Molecular phylogeny of Rotaliida (Foraminifera) based on complete small subunit rdna sequences

Available online at www.sciencedirect.com Marine Micropaleontology 66 (2008) 233 246 www.elsevier.com/locate/marmicro Molecular phylogeny of Rotaliida (Foraminifera) based on complete small subunit rdna sequences Magali Schweizer a,b,, Jan Pawlowski c, Tanja J. Kouwenhoven a, Jackie Guiard c, Bert van der Zwaan a,d a Department of Earth Sciences, Utrecht University, The Netherlands b Geological Institute, ETH Zurich, Switzerland c Department of Zoology and Animal Biology, University of Geneva, Switzerland d Department of Biogeology, Radboud University Nijmegen, The Netherlands Received 18 May 2007; received in revised form 8 October 2007; accepted 9 October 2007 Abstract The traditional morphology-based classification of Rotaliida was recently challenged by molecular phylogenetic studies based on partial small subunit (SSU) rdna sequences. These studies revealed some unexpected groupings of rotaliid genera. However, the support for the new clades was rather weak, mainly because of the limited length of the analysed fragment. In order to improve the resolution of the phylogeny of the rotaliids, 26 new complete SSU rdna sequences have been obtained. Phylogenetic analyses of these data, together with seven sequences obtained previously, confirm with stronger statistical support the presence of three major clades among the Rotaliida. The first clade comprises members of the families Uvigerinidae, Cassidulinidae and Bolivinidae. The second clade includes all analysed Discorbidae, Rosalinidae, Planulinidae, Planorbulinidae, Rotaliidae, Elphidiidae, Nummulitidae and one of the Nonionidae. Finally, the third clade comprises the Cibicididae, Pseudoparreliidae, Oridorsalidae, Stainforthiidae, Buliminidae and part of the Nonionidae. The clades 1 and 3 are strongly supported by analyses of the complete SSU rdna, while the monophyly of clade 2 is less certain, probably due to the rapid evolutionary rates of some lineages included in this clade. These results clearly contradict the classical separation of rotaliid foraminifera into two orders: Rotaliida and Buliminida. Relatively good agreement has been found between molecular data and the morphological definition of the families for which more than one genus was sequenced. However, larger taxon sampling will be necessary for a better definition of the three major clades. 2007 Elsevier B.V. All rights reserved. Keywords: Benthic foraminifera; Rotaliida; Molecular phylogeny; Classification; SSU rdna 1. Introduction Benthic foraminifera are important elements of the meiofaunal community; their abundance data as well as Corresponding author. Geological Institute, ETH Zurich, Switzerland. E-mail address: magali.schweizer@erdw.ethz.ch (M. Schweizer). the chemical composition of their fossil tests are extensively used as proxies for environmental changes in the geological past (Van der Zwaan et al., 1999). Together with their potential use for monitoring recent environments, this has led to an increasing interest in their biological functioning. Since the second half of the last century research has focused on the use of foraminifera for environmental and pollution issues (e.g. Bandy et al., 0377-8398/$ - see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.marmicro.2007.10.003

234 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 1964; Vénec-Peyré, 1981; Van der Zwaan and Jorissen, 1991; Alve, 1995; Mojtahid et al., 2006). Proxy research, relevant to applications in the fossil record addresses, for instance, the response to environmental factors such as organic carbon flux (e.g. Smart et al., 1994; Murray, 2001; Ernst et al., 2005). The preservation of foraminiferal tests is important in this type of research (e.g. Berger, 1979; Mackensen and Douglas, 1989; Murray, 1989), and so are the vital effects associated with incorporation into calcareous tests of elements (e.g. Bentov and Erez, 2006) and stable isotopes (e.g. Schmiedl et al., 2004). In addition, research into the role of benthic foraminifera in biogeochemical cycling has recently resulted in the discovery that some species of foraminifera are able to perform complete denitrification (Risgaard-Petersen et al., 2006) and in doing so could play an important, though as yet not quantified, role in the marine nitrogen cycle. In view of foraminiferal evolution this raises the question whether traits like nitrate reduction developed separately in different clades, and when. Finally, cryptic speciation occurs in planktic (e.g., Darling et al., 2006) as well as in benthic foraminifera (e.g. Hayward et al., 2002); although in Uvigerina the opposite has been described recently (Schweizer et al., 2005). Either way, the use of foraminifera as (paleo-) environmental proxies needs to be reconsidered. In view of these developments, phylogenetic relationships between foraminifera, which were previously predominantly based on morphology, need a re-evaluation. The order Rotaliida comprises calcareous hyaline perforate species and represents one third of the extant, described genera of foraminifers (based on data from Decrouez, 1989). The taxon was introduced by Delage and Hérouard in 1896 as the suborder Rotalidae (reference in Loeblich and Tappan, 1964). The hyaline foraminifers were recognized earlier as a group by Williamson (1858), who was the first author to use the wall composition in foraminiferal classification (Cifelli and Richardson, 1990). However, this distinction was not always adhered to in early classifications. For example, Reuss (1861), Carpenter et al. (1862), Brady (1884), Rhumbler (1895) and Cushman (1922) placed arenaceous textulariids and hyaline bolivinids (sometimes with buliminids and cassidulinids) within the same group; this was also the case in some more recent publications (Hofker, 1951, 1956; Mikhalevich and Debenay, 2001). The classifications of Cushman (1928) and Galloway (1933) separated the hyaline genera from other foraminifera and placed them in several families. In 1964, Loeblich and Tappan grouped all hyaline perforate foraminifera in the suborder Rotaliina which included ten superfamilies (i.e. Nodosariacea, Buliminacea, Discorbacea, Spirillinacea, Rotaliacea, Globigerinacea, Orbitoidacea, Cassidulinacea, Carterinacea and Robertinacea) distinguished on the basis of the wall microstructure as well as coiling and aperture characteristics. In 1981, Haynes retained the wall structure as the primary basis of subdivision; however, more emphasis was given to the shape of the aperture. In his classification, the superfamilies were raised to orders, and the Nodosariida, Robertinida, Buliminida and Globigerinida were separated from the Rotaliida. The Buliminida comprised the hyaline perforate foraminifers with a toothplate and included the following superfamilies: Buliminacea, Bolivinitacea and Cassidulinacea, whereas the Rotaliida contained the Spirillinacea, Discorbacea, Asterigerinacea and Orbitoidacea (Haynes, 1981). In their monumental work, Loeblich and Tappan (1988) defined supplementary suborders for the hyaline perforate calcareous foraminifers in addition to their former classification (1964): Involutinina, Spirillinina, Carterinina, Silicoloculinina, Lagenina, Robertinina and Globigerinina. The suborder Rotaliina was divided into 24 superfamilies; the criteria used were the number of chambers, the presence or absence of perforations, canals and cavities in the test, and the aperture. In 1992, to solve some of the inconsistencies reported by Haynes (1981, 1990), Loeblich and Tappan raised the foraminifera from an order to a class (the foraminiferal suborders were thus given order status) and recognized the order Buliminida Fursenko, 1958. In the most recent classification, Sen Gupta (2002) followed the classification of Loeblich and Tappan (1992) with slight modifications (Fig. 1). According to this view, Buliminida are distinguished from Rotaliida by the presence of a toothplate, a loopshaped aperture and a high trochospiral coil. Since the mid-90s, the molecular approach has shed new light on foraminiferal phylogeny. The first molecular results showed a clade combining the Textulariida and the Rotaliida (Pawlowski et al., 1997; Pawlowski, 2000). This mixed clade was explained either by a radiation occurring in a relatively short time, or by slow rates of evolution of both groups (Pawlowski et al., 1997, 2003). More recent analyses were able to separate the rotaliids and textulariids, albeit without strong statistical support (Holzmann et al., 2003; Ertan et al., 2004; Bowser et al., 2006). Within the rotaliids, Holzmann et al. (2003) examined the links between the Nummulitidae and seven other rotaliid families, Ertan et al. (2004) explored the relationships of eleven genera of Rotallida, whereas Schweizer et al. (2005) investigated the position of uvigerinids.

M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 235 Fig. 1. Diagram showing the taxonomic positions of the genera studied in this paper and in Schweizer et al. (2005) with the classifications of Sen Gupta (2002). The first column represents the 22 extant superfamilies with the ones represented in phylogenetic analyses in bold. The second column represents the studied families and the third one the studied genera. All these studies were based on a fragment of about 1000 base pairs (bp) long, and situated at the 3 end of the SSU rdna. The phylogenetic information contained in this fragment seems to be insufficient to resolve rotaliid phylogeny, as shown by low statistical support for the deep nodes in the analyses of Ertan et al. (2004) and Schweizer et al. (2005). For this reason, we decided to sequence the complete SSU of selected rotaliid species which would result in a three-fold increase in the number of analysed sites. Therefore, 26 new complete sequences belonging to 12 of the 22 extant superfamilies of Rotaliida (according to the classification of Sen

236 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 Gupta, 2002) were obtained. These new data, along with seven previously obtained sequences, were analysed together to investigate the phylogeny of the rotaliids. 2. Materials and methods 2.1. Collection of the samples Live specimens of rotaliids were collected during several different expeditions between 1995 and 2006 (see Table 1 for details). Sediment samples were either taken by hand with a scraper at the shallow sites or collected by boxcoring and multicoring at the deeper sites. The top few centimetres of sediment were collected from the cores with a spoon and immediately sieved using water from the same environment as the sampling site (fractions 500/250/125 μm). The different fractions were stored at temperatures close to that of the collection site. Specimens were selected alive, imaged and extracted as explained in Schweizer et al. (2005). 2.2. DNA extraction, PCR amplification, cloning and sequencing Extraction of DNA from single or multiple specimens was done with DOC lysis buffer, CTAB or guanidine buffer (Pawlowski, 2000), and occasionally with DNeasy Plant Mini Kit (Qiagen) for multispecimen samples. The SSU was amplified in three steps with fragments between 800 and 1600 bp long. The primers sa10 and s13 were used to amplify the 5 end fragment (sa-s6), the primers s6f and s17 for the middle fragment (s6-s14) and the primers s14f3 and sb for the 3 end fragment (s14-sb). A reamplification of the PCR products (nested PCR) was often necessary and was performed using the following primers: sa10-s6ra for the sa-s6 fragment, s6f-s15rot for the s6-s14 Table 1 List of new complete and partial SSU sequences with the origin of DNA samples, the SSU length (nt= nucleotides) and the accession numbers Superfamily Species Locality DNA isolate SSU length Access number Cassidulinacea Islandiella sp. Svalbard, Norway 2643 3278 nt DQ408638 Cassidulinoides parkerianus Terranova Bay, Antarctica 3924 3348 nt DQ408639 Turrilinacea Stainforthia fusiformis Oslo Fjord, Norway 3965 sa-s6 DQ205387 Dunstaffnage, Scotland 3979 s6-s14 DQ452714 Buliminacea Bulimina marginata Oslo Fjord, Norway 3599 3462 nt DQ408646 Uvigerina phlegeri Setúbal Canyon, Portgual U239 3579 nt DQ408641 Uvigerina earlandi McMurdo Sound, Antarctica 2187 3571 nt DQ408640 Uvigerina peregrina Oslo Fjord, Norway U27 3517 nt DQ408642 Discorbacea Discorbis rosea Florida, USA 753 3507 nt DQ408644 Rosalina sp. Culture 3675 3402 nt DQ408643 Discorbinellacea Epistominella vitrea Cape Evans, Antartica 2060 3463 nt DQ408647 Planorbulinacea Hyalinea balthica Oslo Fjord, Norway 3604 3631 nt DQ408645 Cibicides pachyderma Nazaré Canyon, Portugal C86 3409 nt DQ408652 Cibicides pachyderma Nazaré Canyon, Portugal C196 3431 nt DQ408653 Cibicides lobatulus Oslo Fjord, Norway C24 3526 nt DQ408649 Cibicides lobatulus Skagerrak, Sweden C120 3632 nt DQ408650 Cibicides lobatulus Marseille, France C170 3596 nt DQ408648 Cibicides sp. North Atlantic 2524 3536 nt DQ408651 Planorbulinella sp. Elat, Israel 358 3365 nt DQ452687 Nonionacea Melonis pompilioides Skagerrak, Sweden 1400 3556 nt DQ408657 Pullenia subcarinata McMurdo Sound, Antarctica 1148 3471 nt DQ408656 Pullenia subcarinata McMurdo Sound, Antarctica 1850 3472 nt DQ408655 Nonionella labradorica Svalbard, Norway 4836 sa-s6 EF534076 Skagerrak, Sweden 3966 s6-s14 DQ452713 Haynesina germanica Den Oever, The Netherlands 6008 3027 nt EF534074 Chilostomellacea Oridorsalis umbonatus Fram Strait, Arctic 5410 3373 nt EF534075 Rotaliacea Elphidium williamsoni Den Oever, The Netherlands 6009 2772 nt EF534073 Ammonia sp. 3015 nt EF534072 S. fusiformis and N. labradorica SSU sequences are hybrids of two genetically close samples. Concerning the uvigerinids, it was shown earlier (Schweizer et al., 2005; Schweizer, 2006) that the two species attributed to the genera Rectuvigerina (R. phlegeri) and Trifarina (T. earlandi) are in fact closer to Uvigerina peregrina than U. mediterranea and U. elongatastriata. It was therefore decided to give all these species the same genus name (see discussion in Schweizer, 2006, Ch. 6).

M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 237 fragment and s14f1-sb for the s14-sb fragment. The positions and sequences of the primers are summarized in Fig. 2. The PCR conditions were the following: total volume of 50 μl, denaturation at 94 C for 30 s, annealing for 30 s at 50 C for the amplification and at 52 C for the reamplification, extension at 72 C for 2 min with 40 cycles for the amplification and 35 cycles for the reamplification, final elongation for 5 min at 72 C. The positive PCR products were purified using High Pure PCR Purification Kit (Roche Diagnostics). A few amplifications of the s14-sb fragment were sequenced directly; all the others were cloned. Purified products were ligated in the pgem-t Vector system (Promega) or the Topo Cloning vector (Invitro Gene), and cloned using ultracompetent cells XL2-Blue MRF' (Stratagene). Sequencing reactions were prepared using an ABI-PRISM Big Dye Terminator Cycle Sequencing Kit and analysed with an ABI-377 DNA sequencer or an ABI-PRISM 3100 (Applied Biosystems), all according to the manufacturer's instructions. 2.3. Phylogenetic analysis The new sequences presented here were deposited in the EMBL/GenBank Nucleotide Sequence Database; their accession numbers are reported in Table 1. To extend the data set, other complete SSU sequences from the EMBL/GenBank database have been added (deposited by M. Holzmann (unpublished) and Pawlowski et al. (1996, 1999)). Out-group sequences have been chosen among textulariids, the closest relatives of rotaliids (Bowser et al., 2006). Presently, there are only two sequences available for the complete SSU and they are used in our analyses. Sequences were aligned manually by using Seaview software (Galtier et al., 1996), and the regions which were too variable to be properly aligned were removed. From an alignment of 5565 sites, 1766 sites were removed to obtain a final alignment of 3799 sites, 1745 of them containing no gap. The GTR + I + Γ model was chosen by Modeltest (Posada and Crandall, 1998). However, we preferred to also use another model (HKY +I+Γ) to make comparisons, as Modeltest tends to favour complex models (Kelchner and Thomas, 2007). Therefore, the maximum likelihood (ML) trees were obtained using PhyML 2.4.4 (Guidon and Gascuel, 2003) with the HKY (Hasegawa, Kishino, Yano) model (Hasegawa et al., 1985) allowing transitions and transversions to have potentially different rates, and the GTR (General Time Reversible) model allowing all types of substitution to have different rates (Lanave et al., 1984; Rodriguez et al., 1990). To correct for the among-site rate variations, the proportion of invariable sites (I) and the alpha parameter of gamma distribution (Γ), with eight rate categories, were Fig. 2. Localization and sequences of the primers used to amplify the three fragments of the complete SSU rdna.

238 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 estimated by the program and taken into account in all analyses. Non-parametric bootstrapping (BS) (Felsenstein, 1985) was performed with 100 replicates to assess the reliability of internal branches. The Bayesian analysis was performed with MrBayes 3.1.2 (Huelsenbeck and Ronquist, 2001), using the GTR + I + Γ model. Two independent analyses were performed at the same time with four simultaneous chains (one cold and three heated) run for 1,000,000 generations, and sampled every 100 generations with 2500 of the initial trees discarded as burn-in. The posterior probabilities (PP) were calculated at the same time. Fig. 3. Localization of the inserts found in Rotaliida (in black) compared to the schematised SSU secondary structure of Chlamydomonas reinhardtii (in grey, redrawn from Wuyts et al., 2004) with the names of the helices (in grey, Wuyts et al., 2001). F1 F6 are inserts found only in foraminifers, whereas V1 V9 are variable regions recognized in all eukaryotes (Wuyts et al., 2000). The minimum and maximum numbers of nucleotides (=nt) observed in rotaliids for each insert are indicated, with the number found in C. reinhardtii between brackets for the variable regions (V2 V9).

M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 239 3. Results 3.1. Sequence data The 26 complete SSU sequences presented here have a length ranging from 2772 (Elphidium) to 3768 nucleotides (Nonionella); the four shortest sequences of which belong to the fast evolving taxa Elphidium, Ammonia and Haynesina (see Table 1 for details). The GC content of the studied rotaliids is between 37 and 46%, with the lowest value observed in Cassidulinoides and the highest one in Ammonia. These values are higher than those recorded for Miliolida and Astrorhizida (29 32%), but comparable to the value of 45% found in Globigerinida (Pawlowski, 2000). The typical insertions observed in other foraminifera (e.g. Pawlowski et al., 1999) are also present in the rotaliid sequences (Fig. 3). Some of these insertions are Fig. 4. Phylogeny of Rotaliida inferred from complete SSU rdna sequences using the ML method (HKY+I+Γ). Tree is rooted on textulariids. Bootstrap values for ML analyses (HKY and GTR) and PP values for Bayesian analysis are indicated at the nodes. Species names written in bold designate new sequences, the others were taken from GenBank (accession numbers are added).

240 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 in regions of the SSU which show variations among eukaryotes (V1 V9), whereas others are only present in certain eukaryotes (8e/1, 23e/15 23e/17, 45e/1), or exclusively in foraminifers (F1 F6). The insertions F4 F6 belong to the s14-sb fragment and have been described elsewhere with different names (e.g. Darling et al., 1997 (regions F1 F3); de Vargas et al., 1997 (regions I, II and V)). 3.2. Phylogenetic analyses The phylogenetic analysis of the 26 complete SSU sequences of rotaliid foraminifera and the seven sequences from GenBank is presented in Fig. 4. Two textulariids (Trochammina sp. and Eggerelloides scabrum) are used as an outgroup. Analyses performed with PhyML (ML/HKY and ML/GTR) and MrBayes (GTR) give exactly the same topology. Three major clades emerge within the Rotaliida. Clade 1 comprises Bolivina, Cassidulinoides, Islandiella and Uvigerina. These genera commonly have an elongate test (Islandiella being an exception) and an aperture with a toothplate. The statistical support is 89% BS (ML/HKY) or more (96% BS for ML/GTR and 1.00 PP for Bayesian inference (BI)). Clade 2 includes Rosalina, Discorbis, Planorbulinella, Hyalinea, Pararotalia, Cycloclypeus, Heterostegina, Operculina, Nummulites, Ammonia, Elphidium and Haynesina, genera with a trochospiral to planispiral test. The statistical support of this clade is 77% BS (ML/ HKY) or higher (ML/GTR: 83% BS, BI: 1.00 PP). This second clade can be divided in three subgroups. The first subgroup comprises species with a trochospiral to planispiral test and an interiomarginal arch-shaped aperture, sometimes with the presence of a second identical aperture (Planorbulinella) or secondary openings (Discorbis). Genera included in this first subgroup are Rosalina, Discorbis, Planorbulinella and Hyalinea. The statistical support is rather good (88% BS or higher and 1.00 PP). The second subgroup includes Pararotalia, a genus with a low trochospiral test and an imperforate toothplate and the Nummulitidae, a family with planispiral tests and a complex canal system. This subgroup is less well supported statistically: 77% for ML/HKY, 83% for ML/GTR and 0.93 PP. The third subgroup is composed of three shallow-water and fast evolving genera: Ammonia, Elphidium and Haynesina. The test is a low trochospiral (Ammonia) or planispiral coil (Elphidium and Haynesina) with primary and secondary apertures. This is the best-supported subgroup of clade 2 (95% BS or higher). Clade 3 includes Nonionella, Bulimina, Stainforthia, Epistominella, Oridorsalis, Pullenia, Melonis and Cibicides, and combines species with elongate and lenticular tests. This clade has a high statistical support (ML/HKY+ML/GTR: 99% BS, BI: 1.00 PP). A first subgroup comprises elongate triserial tests with a loopshaped aperture (Bulimina, Stainforthia) and low trochospiral ones with an interiomarginal arch-like aperture (Nonionella, Epistominella). The statistical support of this subgroup is at least 83% BS. A second subgroup includes species with a planispiral (Pullenia, Melonis) or low trochospiral (Oridorsalis, Cibicides) test and a slit-like interiomarginal aperture. The support of this subgroup is even better with 97% BS or higher. 4. Discussion 4.1. Ribosomal RNA secondary structure and foraminiferal insertions The secondary structure of the s14-sb fragment has been studied by Ertan et al. (2004) and Habura et al. (2004). There is evidence that the three foraminiferal insertions found in this fragment (here referred to as F4, F5 and F6, see Fig. 3) are not introns and are retained in the final rrna inside the ribosome (Habura et al., 2004). However, at the same time it has been suggested by Pawlowski et al. (1996) that about 1000 nt are removed from the SSU of Ammonia during the maturation process of RNA. Ertan et al. (2004) have used the variable regions (here referred to as V7 V9, see Fig. 3) and the foraminiferal insertions (see above) to improve the alignment and for taxonomic purposes, but it was noticed that these regions were not always diagnostic and that convergent evolution could occur. In the data presented here, representing the complete SSU, half of the insertions present similar patterns in rotaliids, textulariids (Trochammina sp. and Eggerelloides scabrum) and globigerinids (Globorotalia inflata, G. hirsuta, Neogloboquadrina dutertrei) showing that these groups are closely related. If all the foraminiferal insertions are retained in the rrna, as could be the case for F4 F6 (Habura et al., 2004), there are certainly evolutionary constraints to modelling their DNA sequences. Therefore, the use of these insertions for taxonomic purposes, as was initiated by Ertan et al. (2004), is promising. However, constructing the secondary structure of foraminiferal rrna is rather difficult, because of all the original insertions and the rate heterogeneity, and therefore any conclusions based on rrna structure should be treated with caution.

M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 241 4.2. Complete versus partial SSU phylogeny A question hotly debated is whether adding more taxa, or more sites will improve phylogenetic accuracy (e.g. Hedtke et al., 2006 and references therein). It has been shown that increasing the taxon sampling could improve the accuracy of phylogenetic inferences, particularly in cases of whole genomes sequenced for a few taxa (Geuten et al., 2007 and references therein). In the case of foraminifers, however, the situation is rather different. For the moment, there is a wide taxon sampling on a portion of the SSU and we are therefore in the situation described by Hillis et al. (2003), where a few characters are obtained from the taxa analysed. The authors argue that in this case, it is better to add more characters per taxon and we followed this suggestion by sequencing the complete SSU for some taxa, for which the partial SSU sequences were obtained previously. The 24 rotaliid genera present in the tree based on the complete SSU (Fig. 4) represent 65% of the rotaliid genera for which partial SSU rdna sequences were deposited in GenBank. The general topology of this tree is relatively congruent with previous studies based on partial SSU sequences (Holzmann et al., 2003; Ertan et al., 2004; Schweizer et al., 2005). However, important differences exist in relations between higher taxa and the support for particular groupings. The three clades found in the complete SSU tree were also identified in our previous analysis based on the s14- sb fragment (Schweizer et al., 2005, Fig. 7). However, the statistical support of the lower nodes is particularly improved with the analysis of the complete SSU compared to our former analysis, where the clade 1 branched as a sister-group of clade 3 instead of clade 2, albeit the relations between these clades were not supported (33% for the grouping of clades 1 and 3 in the partial SSU). The structure of particular clades is similar both in the present analyses and those in Schweizer et al. (2005). In clade 1, the relations between Bolivina, Cassidulinidae and Uvigerinidae are the same, but this clade comprised also Globobulimina in analyses of the partial SSU. In clade 2 the two subgroups are present with similar topology, except for the first subgroup, where Hyalinea and Planorbulinella grouped together in the partial SSU phylogeny with a good support (75% BS with PhyML and HKY). The second subgroup comprises the Nummulitidae, which always forms a homogeneous and highly supported clade (N90% BS) and Pararotalia, which in the partial SSU phylogeny branched with a high support close to the Calcarinidae (a family not included in the complete SSU analysis). Ammonia, Elphidium and Haynesina were not represented in the s14-sb analysis. Clade 3 includes the same two subgroups in both analyses, but with slightly different topologies. Bulimina and Nonionella were also closely related in the partial SSU analysis. However, Virgulina and Virgulinella, two taxa not represented in the complete SSU phylogeny, branched closer to Nonionella. Epistominella and Stainforthia were sistergroups in the partial SSU phylogeny, albeit with a low statistical support (31% BS with PhyML and HKY). In the second subgroup, Chilostomella (not included in the present analysis) was the closest relative of Pullenia with a reasonable support (74% BS with PhyML and HKY), and Cibicides was branching with that subgroup instead of with Melonis, albeit with no statistical support. The global topology of the second clade also appeared in an analysis performed by Holzmann et al. (2003) on the s14-sb fragment and focusing on Nummulitidae. The Nummulitidae were equally well supported (87% BS or higher) and branched together with Pararotalia and the Calcarinidae, despite a low support with the ML analysis (51% BS). The general topology of the other subgroup was similar to the one observed in the present analysis. However, the Cassidulinidae and Bolivinidae (belonging to the first clade in this study) appeared as a sister-group of the sub-clade Planorbulina Rosalina, with Hyalinea less closely related. Moreover, the branching of Cassidulinidae Bolivinidae inside this subgroup in Holzmann et al. (2003) was not statistically supported. The main discrepancy between our present study and partial SSU analysis of Ertan et al. (2004) concerns the relations between members of our clade 3. In Fig. 7 of Ertan et al. (2004), Bulimina andvirgulinella, belonging to the first subgroup in our clade 3 branched as a sistergroup of taxa belonging to our clade 1, whereas Chilostomella and Melonis formed a clade sister to all the other rotaliid sequences. These relations were interpreted as evidence for partition between the orders Rotaliida and Buliminida, however, the support of these two deeper nodes was very weak (respectively 0.19 and 0.38 PP). Despite this discrepancy, the relations between other rotaliids were quite congruent with the present study. The study of Ertan et al. (2004) confirmed with good support (0.88 PP) the grouping between members of our clade 2 (Rosalina, Haynesina, Ammonia, Elphidium). Representatives of clade 1 (Bolivina, Uvigerina, Globobulimina) equally branched together with a good support (0.80 PP), but with a different topology than in Schweizer et al. (2005): Globobulimina branched closer to Bolivina instead of Uvigerina in the study of Ertan et al. (2004).

242 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246

M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 243 4.3. Molecular phylogeny and morphology-based classification of Rotaliida Because there is a good congruence between the trees based on complete and partial SSU sequences, an analysis has been performed including the complete SSU sequences from the present study and the partial sequences (s14f-sb) from Schweizer et al. (2005, Fig. 7)(Fig. 5). This recapitulative tree is compared to the mainstream morphological classifications, with the one of Sen Gupta (2002) as a summary (Fig. 1). All genera represented by more than one species were monophyletic, except Rosalina (Fig. 5). Four morphologically defined families for which different taxa could be sampled, i.e. Cassidulinidae, Uvigerinidae, Calcarinidae and Nummulitidae, received a high statistical support in the phylogenetic analyses. On the other hand, the Nonionidae (Melonis, Pullenia, Nonionella, Haynesina), the Rotaliidae (Pararotalia, Ammonia) and the Buliminidae (Bulimina, Globobulimina) appeared polyphyletic, questioning the morphological basis of their grouping and asking for further, preferably integrated morphological and genetic studies. At a higher taxonomic level, the superfamilies Buliminacea, Nonionacea and Planorbulinacea appear polyphyletic. The polyphyly of the first two superfamilies is already clear at the family level. The Planorbulinacea, which include Hyalinea, Planorbulinella, Planorbulina, and Cibicides, are split between clades 2 and 3 (Fig. 5). This is quite surprising given that Hyalinea has always been grouped with Cibicides (e.g. Loeblich and Tappan, 1964, 1988; Haynes, 1981) and that Planorbulina was thought to be a stage in the life cycle of Cibicides (Nyholm, 1961). Some clades observed in the molecular analyses were proposed by Haynes (1981) based on morphological studies. The genera Pullenia and Chilostomella were classified in the family Chilostomellidae and the genera Bulimina, Stainforthia, Virgulina and Virgulinella in the family Buliminidae, superfamily Buliminacea (Haynes, 1981, 1990). Haynes (1981) also included Epistominella in the Buliminacea, but in a different family (Turrilinidae). Our results partially corroborate Haynes' subdivision, although other Buliminacea (Uvigerina and possibly Globobulimina) branch well separated from this group. Our data do not support the separation between the orders Buliminida and Rotaliida as proposed in the latest classifications (Haynes, 1981; Loeblich and Tappan, 1992; Sen Gupta, 2002). Although the two main groups observed in the phylogenetic analyses of the complete SSU (clade 1 and clades 2 + 3, see Fig. 4) roughly correspond to these orders, several buliminid genera (Bulimina, Stainforthia, Virgulina, Virgulinella), branch within clade 3 together with some of the rotaliids. The position of these genera questions the pertinence of the criteria used to separate the two orders. Apparently, the presence of a toothplate, which is the main feature to distinguish the two orders, is not as important as was previously thought (e.g., Hofker, 1951, 1956; Mikhalevich and Debenay, 2001) and was already questioned by Haynes (1981, p. 24). Moreover, the fact that Cassidulina, a genus without a toothplate, is traditionally placed in the Cassidulinidae (position confirmed by the molecular analyses, see Schweizer et al., 2005), a family where other genera represented in this study do possess a toothplate, shows that the taxonomic value of this character is questionable. Another representative feature of the Buliminida is the high trochospiral coil. However, this characterizes only 70% of the members of this order (Haynes, 1981). In view of (paleo-) ecological applications of foraminifera it is interesting to note that Nonionella, which appears closely related to Bulimina, includes species that are able to denitrify nitrate, as well as Stainforthia and Globobulimina (Risgaard-Petersen et al., 2006), a process which was formerly recognized only in prokaryotes. The results presented here indicate that the denitrification pathway exists in at least two rotaliid clades: clade 1 with Globobulimina (some Uvigerina and Bolivina species are also possible candidates but further tests need to confirm this (N. Risgaard-Petersen and E. Geslin, pers. comm.)) and clade 3 with the pair Nonionella Stainforthia. 5. Conclusions and perspectives Currently, our results based on complete and partial SSU rdna sequences favour the existence of a unique order Rotaliida. This order is subdivided into three clades which could be considered as suborders. However, further molecular studies are needed to confirm this partition and to better define the three clades morphologically. Sequencing of the complete SSU rdna clearly improved the phylogenetic signal contained in analyses of the s14- Fig. 5. Phylogeny of Rotaliida based on complete (this study, genera written in black) and partial (Schweizer et al., 2005, genera written in grey) sequences using the ML method (HKY+I+Γ). The number of sequences is indicated between brackets for genera represented by more than one sequence. Tree is rooted on textulariids. Bootstrap values (100 replications) are indicated at the nodes by white dots when they are 60% or higher and by black dots when they are 95% or higher.

244 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 sb fragment. By multiplying the number of analysed sites by a factor three, very strong statistical support was obtained for most of the clades. Still, problems persist with some clades composed of fast evolving species (i.e. Ammonia, Elphidium, Haynesina), which are more sensitive to the long branch attraction (LBA) phenomenon. Analysis of protein-coding genes could resolve these problems. However, it is already known that at least two genes for which data are available (actin, β-tubulin), are too conserved to resolve the phylogeny of Rotaliida (Flakowski et al., 2005; Ujiie and Pawlowski, in preparation). Although it is expected that some mitochondrial genes could be helpful, the data are not yet available. At the same time, to revise the classification of Rotallida it will be compulsory to obtain complete SSU sequences for taxa belonging to the families that have not yet been examined. Acknowledgements We thank Elisabeth Alve, Stefan Agrenius, Henko de Stigter, Christoffer Schander, the crews of the R/V Trygve Braarud, Arne Tiselius, Pelagia and Hans Brattström, Sam Bowser, Sergei Korsun, Tomas Cedhagen and Maria Holzmann for their help in sampling, and José Fahrni and Delphine Berger for their technical assistance. This study was supported by the Dutch NWO/ALW grant 811.32.001 and Swiss NSF grants 3100A0-100415 and 200020-109639/1. Appendix A. Taxonomic notes Bulimina marginata d'orbigny, 1826, Ann. Sci. Nat., sér. 1, 7, p. 269, pl. 12, figs 10 12. Cassidulinoides parkerianus (Brady): Cassidulina parkeriana Brady, 1884, Rep. Chall. Exp., Zool., 9, 432, pl. 54, Figs. 11 16. Cibicides lobatulus (Walker and Jacob): Nautilus lobatulus Walker and Jacob, 1798, in Kanmacher, Adam's Ess. Micr., p. 642, pl. 14, Fig. 36. Cibicides pachyderma (Rzehak): Truncatulina pachyderma Rzehak 1886, Naturf. Ver. Brünn, 24, p. 87, pl. 1, Fig. 5a c. Discorbis rosea (d'orbigny): Rotalia rosea d'orbigny, 1826, Ann. Sci. Nat., sér. 1, 7, p. 272. Elphidium williamsoni Haynes, 1973, British Mus. Bull., Zool. Suppl., vol. 4, pp. 207 209. Epistominella vitrea Parker, 1953, Cush. Found. Foram. Res, Spec. Publ. 2, p. 9, pl. 4, Figs. 34 36, 40 41. Haynesina germanica (Ehrenberg): Nonionina germanica Ehrenberg, 1840, K. Akad. Wiss. Berlin, Physik.-Math. Kl., Abh., Jahrg. 1839, pl. 2, Fig. 1. Hyalinea balthica (Schroeter): Nautilus balthicus Schroeter, 1783, Einleitung in die Conchylienkenntnis nach Linné, 1, p. 20, pl. 1, Fig. 2. Melonis pompilioides (Fichtel and Moll): Nautilus pompilioides Fichtel and Moll, 1798, Test. Micr. Arg. Naut., p. 31, pl. 2, figs a c. Nonionella labradorica (Dawson): Nonionina labradorica Dawson, 1860, Canadian Nat. Geol., 5, p. 192, Fig. 4. Oridorsalis umbonatus (Reuss): Rotalina umbonata Reuss, 1851, Deutsch. Geol. Ges. Zeitschr., 3, p. 75, pl. 5, Fig. 35. Pullenia subcarinata (d'orbigny): Nonionina subcarinata d'orbigny, 1839, Voyage dans l'amérique Méridionale; Foraminifères, 5, pt. 5, p. 28, pl. 5, Figs. 23 24. Stainforthia fusiformis (Williamson): Bulimina pupoides d'orbigny 1846 var. fusiformis Williamson, 1858, Rec. foram. G. B., p. 63, pl. 5, Figs. 129 130. Uvigerina earlandi (Parr): Angulogerina earlandi Parr 1950, Brit. N. Z. Ant. Res. Exp., 5, p. 341, pl. 12, Fig. 21. Uvigerina peregrina Cushman 1923, U. S. Nat. Mus. Bul., 104, pt. 4, p. 166, pl. 42, figs 7 10. Uvigerina phlegeri (Le Calvez): Rectuvigerina phlegeri Le Calvez 1959, Rec. Trav. Inst. Pêches Maritimes, 23, p. 263, pl. 1, Fig. 11. References Alve, E., 1995. Benthic foraminiferal response to estuarine pollution: a review. Journal of Foraminiferal Research 25, 190 203. Bandy, O.L., Ingle, J.C., Resig, J.M., 1964. Foraminifera, Los Angeles County outfall area, California. Limnology and Oceanography 9, 124 137. Bentov, S., Erez, J., 2006. Impact of biomineralization processes on the Mg content of foraminiferal shells: a biological perspective. Geochemistry, Geophysics, Geosystems 7, Q01P08. doi:10.1029/ 2005GC001015. Berger, W.H., 1979. Preservation of foraminifera. In: Lipps, J.H., Berger, W.H., Buzas, M.A., Douglas, R.G., Ross, C.A. (Eds.), Foraminiferal Ecology and Paleoecology. SEPM Short Course, vol. 6, pp. 105 155. Bowser, S.S., Habura, A., Pawlowski, J., 2006. Molecular evolution of foraminifera. In: Katz, L.A., Bhattacharya, D. (Eds.), Genomics and Evolution of Microbial Eucaryotes. University Press, Oxford, Oxford, pp. 78 93. Brady, H.B., 1884. Report on the foraminifera dredged by H.M.S. Challenger during the years 1873 1876. Reports of the scientific results of the voyage of H.M.S. Challenger, 1873 1876. Zoology 9, 1 814. Carpenter, W.B., Parker, W.K., Jones, T.R., 1862. Introduction to the Study of Foraminifera. Ray Society, London. Cifelli, R., Richardson, S.L., 1990. A history of the classification of foraminifera. Cushman Foundation for Foraminiferal Research 27, 1 119 Special Publications. Cushman, J.A., 1922. The foraminifera of the Atlantic Ocean, 3: Textulariidae. United States National Museum Bulletin 104, 1 149.

M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 245 Cushman, J.A., 1928. Foraminifera: Their Classification and Economic Use. Cushman Laboratory for Foraminiferal Research, Sharon. Darling, K.F., Wade, C.M., Kroon, D., Leigh Brown, A.J., 1997. Planktic foraminiferal molecular evolution and their polyphyletic origins from benthic taxa. Marine Micropaleontology 30 (3 4), 251 266. Darling, K.F., Kucera, M., Kroon, D., Wade, C.M., 2006. A resolution for the coiling direction paradox in Neogloboquadrina pachyderma. Paleoceanography 21, PA2011. doi:10.1029/2005pa001189. Decrouez, D., 1989. Generic ranges of Foraminiferida. Revue de Paléobiologie 8 (1), 263 321. de Vargas, C., Zaninetti, L., Hilbrecht, H., Pawlowski, J., 1997. Phylogeny and rates of molecular evolution of planktonic foraminifera: SSU rdna sequences compared to the fossil record. Journal of Molecular Evolution 45, 285 294. Ernst, S.R., Bours, R., Duijnstee, I.A.P., Van der Zwaan, G.J., 2005. Experimental effects of an organic matter pulse and oxygen depletion on a benthic foraminiferal shelf community. Journal of Foraminiferal Research 35, 177 197. Ertan, K.T., Hemleben, V., Hemleben, C., 2004. Molecular evolution of some selected benthic foraminifera as inferred from sequences of the small subunit ribosomal DNA. Marine Micropaleontology 53, 367 388. Felsenstein, J., 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783 791. Flakowski, J., Bolivar, I., Fahrni, J., Pawlowski, J., 2005. Actin phylogeny of foraminifera. Journal of Foraminiferal Research 35 (2), 93 102. Galloway, J.J., 1933. A Manual of Foraminifera. James Furman Kemp Memorial Series, no. 1. Principia Press, Bloomington. Galtier, N., Gouy, M., Gautier, C., 1996. SEAVIEW and PHYLO_WIN: two graphic tools for sequence alignment and molecular phylogeny. Computer Applications in Biosciences 12, 543 548. Geuten, K., Massingham, T., Darius, P., Smets, E., Goldman, N., 2007. Experimental design criteria in phylogenetics: where to add taxa. Systematic Biology 56 (4), 609 622. Guidon, S., Gascuel, O., 2003. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52 (5), 696 704. Habura, A., Rosen, D.R., Bowser, S.S., 2004. Predicted secondary structure of the foraminiferal SSU 3 major domain reveals a molecular synapomorphy for granuloreticulosean protests. Journal of Eukaryotic Microbiology 51 (4), 464 471. Hasegawa, M., Kishino, H., Yano, T.-A., 1985. Dating of the humanape splitting by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution 22, 160 174. Haynes, J.R., 1981. Foraminifera. Macmillan, London. Haynes, J.R., 1990. The classification of the foraminifera: a review of historical and philosophical perspectives. Palaeontology 33 (3), 503 528. Hayward, B.W., Grenfell, H.R., Holzmann, M., Pawlowski, J., 2002. The world's most misidentified forams; morphological discrimination of molecular types in ammonia. In: Grenfell, H.R. (Ed.), Geological Society of New Zealand Annual Conference, Northland 2002; Programme and Abstracts, vol. 112A. Geological Society of New Zealand, p. 24. Miscellaneous Publication. Hedtke, S.M., Townsend, T.M., Hillis, D.M., 2006. Resolution of phylogenetic conflict in large data sets by increased taxon sampling. Systematic Biology 55 (3), 522 529. Hillis, D.M., Pollock, D.D., McGuire, J.A., Zwickl, D.J., 2003. Is sparse taxon sampling a problem for phylogenetic inference? Systematic Biology 52 (1), 124 126. Hofker, J., 1951. The foraminifera of the Siboga expedition, 3: Ordo Dentata, sub-ordines Protoforaminata, Biforaminata, Deuteroforaminata. Uitk. Zool. Bot. Oceanogr. Geol. Geb. IVa, 1 513. Hofker, J., 1956. Foraminifera Dentata: foraminifera of Santa Cruz and Thatch-Island Virginia-Archipelago, West Indies. Spolia Zoologica Musei Hauniensis 15, 1 237. Holzmann, M., Hohenegger, J., Pawlowski, J., 2003. Molecular data reveal parallel evolution in nummulitid foraminifera. Journal of Foraminiferal Research 33 (4), 277 284. Huelsenbeck, J.P., Ronquist, F., 2001. MRBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17 (8), 754 755. Kelchner, S.A., Thomas, M.A., 2007. Model use in phylogenetics: nine key questions. Trends in Ecology and Evolution 22 (2), 87 94. Lanave, C., Preparata, G., Saccone, C., Serio, G., 1984. A new method for calculating evolutionary substitution rates. Journal of Molecular Evolution 20, 86 93. Loeblich, A.R., Tappan, H., 1964. Sarcodina, chiefly Thecamoebians and Foraminiferida. In: Moore, R.C. (Ed.), Treatise on Invertebrate Paleontology. Geological Society of America/University of Kansas, New York. Loeblich, A.R., Tappan, H., 1988. Foraminiferal Genera and Their Classification, 1. Van Nostrand Reinhold, New York. Loeblich, A.R., Tappan, H., 1992. Present status of foraminiferal classification. In: Takayanagi, Y., Saito, T. (Eds.), Studies in Benthic Foraminifera: Proceedings of the Fourth International Symposium on Benthic Foraminifera, Sendai, 1990. Tokai University Press, Tokyo, pp. 93 102. Mackensen, A., Douglas, R.G., 1989. Downcore distribution of live and dead deep-water benthic foraminifera in box cores from the Weddell Sea and the California continental borderland. Deep-Sea Research 36, 879 900. Mikhalevich, V., Debenay, J.-P., 2001. The main morphological trends in the development of the foraminiferal aperture and their taxonomic signifiance. Journal of Micropalaeontology 20, 13 28. Mojtahid, M., Jorissen, F., Durrieu, J., Galgani, F., Howa, H., Redois, F., Camps, R., 2006. Benthic foraminifera as bio-indicators of drill cutting disposal in tropical east Atlantic outer shelf environments. Marine Micropaleontology 61, 58 75. Murray, J.W., 1989. Syndepositional dissolution of calcareous foraminifera in modern shallow-water sediments. Marine Micropaleontology 15, 117 121. Murray, J.W., 2001. The niche of benthic foraminifera, critical thresholds and proxies. Marine Micropaleontology 41, 1 7. Nyholm, K.-G., 1961. Morphogenesis and biology of the foraminifer Cibicides lobatulus. Zoologiska Bidrag Fran Uppsala 33, 157 196. Pawlowski, J., 2000. Introduction to the molecular systematics of foraminifera. Micropaleontology 46 (suppl. 1), 1 12. Pawlowski, J., Bolivar, I., Fahrni, J.F., Cavalier-Smith, T., Gouy, M., 1996. Early origin of foraminifera suggested by SSU rrna gene sequences. Molecular Biology and Evolution 13 (3), 445 450. Pawlowski, J., Bolivar, I., Fahrni, J.F., de Vargas, C., Gouy, M., Zaninetti, L., 1997. Extreme differences in rates of molecular evolution of foraminifera revealed by comparison of ribosomal DNA sequences and the fossil record. Molecular Biology and Evolution 14 (5), 498 505. Pawlowski, J., Bolivar, I., Fahrni, J.F., de Vargas, C., Bowser, S.S., 1999. Molecular evidence that Reticulomyxa filosa is a freshwater naked foraminifer. Journal of Eukaryotic Microbiology 46 (6), 612 617. Pawlowski, J., Holzmann, M., Berney, C., Fahrni, J., Gooday, A.J., Cedhagen, T., Habura, A., Bowser, S.S., 2003. The evolution of early foraminifera. Proceedings of the National Academy of Sciences of the United States of America 100 (20), 11494 11498.

246 M. Schweizer et al. / Marine Micropaleontology 66 (2008) 233 246 Posada, D., Crandall, K.A., 1998. MODELTEST: testing the model of DNA. Bioinformatics 14 (9), 817 818. Reuss, A.E., 1861. Entwurf einer Systematischen Zummenstellung der Foraminiferen. Sitzungsberichte der Kaiserlichen Akademie der Wissenschaften, Wien 44, 354 396. Rhumbler, L., 1895. Entwurf eines naturlichen systems der Thalamorphoren. Nachrichten von der Königlichen Gesellschaft der Wissenschaften zu Göttigen, Mathematisch-Physikalische Klasse 1895 (1), 51 98. Risgaard-Petersen, N., Langezaal, A.M., Ingvardsen, S., Schmid, M.C., Jetten, M.S., Op den Camp, H.J., Derksen, J.W., Pina-Ochoa, E., Eriksson, S.P., Nielsen, L.P., Revsbech, N.P., Cedhagen, T., van der Zwaan, G.J., 2006. Evidence for complete denitrification in a benthic foraminifer. Nature 443 (7107), 93 96. Rodriguez, F., Oliver, J.F., Martin, A., Medina, J.R., 1990. The general stochastic model of nucleotide substitution. Journal of Theoretical Biology 142, 485 501. Schmiedl, G., Pfeilsticker, M., Hemleben, C., Mackensen, A., 2004. Environmental and biological effects on the stable isotope composition of recent deep-sea benthic foraminifera from the western Mediterranean Sea. Marine Micropaleontology 51, 129 152. Schweizer, M., 2006. Evolution and molecular phylogeny of Cibicides and Uvigerina (Rotaliida, Foraminifera). Geologica Ultraiectina 261, 1 167. Schweizer, M., Pawlowski, J., Duijnstee, I.A.P., Kouwenhoven, T.J., Van der Zwaan, G.J., 2005. Molecular phylogeny of the foraminiferal genus Uvigerina based on ribosomal DNA sequences. Marine Micropaleontology 57, 51 67. Sen Gupta, B.K., 2002. Systematics of modern foraminifera. In: Sen Gupta, B.K. (Ed.), Modern Foraminifera. Kluwer, Dordrecht, pp. 7 36. Smart, C.W., King, S.C., Gooday, A.J., Murray, J.W., Thomas, E., 1994. A benthic foraminiferal proxy of pulsed organic matter paleofluxes. Marine Micropaleontology 23, 89 99. Van der Zwaan, G.J., Jorissen, F.J., 1991. Biofacial patterns in river induced shelf anoxia. In: Tyson, R.V., Pearson, T.H. (Eds.), Modern and Ancient Continental Shelf Anoxia, vol. 58. Geological Society, pp. 65 82. Special Issue. Van der Zwaan, G.J., Duijnstee, I.A.P., den Dulk, M., Ernst, S.R., Jannink, N.T., Kouwenhoven, T.J., 1999. Benthic foraminifers: proxies or problems? A review of paleoecological concepts. Earth- Science Reviews 46, 213 236. Vénec-Peyré, M.T., 1981. Les foraminifères et la pollution: étude de la microfaune de la cale du Dourduff (embouchure de la rivière de Morlaix). Cahiers de Biologie Marine 25 33. Williamson, W.C., 1858. On the Recent Foraminifera of Great Britain. The Ray Society, London. Wuyts, J., de Rijk, P., van de Peer, Y., Pison, G., Rousseeuw, P., de Wachter, R., 2000. Comparative analysis of more than 3000 sequences reveals the existence of two pseudoknots in area V4 of eukaryotic small subunit ribosomal RNA. Nucleic Acids Research 28 (23), 4698 4708. Wuyts, J., van de Peer, Y., de Wachter, R., 2001. Distribution of substitution rates and location of insertion sites in the tertiary structure of ribosomal RNA. Nucleic Acids Research 29 (24), 5017 5028. Wuyts, J., Perrière, G., van de Peer, Y., 2004. The European ribosomal RNA database. Nucleic Acids Research 32, D101 D103.