHITONIDAE Tonicella marmorea (Fabricius, O. 1780) Tonicella rubra (Linné, 1767) Ischnochiton albus (Linné, 1767) GASTROPODA PROSOBRANCHIA FISSURELLIDAE Puncturella noachina (Linné, 1771) LOTHIDAE Erginus rubellus (Fabricius, O. 1780) LEPETIDAE Lepeta caeca (Müller, 1776) TROCHIDAE Margarites costalis (Gould, 1841) Margarites groenlandicus (Gmelin, 1791) Margarites helicinus (Phipps, 1774) Margarites olivaceus Brown, 1827 Solariella varicosa (Mighels & Adams, 1842) TURBINIDAE Moelleria costulata (Møller, 1842) RISSOIDAE Frigidoalvania janmayeni (Friele, 1878) Frigidoalvania cruenta (Odhner, 1915) Boreocingula castanea (Møller, 1842) ELACHISIMIDAE Elachisina globuloides (Warén, 1972) TURRITELLIDAE Tachyrhynchus reticulatus (Mighels & Adams, 1842) NATICIDAE Euspira pallida (Broderip & Sowerby, G.B. 1829) Cryptonatica affinis (Gmelin, 1791) MURICIDAE Trophon clathratus (Linné, 1767) Boreotrophon truncatus (Ström, 1767) BUCCINIDAE Buccinum undatum (Linné, 1758) Buccinum scalariforme, Møller, 1842 Buccinum glaciale Linné, 1761 Buccinum polare Gray J.E., 1839 Colus kroeyeri (Møller, 1842) Colus sabini (Gray, 1824) CANCELLARIDAE Admete viridula (Fabricius, O. 1780) CONIDAE Oenopota pyramidalis (Ström, 1788) Oenopota exarata (Møller, 1842) Oenopota impressa (Mørch, 1969) Oenopota nobilis (Møller, 1842) Oenopota sp GASTROPODA HETEROBRANCHIA MATHILDIDAE Turitellopsis stimpsoni (Stimpson, 1851) PYRAMIDELLIDAE Menestho albula (Fabricius, O. 1780) Menestho truncatula Odhner, 1915 GASTROPODA OPISTHOBRANCHIA DIAPHANIDAE Diaphana sp RETUSIDAE Retusa sp. PHILINIDAE Philine sp. CYLICHNIDAE Cylichna cf alba (Brown, 1827) Cylichna cf occulta (Mighels & Adams, 1842) BIVALVIA NUCULIDAE Ennucula tenuis (Montagu, 1808) NUCULANIDAE Nuculana pernula Müller, 1779 Nuculana minuta (Müller, 1776) YOLDIIDAE Yoldia hyperborea Torell, 1859 M. Wlodarska-Kowalczuk Molluscs in Kongsfjorden (Spitsbergen, Svalbard) Polar Research 26 (2007) 48 – 63 © 2007 The Author 63 Yoldiella solidula Warén, 1989 Yoldiella lucida (Lovén, 1846) Yoldiella lenticula (Møller, 1842) Yoldiella frigida (Torell, 1859) Yoldiella intermedia (Sars, M. 1865) Portlandia arctica (Gray, 1824) MYTILIDAE Crenella decussata (Montagu, 1808) Musculus corrugatus (Stimpson, 1851) Musculus discors (Linné, 1767) Dacrydium vitreum (Møller, 1842) ARCIDAE Bathyarca glacialis (Gray, J.E. 1824) PECTINIDAE Chlamys islandica (Müller, O.F. 1776) PROPEAMUSSIIDAE Arctinula groenlandica (Sowerby, G.B. 1842) THYASIRIDAE Axinopsida orbiculata (Sars, G.O. 1878) Thyasira gouldi (Philippi, 1845) Thyasira dunbari Lubinsky, 1976 Thyasira sp. n. (Bouchet & Warén, 1979) UNGULINIDAE Diplodonta torelli Jeffreys, 1847 KELLIDAE Kellia sp. MONTACUTIDAE Montacuta maltzani Verkrüzen 1876 Montacuta spitzbergensis Knipowitsch, 1901 Mysella sovaliki MacGinitie, 1959 ASTARTIDAE Astarte crenata (Gray, 1824) Astarte borealis (Schumacher, 1817) Astarte elliptica (Brown, 1827) Astarte montagui (Dillwyn, 1817) CARDIDAE Ciliatocardium ciliatum (Fabricius, O. 1780) Serripes groenlandicus (Bruguière, 1789) TELLINIDAE Macoma moesta Deshayes, 1855 Macoma calcarea (Gmelin, 1791) VENERIDAE Liocyma fluctuosa (Gould, 1841) MYIDAE Mya truncata Linné, 1758 HIATELLIDAE Hiatella arctica (Linné, 1767) THRACIIDAE Thracia myopsis Møller, 1842 Thracia devexa Sars, G.O. 1878 LYONSIIDAE Lyonsia arenosa (Møller, 1842) PANDORIDAE Pandora glacialis Leach, 1819 CUSPIDARIIDAE Cuspidaria subtorta (Sars, G.O. 1878) Cuspidaria arctica (Sars, M. 1859)