131

RedescRiption of Unio gibbUs spengleR, 1793, a west palaeaRctic fReshwateR mussel with hookless glochidia

Rafael araujo*, carlos toledo & annie machordom

Museo nacional de Ciencias naturales (CsiC), José gutiérrez Abascal 2, 28006 Madrid, spain

aBstRact

as part on an ongoing effort to unravel the taxonomy and phylogeny of palaearctic freshwater mussels, this paper describes an overlooked Unio species that inhabits the atlantic rivers of south iberia and morocco, based on morphological, reproductive, and molecular characters. live specimens of Unio gibbus were recently found in the Barbate River in southwest spain and in the Beth River in the sebou basin, northwest morocco.

phylogenetic relationships inferred from partial coi and 16s rRna gene sequences point to the recognition of U. gibbus as an evolutionary unit. genetic divergences with respect to other Unio species ranged from 10.24% and 6.18% (compared to U. tumidus) to 12.91% and 8.89% (compared to U. tumidiformis) for coi and 16s rRna respectively.

in the spanish specimens, the entire internal cavity of the external demibranchs (homoge-neity) acts as a marsupium (ectobranchy), yet in the moroccan population, the four gills are occasionally filled with glochidia (occasional tetrageny). glochidia are rounded triangular, intermediate between the shapes of those in other species of Unio and Potomida. the border of the larval shell bears numerous small lumps, but these fail to form the ventral styliform hook typical of the genus Unio. these two characters, tetrageny and hookless glochidia, both of major significance in phylogenetic studies of the group, are reported here for the first time in the genus Unio.

the spanish populations of U. gibbus should be a main priority for european invertebrate conservation measures.

key words: Unio, hookless glochidia, iberian peninsula, morocco, mitochondrial dna.

intRoduction

most specialists would agree that the seminal work of haas (1969) on unionoid taxonomy forms the basis for any modern study on the taxonomy and phylogeny of this mollusc group. the genus Unio was considered by haas (1969) as a series of 12 “fundamental Unio species” comprised of different “races” or incipi-ent species. although this author’s work was mostly based on conchology, haas laid down the basic framework for current freshwater mussel systematics. in a recent review of the systematics and global diversity of the freshwa-ter mussel species, graf & cummings (2007) only considered seven species under the genus Unio. thus, the taxonomy of this polymorphic group now needs to be reconstructed as new molecular data and further biological char-acters, such as shell morphology, anatomy,

malacologia, 2009, 51(1): 131−141

*corresponding author: rafael@mncn.csic.es

glochidium, come to light. Recent studies on european unionoid taxonomy necessitate the formulation of new hypotheses regarding the number of evolutionary independent taxa in this imperilled mollusc group. with haas’ work as the base, recent conchological, anatomical, reproductive and molecular data have revealed several new features of the taxonomy of west palaearctic naiads: (1) the two iberian mediter-ranean Unio discussed by haas (1969) – U. elongatulus penchinatianus Bourguignat, 1865, and U. elongatulus valentinus Rossmässler, 1854 – are actually synonyms of U. mancus lamarck, 1819 (araujo et al., 2005); (2) the two iberian Potomida discussed by haas (1969) – P. littoralis littoralis cuvier, 1798, and P. littoralis umbonata Rossmaessler, 1844 – are synonyms of P. littoralis cuvier, 1798 (Reis et al., in review); (3) Unio tumidiformis castro, 1885, an overlooked iberian endemic, is being

aRaujo et al.132

redescribed (Reis & araujo, in review); and (4) Margaritifera marocana pallary, 1918, has also been redescribed as a valid species that differs from M. auricularia (spengler, 1793) (araujo et al., in press).

as part of an ongoing study of the phylogeny of the palaearctic unionoids, the present paper describes another overlooked Unio species based on the character set mentioned above. the species was previously considered a synonymy of the southwest atlantic iberian U. pictorum delphinus spengler, one of the 13 taxa into which haas (1969) separated the west palaearctic species of Unio pictorum (linnaeus, 1758).

mateRials and methods

to determine the historical distribution of the species, we reviewed spengler (1793), Bour-guignat (1864, 1865), drouet (1893), simpson (1900), haas (1913, 1917, 1969), pallary (1923, 1927, 1936), azpeitia (1933), daget (1998), soriano et al. (2001), and knudsen et al. (2003). Recent specimens were obtained by field sampling the Iberian Peninsula and Mo-rocco in 2006 and 2007; additional material was kindly provided by collaborators. specimens were collected by wading the rivers, sometimes using a water-scoop, or by snorkelling. some specimens (six from spain and eight from morocco) were transported alive in the cold (without water) to the laboratory where they were dissected and photographed.

shell characters were examined in adult and juvenile specimens from the two rivers where the species were found. detailed descriptions were made of shell shape and outline, colour, umbonal sculpture, and hinge characteristics. all relevant features were photographed.

anatomical studies were conducted on live and preserved specimens from the two rivers, paying close attention to characters identified by other authors as taxonomically relevant in unionoids (ortmann, 1911; haas, 1924; nagel, 1999; graf & cummings, 2006).

to characterize the larval stage, glochidia were collected from live mussels by their separation from gravid gills, cleaned with koh and gold-coated for scanning electron microscopy.

for molecular analyses, two foot samples from 15 specimens (six from spain and nine from morocco) were obtained and pre-served both frozen at -75°c and in absolute

ethanol. total dna was extracted using the chargeswitch gdna micro tissue (invitrogen) extraction kit or classic phenol/chloroform pro-cedures (sambrook et al., 1989). partial coi and 16S rRNA gene sequences were amplified by polymerase chain reaction (pcR) using the same primers as in machordom et al. (2003). the pcR mix contained 3µl dna, 5µl of the corresponding buffer (with 10x 2mm mgcl2), 1µl of dntps mix (10mm), 0.8µl of both prim-ers (10µm), 0.3µl Taq dna polymerase (5u/µl) (Biotools) and ddh2o in a total volume of 50µl. The amplification conditions for both genes were as follows: 94°c (4 min), 40 cycles of 94°c (45 s), 45°c (1 min), 72°c (1 min), and a final extension at 72°C (10 min).

the products were visualised with blue light in 0.8% agarose gels stained with sYBR safe (invitrogen). a co-migrating 100 pb or 1kb lad-der molecular-weight marker served to confirm correct amplification. The amplified fragments (around 700 bp) were purified by ethanol precipitation prior to sequencing both strands using Bigdye terminator (applied Biosystems, aBi). products were electrophoresed on an aBi 3730 genetic analyzer (applied Biosystems). the forward and reverse dna sequences obtained for each specimen were aligned and checked using the sequencher program (gene code corporation) after removing primer re-gions. when necessary, clustal X (thompson et al., 1997) was used to align the obtained sequences of 16S, but the final alignment was adjusted by eye.

for the phylogenetic analyses, we selected genBank Unio sequences representing the type localities of various species (or at least populations close to them), such as Unio pic-torum (linnaeus, 1758); U. tumidus philipsson, 1788; U. crassus philipsson, 1788; U. dougla-siae Gray, in Griffith & Pidgeon, 1833; U. caffer krauss, 1848; and U. mancus lamarck, 1819, and those from the iberian peninsula studied by Reis et al. (in review). three other species belonging to the same family, Potomida lit-toralis (cuvier, 1798), Amblema plicata (say, 1817) (both from the subfamily ambleminae) and Anodonta cygnea (linnaeus, 1758) (sub-family anodontinae), were also included in the analyses. Margaritifera auricularia (spengler, 1793), representing the family margaritiferidae, was used as an outgroup.

analyses were performed on both genes to-gether and separately for each gene. a partition homogeneity test (implemented in paup* as ild test: mickevich & farris, 1981; farris et al.,

RedescRiption of Unio gibbUs 133

1994) was used to assess congruence among data for the two genes. phylogenetic analyses (maximum parsimony – mp, maximum likelihood – ml, and neighbour-joining – nj) were con-ducted using paup* 4.0b10 (swofford, 2001). equal-weighted parsimony analyses were performed through heuristic searches with tBR branch swapping and ten random additions.

the appropriate model and starting param-eters for ml and nj analyses were chosen for each of the datasets using the akaike infor-mation criterion implemented in model test 3.06 (posada & crandall, 1998). maximum likelihood analysis was performed by heuristic searches, and the robustness of the topologies was assessed by bootstrap replications (1,000) (felsenstein, 1985).

Bayesian analyses using mrBayes 3.1.1 (huelsenbeck, 2000; huelsenbeck & Ronquist, 2001) were conducted for each dataset, each gene separately and both together. the Bayes-ian analysis using one cold and three incre-mentally heated chains started from a random tree. the models of nucleotide evolution were estimated following a model with rate classes, unequal base frequencies and the “invgamma”

option, for each gene separately and consider-ing each codon position in the case of the coi gene. the metropolis-coupled markov chain monte carlo (mcmcmc) analysis was run for 5,000,000 generations, sampling each 100 generations. Burn-in was 10%, after which likelihood values were stable. majority rule con-sensus trees were constructed using paup.

Results

Unio gibbus spengler, 1793Unio gibbus spengler, 1793: 64.Unio turdetanus drouet, 1893: 66, 67; pl. 1,

fig. 4.Unio tifleticus pallary, 1923: 78. figured in pal- 78. figured in pal-

lary, 1927: pl. 7, figs. 1, 2.Unio (Limniun) foucauldiana pallary, 1936: 63,

64; pl. 4, fig. 2.

Holotype: Zoological museum of the univer-sity of copenhagen, denmark. Zmuc BiV-434. illustrated by haas (1913) and knudsen et al. (2003). type locality: “tranquebar”, india, probably in error.

fig. 1. map showing the distribution of Unio gibbus. Black points indicate sites where live specimens have been recently discovered - 1: Barbate River; 2: Beth River; 3: Tiflet River [Unio tifleticus pallary, 1923]; 4: sous River [Unio (Limniun) foucauldiana pallary, 1936].

aRaujo et al.134

Distribution (fig. 1): atlantic rivers of south iberia and morocco. live specimens of U. gibbus have been recently found in two rivers only: Barbate River, cádiz, southwest spain, and two localities in the Beth River, sebou basin, northwest morocco. in the spanish river, the species shares its habitat with Potomida littoralis, Unio cf. pictorum and Anodonta cf. anatina (linnaeus, 1758), whereas in morocco it was found living with another species of Unio and with Potomida sp.

pallary (1923, 1927, 1936) cited Unio tifleti-cus in the Tiflet River, Sebou basin, and Unio (Limniun) foucauldiana in the sous River, south of agadir, morocco.

External Morphology (figs. 2, 3): shell swol-len, oval or rounded, equivalve, inequilateral, usually elevated at posterior end. outline, es-pecially in juvenile shells, resembling Potomida littoralis. anterior end rounded in a circular arc, posterior end rounded but dorsally elevated, forming a marked keel in ligamental area. shell thin, light; periostracum flaking in posterior and ventral areas. colour green (more common in moroccan specimens) to brown, sometimes with radial bands of clearer colour. umbones rounded, prominent, never eroded, clearer in colour that rest of shell. it is common to see the glochidium as a protoconch in the apex of the umbo (fig. 3B). umbo sculpture more marked

fig. 2. Unio gibbus specimens from different localities. a: Barbate River, spain; B: Beth River, morocco; c: juvenile specimens (Barbate River); d: umbonal sculpture (Barbate River).

RedescRiption of Unio gibbUs 135

in spanish specimens, consisting of two or three rows of strong tubercula parallel to growth lines of the shell. in juvenile shells, posterior tubercula extend along a diagonal row that is absent in adult shells (fig. 3a). left valve with two laminar pseudo-cardinal teeth, sometimes crenulated, which may or may not be well separated; posterior tooth beaked and more elevated. two laminar posterior lateral teeth variable in length, the lower tooth always more elevated. Right valve with pseudo-cardinal tooth generally beaked, sometimes flat, and very tall laminar lateral posterior tooth. maxi-mum measured shell length 85.6 mm.

Anatomy (fig. 4): soft parts typical of Unio. papillae of the inhalant siphon generally not bifurcated; exhalant opening confined by two lateral muscular strips with external black trimming, and separated dorsally from the supra-anal aperture by a bridge shorter than the supra-anal aperture.

Life Cycle: in the Barbate River, spain, females carried embryos in their gills at the end of february 2006, whereas in the Beth River, morocco, ripe glochidia were found in specimens collected in early june 2007. in the spanish specimens, the entire internal cavity of the external demibranchs (homogeneity) acted as a marsupium (ectobranchy) (fig. 5a), but in the moroccan population the four gills were oc-casionally filled with glochidia (occasional tetrag-eny) (fig. 5B, c). glochidia rounded triangular

(fig. 6), intermediate between glochidia shapes in Unio and Potomida. Border of the larval shell bears numerous small lumps enlarging toward the ventral portion where they become spicules (fig. 6f, g, h) but never form the ventral styli-form hook typical of the genus. mean glochidial dimensions (measured by sem): length: 209.17 µm (sd = 2.83; n = 19), height: 211 µm (sd = 3.93; n = 21), width: 67.67 µm (sd = 5.99; n = 3).

the phylogenetic relationships inferred from our data (genBank accession numbers: eu735753 to eu735781) point to the indepen-dence of U. gibbus as an evolutionary unit (fig. 7). in molecular terms, our results indicate a clear distinction between U. gibbus and the rest of the species analysed (fig. 7). the spanish specimens showed an absolute lack of genetic variability, with only one haplotype observed for each gene studied. the iberian and moroc-can specimens of U. gibbus displayed a small amount of intraspecific divergence (0.37% for coi), and all shared the same haplotype for the 16s gene. in contrast, divergences were 30 times higher when compared to other Unio species – 10.24% with respect to the clos-est U. tumidus to 12.91% with respect to U. tumidiformis da silva e castro, 1885, for coi, and 6.18% to 8.89% for 16s with respect to the same species, respectively. these results are based on the analysis of 657 coi and 500 16s characters. although support was low, especially for 16s, global phylogenetic analysis revealed U. tumidus as the sister species of U. gibbus (fig. 7). moreover, U. gibbus always

fig. 3. Unio gibbus specimen. a: detail of the umbonal sculpture; B: Remains of the glochidia present in the umbo.

aRaujo et al.136

fig. 4. anatomy of Unio gibbus from the Barbate River. a: general view; B: inhalant siphon; c: supra anal aperture.

fig. 5. Unio gibbus specimens. a: gravid external gill with embryos (Barbate River); B, c: gravid gills with glochidia (Beth River).

appeared as an independent lineage from a well-supported cluster that included the palae-arctic species U. cf. pictorum, U. pictorum, U. mancus, U. tumidiformis, and U. crassus (fig. 7), and even the african U. caffer according to the coi gene (fig. 7B). the chinese U.

douglasiae occupied a basal position in the 16s analyses (fig. 7a).

Anodonta cygnea appeared as sister group of the genus Unio in both, coi and global analyses, but surprisingly it was included in the Unio cluster in the 16s dataset.

RedescRiption of Unio gibbUs 137

discussion

simpson (1900: 861) considered U. gibbus as an indeterminate oriental unionidae. the synonymization of Unio turdetanus drouet, 1893, with Unio gibbus spengler, 1793, was mentioned by haas (1913), who also suggested

that the tranquebar locality of the spengler specimen was erroneous, and proposed that it had actually come from spain. this is a similar error to that concerning the locality of the euro-pean Unio auricularius spengler (= Margaritifera auricularia), which obviously did not come from east india, as indicated by spengler (1793).

fig. 6. glochidium of Unio gibbus from the Beth River. a, B: general view; c: external view; d: lateral view; e: posterior view; f-h: spicules; i: ligament.

aRaujo et al.138

fig. 7. majority-rule consensus trees from Bayesian phylogenetic relationships among members of the palaearctic genus Unio based on partial sequences of – a: 16s; B: coi; c: 16s + coi. numbers above branches indicate Bayesian posterior probabilities and maximum parsimony bootstrap values; those be-low branches indicate bootstrap values for maximum likelihood and neighbour-joining treatments.

RedescRiption of Unio gibbUs 139

later, haas (1917, 1969) listed U. turdetanus drouet, 1893, and U. gibbus, as synonyms of the southwest atlantic iberian Unio pictorum delphinus spengler, 1793, one of the 13 taxa into which this author divided U. pictorum. this was probably also why azpeitia (1933) ascribed U. gibbus to the same group as U. delphinus spengler and U. hispanus Rossmässler, 1844, both synonyms of Unio pictorum, although he distinguished U. gibbus from the other spe-cies.

Unio (Limniun) foucauldiana, which we syno-nymize here with U. gibbus, was included by daget (1998) as a synonym of the north african U. elongatulus durieui deshayes, 1847, one of the 17 taxa into which haas (1969) divided the mediterranean species U. elongatulus c. pfeiffer, 1825; nevertheless, U. (Limniun) foucauldiana was not included in haas (1969). surprisingly, Unio tifleticus, the other pallary species that we consider to be a synonym of U. gibbus, was synonymized by both haas (1969) and daget (1998) with Potomida littoralis fell-mani (deshayes, 1848), the north african race of P. littoralis.

according to the characters we examined, es-sentially morphological, molecular, and the glo-chidia, U. gibbus differs considerably from both U. pictorum and U. elongatulus, and logically from P. littoralis; it is clearly a different species, probably restricted to some atlantic iberian and moroccan rivers. our molecular data also sug-gest that U. gibbus is more closely related to U. tumidus than to other Unio species.

the genetic divergences we found were simi-lar or even greater to those observed among other unionoid species (lydeard et al., 1996; araujo et al., in press), but despite this clear distinction, the biogeographic implications to be inferred from the phylogenetic results are not so obvious. the current distribution of U. gibbus suggests differentiation in the southern iberian peninsula and morocco most likely, when both areas comprised a single landmass, in the Betic-Rifian massif (Steininger & Rögl, 1984). Now, given that the Betic-Rifian massif extended as far as the current algerian kabilies mountains, the possible spread of U. gibbus to the mediterranean rivers of north africa should be considered. this possibility is also suggested by the presence of Unio (Limniun) seurati pallary, 1936, in chèlif, algeria, a very similar species to U. gibbus that could probably be synonymized with it.

today’s distribution of U. gibbus is greater in the maghreb region than on the iberian penin-sula. the lower molecular variation noted within

the Iberian sample could reflect this reduced distribution range, indicating poor conditions in the iberian population resulting in reduction in the effective population size and a lack of gene flow or extinction across some of its potential distribution.

it is possible that due to the long ancient history of the group, true relationships among its members or their actual biogeographical patterns are difficult or impossible to recover. despite this, the results of the few recent stud-ies performed on west palaearctic unionoids (Reis et al., in review; araujo et al., in press) are beginning to unveil a much richer diversity than currently recognized. thus, further insight into the evolutionary history of Unio will have to wait until areas of such potentially high diversity areas as turkey, greece, and the maghreb region are subjected to detailed investigation.

Unio gibbus has not been found in recent surveys of many spanish south atlantic and mediterranean rivers, including several tributar-ies of the guadalquivir basin, and the locality described by drouet (1893) for U. turdetanus (= U. gibbus). the living population examined here was found in a short stretch of the Barbate River upstream from a large reservoir. the lack of genetic variability among these specimens, along with seasonal droughts and excessive water extractions from this river, make the spanish populations of U. gibbus a main prior-ity for invertebrate conservation measures. in morocco, at least in the Beth River, U. gibbus is a very common species.

Besides the external appearance of the shell and the molecular markers rendering the spe-cies unmistakable, the most important charac-ters that separate U. gibbus from other Unio species are tetrageny and hookless glochidia, both features informative in phylogenetic studies of the group. interestingly, the trait of tetrageny was only detected in one moroccan specimen, the four gills of which were filled with glochidia, the external gills being much more swollen than the internal ones. two other gravid moroccan specimens were carrying embryos but only in the external gills, as also occurred in the three gravid spanish specimens. these observa-tions suggest that the internal gills may be filled later, once the external ones are full: could the absence of embryos in the inner gills thus only be a question of time, or was the tetragenous specimen identified just an isolated case of �oc-“oc-casional tetrageny” with no phylogenetic implica-tions? this question will likely be answered as we obtain more information on the reproductive strategy of other populations of this species.

aRaujo et al.140

although a little broader, the size of the U. gibbus glochidium falls within the range cited for the already known species of the genus (pekkarinen & englund, 1995). however, its rounded shape, and the absence of the ventral hook make the larval shell a specific diagnostic character. it has been suggested that hooks might develop later in juvenile glochidia, such that young glochidia have no hooks, whereas older ones do (ortmann, 1911: 301). the glo-chidia in our specimen were not encapsulated, so we can assume they were fully developed. nevertheless, the idea that U. gibbus glochidia lack well-developed median ventral teeth should be also considered a hypothesis for fu-ture testing. as recently stated by graf & cum-mings (2006), it is yet too soon to know with certainty if tetrageny and hookless glochidia are, respectively, the plesiomorphic brooding morphology and larval type of the unionoida. the peculiar reproductive habits discovered in U. gibbus fuel this rationale and reveal there are still large gaps in our knowledge of the reproductive strategies of the naiad species of the palaearctic. perhaps the time has come to rethink the characters presently describing the genus Unio.

acknowledgements

the authors thank josé miguel Barea, elena Ballesteros, and joaquim Reis for their invaluable help collecting specimens, and ab-delhamid azeroual for sending us preserved specimens. plates were compiled by j. muñoz of the photography facility of the museo nacio-museo nacio-nal de ciencias naturales in madrid, spain. ana Burton revised the english version of the manuscript. the authors are also indebted to two anonymous referees for their constructive comments.

liteRatuRe cited

aRaujo, R., i. gómeZ & a. machoRdom, 2005, the identity and biology of Unio mancus (= U. elongatulus) (Bivalvia: unionidae) in the iberian peninsula. Journal of Molluscan studies, 71(1): 25–31.

aRaujo, R., c. toledo, d. Van damme, m. ghamiZi & a. machoRdom, in press, Marga-ritifera marocana (pallary, 1918) a valid current species inhabiting the moroccan rivers. Journal of Molluscan studies.

aZpeitia moRos, f., 1933, Conchas bivalvas de agua dulce de España y Portugal. memorias

del instituto geológico y minero de españa, madrid. 2 vols., 756 pp + 36 pls.

BouRguignat, m. j. R., 1864, Malacologie de L’Algérie. Histoire naturelle des animaux mollus-ques terrestres et fluviatiles recuellis jusqu’à ce jour dans nos possessions du nord de l’Afrique, Vol. 2. alger, paris. 380 pp., 17 pls.

BouRguignat, m. j. R., 1865, mollusques nouveaux, litigieux ou peu connus. Revue et Magazin de Zoologie Pure et Appliquée, 17: 337–347 + pls. 16–24.

daget, j., 1998, Catalogue raisonné des Mol-lusques bivalves d’eau douce africains. Back-huys, orstom. 329 pp.

dRouet, h., 1893, Unionidae de l’Espagne. j.-B. Baillière, paris. 88 pp., 2 pls.

faRRis, j. s. k., m. källeRsjö, a. g. kluge & c. Bult, 1994, Testing significance of incon-gruence. Cladistics, 10: 315–319.

felsenstein, j., 1985, Confidence limits on phylogenies: an approach using the bootstrap. Evolution, 39: 783–791.

gRaf, d. & k. cummings, 2006, palaeo-heterodont diversity (mollusca: trigonioida + unionoida): what we know and what we wish we knew about freshwater mussel evolution. Zoological Journal of the Linnean society, 148: 343–394.

gRaf, d. & k. cummings, 2007, Review of the systematics and global diversity of freshwater mussel species (Bivalvia: unionoida). Journal of Molluscan studies, 73(4): 291–314.

haas, f., 1913, Bemerkungen über spenglers unionen. Videnskabelige Meddelelser fra Dansk Naturhistorisk Forening I Kjobenhavn, 65: 51–66.

haas, f., 1917, estudio para una monografía de las náyades de la península ibérica. Anuari de la Junta de Ciencies naturals, 2(2): 131–190.

haas, f., 1924, anatomische untersuchungen an europäischen najaden. i. Archiv für Mollus-kenkunde, 56: 66–82.

haas, f., 1969, superfamilia unionacea. Das Tierreich, 88: 1–663.

huelsenBeck, j. p., 2000, Mrbayes: bayesian inference of phylogeny. university of Rochester, department of Biology, Rochester, new York.

huelsenBeck, j. p. & f. Ronquist, 2001, mrBayes: Bayesian inference of phylogeny. bioinformatics, 17: 754–755.

knudsen, j., k. R. jensen, c. nielsen & R. i. johnson, 2003, lorentz spengler´s descriptions of freshwater mussels (mollusca: unionacea): translation and notes. steenstrupia, 27(2): 263–279.

lYdeaRd, c. h., m. mulVeY & g. m. daVis, 1996, molecular systematics and evolution of reproductive traits of north american freshwa-ter unionacean mussels (mollusca: Bivalvia) as inferred from 16s rRna gene sequences. Philosophical Transactions of the Royal society of London, B, 351: 1593–1603.

machoRdom, a., R. aRaujo, e. eRpenBeck & m. a. Ramos, 2003, phylogeography and conservation genetics of endangered euro-pean margaritiferidae (Bivalvia: unionoidea).

RedescRiption of Unio gibbUs 141

biological Journal of the Linnean society, 78(2): 235–252.

mickeVich, m. f. & j. s. faRRis, 1981, the im-plications of congruence in Menidia. systematic biology, 30: 351–370.

nagel, k.-o., 1999, anatomische und mor-phologische merkmale europäischer najaden (unionoidea: margaritiferidae und unionidae) und ihre Bedeutung für die systematik. Heldia, 2(3): 33–48.

oRtmann, a. e., 1911, a monograph of the na-jades of pennsylvania. Memoirs of the Carnegie Museum, 4(6): 279–347.

pallaRY, p., 1923, descriptions de trois mol-lusques nouveaux du maroc. bulletin de la société des sciences naturelles du Maroc, 3: 78–79.

pallaRY, p., 1927, complément à la faune malacologique de la Berbérie. Journal de Con-chyliologie, 71: 197–277 + pls. 1–7.

pallaRY, p., 1936, deuxiéme complément à la faune malacologique de la Berbérie. Journal de Conchyliologie, 80: 5–65 + pls. 1–4.

posada, d. & k. a. cRandall, 1998, MOD-ELTEST: testing the model of dna substitution. bioinformatics, 14: 817–818.

pekkaRinen, m. & p. m. englund, 1995, sizes of intramarsupial unionacean glochidia in finland. Archiv Für Hydrobiologie, 134(3): 379–391.

Reis, j. & R. aRaujo, in review, Redescription of Unio tumidiformis castro 1885 (Bivalvia, unioni-dae), an endemism from southwestern iberian peninsula. Journal of natural History.

Reis, j., a. machoRdom & R. aRaujo, in review, Evolutionary and taxonomic significance of phenotypic plasticity in freshwater mussels (Bivalvia: unionidae) at the western extreme of

their european range. Zoological Journal of the Linnean society.

samBRook, j., e. f. fRitsch & t. maniatis, 1989, Molecular cloning: a laboratory manual, 2nd ed., 3 vols. cold spring harbor laboratory, cold spring harbor, new York.

simpson, c. t., 1900, synopsis of the naiades or pearly fresh–water mussels. Proceedings of the United states national Museum, 22(1205): 501–1044.

soRiano, o., m. Villena & m. s. alonso, 2001, catálogo de los uniónidos (mollusca, unionidae) de la península ibérica conservados en el museo nacional de ciencias naturales (csic). graellsia, 57(1): 133–152.

spengleR, l., 1793, om slaegterne Chaena, Mya og Unio. skrifter af naturhistorie selskabet, 3: 16–69.

steiningeR, f. f. & f. Rögl, 1984, paleoge-ography and palinspastic reconstruction of the neogene of the mediterranean and paratethys. pp. 659–668, in: j. e. diXon & a. h. f. RoBeRtson, eds., The geological Evolution of the Eastern Mediterranean. geological society special Publication 17. Blackwell Scientific, Palo Alto, california. vii + 824 pp.

swoffoRd, d. l., 2001, PAUP*: Phylogeny Analysis Using Parsimony (*and other meth-ods), Version 4.0b10. sinauer associates, sunderland, massachusetts.

thompson, j. d., t. j. giBson, f. plewniak, f. jeanmougin & d. g. higgins, 1997, the ClustalX windows interface: flexible strategies for multiple sequence alignment aided by qual-ity analysis tools. nucleic Acids Research, 24: 4876–4882.

Revised ms. accepted 5 november 2008