Zootaxa 3309: 1–35 (2012) ISSN 1175-5326 (print edition) www.mapress.com/zootaxa/ Article ZOOTAXA Copyright © 2012 · Magnolia Press ISSN1175-5334(online edition) Bothriocephalidean tapeworms (Cestoda) of freshwater fish in Africa, including erection of Kirstenella n. gen. and description of Tetracampos martinae n. sp. Ů ROMAN KUCHTA1, ALENA BURIANOVÁ1, MILOSLAV JIRK 1, ALAIN de CHAMBRIER2, MIKULÁŠ OROS1,3, JAN BRABEC1 & TOMÁŠ SCHOLZ1 Č 1Institute of Parasitology, Biology Centre of the Academy of Sciences of the Czech Republic, Branišovská 31, 370 05 eské ě Bud jovice, Czech Republic; e-mail: [email protected] 2Department of Invertebrates, Natural History Museum, P.O. Box 6434, CH-1211 Geneva 6, Switzerland 3Institute of Parasitology, Slovak Academy of Sciences, Hlinkova 3, 040 01 Košice, Slovakia Table of contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 Material and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3 Bothriocephalus acheilognathi Yamaguti, 1934 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 Ichthybothrium ichthybori Khalil, 1971 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Kirstenella Kuchta n. gen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8 Kirstenella gordoni (Woodland, 1937) Kuchta n. com. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9 Polyonchobothrium polypteri (Leydig, 1853) Lühe, 1900 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12 Tetracampos martinae Kuchta n. sp. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Tetracampos ciliotheca Wedl, 1861. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16 Bothriocephalus claviceps (Goeze, 1782) Rudolphi, 1810 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24 Phylogenetic relationships . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24 Key to the freshwater bothriocephalideans from Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Acknowledgments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Abstract A survey of bothriocephalidean tapeworms (Cestoda) parasitizing African freshwater fish is provided. Based on critical evaluation of type specimens and extensive, newly collected material, only the following seven species, instead of 19 taxa listed in the literature, are considered to be valid and their redescriptions are provided: Bothriocephalus acheilognathi Yamaguti, 1934 (with 3 synonyms from Africa); Bothriocephalus claviceps (Goeze, 1782) (marginally in Africa); Ichthybothrium ichthybori Khalil, 1971; Kirstenella gordoni (Woodland, 1937) n. comb. (1 synonym); Polyonchobothrium polypteri (Leydig, 1853) (4 synonyms); and Tetracampos ciliotheca Wedl, 1861 (4 synonyms). In addition, Tetracampos martinae Kuchta n. sp. is proposed for tapeworms from the catfish Bagrus meridionalis from Lake Malawi. The new species differs from T. ciliotheca in a much larger body (19 cm versus 3 cm), dorsoventally flattened strobila and numerous (39 versus 25–35) and longer apical hooks (up to 98 μm versus less than 50 μm). Kirstenella Kuchtan. gen. is proposed to accommodate Senga gordoni Woodland, 1937 as its type species. The new genus is distinguished from other genera of the Bothriocephalidae by the presence ofan apical disc armed with two lateral semicircles of large hooks, cortical vitelline follicles and large-sized cirrus-sac. All but one valid species were recollected. Bothriocephalidean cestodes are widely distributed throughout Africa, but only two species, B. acheilognathi and T. ciliotheca, occur in other continents. All but one species (B. acheilognathi) exhibit narrow host specificity, being limited either to one host species (K. gordoni in Heterobranchus bidorsalis and T. martinae in Bagrus meridionalis) or one host genus (I. ichthybori in Ichthyborus spp., P. polypteri in Polypterus spp. and T. ciliotheca in Clarias spp.). Molecular data based on partial sequences of the large subunit rDNA (lsrDNA) show monophyletic position of all African taxa analysed (B. acheilognathi,I. ichthybori,K. gordoni,P. polypteri and T. ciliotheca). Key words: Taxonomic revision, morphology, redescriptions, new genus, new species, phylogeny, identification key, zoogeography, host specificity Accepted by N. Dronen: 29 Feb. 2012; published: 11 May 2012 1 Introduction In Africa, more than 3,000 species of freshwater fish have been recorded, including members of the most ancient groups, such as lungfish (Dipnoi) and bichirs (Polypteriformes) (Lévêque et al. 2008; Froese & Pauly 2011). The helminth fauna of African teleosts has been studied since the middle 19th century, when Leydig (1853) and Wedl (1861) described the first tapeworms from bichirs and clariid fish, respectively. Khalil (1971a) published the first checklist of parasites of freshwater fish in Africa and in its updated edition (Khalil & Polling 1997), a total of 359 species of helminths, including 61 species of adult and larval tapeworms (Cestoda), were reported. Adult tapeworms identified to the species level belong to the orders Amphilinidea (1 species), Caryophyllidea (20 species in 7 genera), Bothriocephalidea (13/3) and Proteocephalidea (19/6). Recently, several taxonomic accounts, which contained critical reviews of tapeworms of two of these orders, Proteocephalidea and Caryophyl- lidea, have been published (de Chambrier et al. 2007, 2008, 2011; Scholz et al. 2009, 2011a; Schaeffner et al. 2011) and a new genus and species of proteocephalidean cestodes were described (de Chambrier et al. 2009). However, only limited information exists on the actual species composition, host specificity and distribution of members of a newly established order, Bothriocephalidea Kuchta, Scholz, Brabec & Bray, 2008 proposed to accommodate tapeworms with paired bothria on their scoleces, previously placed in the suppressed order Pseudo- phyllidea (see Kuchta et al. 2008a). The new order was revised by Kuchta et al. (2008b), who provided amended generic diagnoses, whereas Kuchta & Scholz (2007) proposed numerous synonymies of bothriocephalidean spe- cies, including taxa reported from African freshwater fish. Since publication of the checklist by Khalil & Polling (1997), a number of new host and geographical records from African freshwater fish have been published. However, unsatisfactorily resolved taxonomy of the group, questionable validity of several taxa and continuing use of names of apparently invalid species, such as Polyoncho- bothrium clarias (see Kuchta et al. 2008b), make these new records unreliable or confusing. It is thus impossible to use literary data for a reliable assessment of the actual diversity, distribution of individual taxa and their relation- ships with fish hosts. Recently, extensive new material was collected by the present authors and their co-workers in several African countries, including type-localities of several taxa (see below). Consequently, bothriocephalidean tapeworms para- sitizing freshwater fish in Africa were revised on the basis of morphological and taxonomic evaluation of this new material, supplemented by a study of all available type and voucher specimens and a critical analysis of literary data. Results of this revision are presented herein, including redescriptions of all but one species (Bothriocephalus acheilognathi, which is also distributed outside Africa), and data on their fish hosts and distribution. Tapeworms found in Bagrus meridionalis Günther from Lake Malawi represents a new, hitherto undescribed species, which is described in this paper. In addition, a new genus is proposed to accommodate Polyonchobothrium gordoni Wood- land, 1937 from clariid catfish. Material and methods Material collected by the present authors and their collaborators was obtained by the examination of more than 2,000 freshwater fish of 120 species of 23 families, carried out from 2006 to 2010 in the Democratic Republic of the Congo, Ethiopia, Gabon, Kenya and the Sudan (A. C., T. S., M. J., M. O.) (Appendix 1). Additional material from the following countries has been provided by co-workers (see sections Material studied): Egypt, Malawi, Sen- egal and South Africa. Most tapeworms were obtained by dissection of fresh fish. New material used in this study originates from the following localities. Congo River basin – Democratic Repub- lic of the Congo: Pioka (left bank) 4°54'23"S, 14°23'55"E. Gambia basin – Senegal: Niokolo-Koba National Park 13°1'44"N, 12°59'23"W. Nile River basin – Ethiopia: Beshelo River near Old Bridge, 11°28'27.78"N, 39°13'47.84"E, Lake Awasa 7°4'3.05"N, 38°26'30.86"E, Lake Chamo 5°51'10.03"N, 37°34'3.97"E, Lake Langano 7°35'26.47"N, 38°45'12.26"E, Lake Tana 12°2'29.64"N, 37°19'6.25"E, Lake Ziway 7°59'44.72"N, 38°49'42.51"E. The Sudan: Kostí, White Nile 13°10'20"N, 32°40'20"E, Sennar Dam, Blue Nile 13°32'37"N, 33°38'12"E, Khartoum 15°35'03"N, 32°32'13"E, Al Kawah 13°44'48.66"N, 32°29'49.56"E, Er Roseires Dam 11°47'5.83"N, 34°23'45.03"E, Lake Nubia (Asuan Dam) 21°46'41.65"N, 31°18'36.24"E. Turkana basin – Kenya: Lake Turkana, Todonyang, Omo River delta 4°27'10"N, 35°56'30"E. Zambezi basin – Malawi: southeast arm of Lake Malawi 14°9'41.00"S, 35°0'48.00"E. 2 · Zootaxa 3309 © 2012 Magnolia Press KUCHTA ET AL. The worms for morphological studies, including scanning electron microscopical (SEM) observations and histol- ogy, were fixed with hot 4% formaldehyde solution (see Oros et al. 2010), whereas some tapeworms or their pieces were preserved in 96% molecular grade ethanol for DNA sequencing (see below). Whole mounts were stained with μ Mayer’s hydrochloric carmine and mounted in Canada balsam. Cross sections of the strobila (thickness 15 m) were stained with hematoxylin-eosin, using standard histological methodology (Scholz & Hanzelová 1998). Several scoleces and segments were prepared for SEM following the procedure outlined by Kuchta & Caira (2010). Terminol- ogy of microtriches follows Chervy (2009); names of fish follow those of Froese & Pauly (2011). Illustrations were made using a drawing attachment on an Olympus BX51 microscope with differential interference contrast optics. Measurements are given in the text as ranges followed in parentheses by the mean, standard deviation, and number of measurements made. Measurements in descriptions are given in micrometers unless otherwise stated. Additional material of bothriocephalidean cestodes, which is listed in redescriptions of individual species, was provided by M. Barson (specimens from South Africa), R. A. Bray (Malawi), B. Koubková (Senegal) and E. N. Protasova (Ethiopia). In addition, type and voucher specimens were borrowed from several museum collections: The Natural History Museum, London, UK Č(acronym BěMNH); Göteborgs Naturhistoriska Museum, Göteborg, Sweden (GNM); Institute of Parasitology, eské Bud jovice, Czech Republic (IPCAS); Muséum d’Histoire Naturelle, Geneva, Switzerland (MHNG); Muséum National d’Histoire Naturelle, Paris, France (MNHNP); Meguro Parasitological Museum, Tokyo, Japan (MPM); Naturhistorisches Museum, Vienna, Austria (NMW); Musée Royal de l’Afrique Centrale, Tervuren, Belgium (RMCA); The Royal Veterinary College, London, UK (RVC); United States National Parasite Collection, Beltsville, Maryland, USA (USNPC); Naturhistorisches Museum, Berlin, Germany (ZMB). To assess phylogenetic relationships of African bothriocephalideans, a comparative analysis of partial sequences of the large subunit rDNA (lsrDNA) gene was performed. Seven samples were analysed: B. acheilognathi exLabeo- barbusnedgia Rüppell from Ethiopia (MHNG 55310; GenBank Accession No. JQ811839), B. acheilognathi exPoe- cilia reticulata Peters from Czech Republic (IPCAS C-15; JQ811834), I. ichthybori ex Ichthyborus besse (Joannis) from the Sudan (IPCAS C-455; JQ811837), K. gordoni ex Heterobranchus bidorsalis Geoffroy Saint-Hilaire from Ethiopia (IPCAS C-609; JQ811838), P. polypteri ex Polypterus senegalus Cuvier from the Sudan (IPCAS C-464; JQ811836), T. ciliotheca ex Clarias gariepinus (Burchell) from Ethiopia (IPCAS C-466; JQ811835) and Sengasp. ex Channa micropeltes (Cuvier) from Cambodia (IPCAS C-495; JQ811840). Genomic DNA was extracted using a standard phenol-chloroform extraction (Sambrook & Russell 2001) from 96% ethanol preserved samples. The D1–D3 region of lsrDNA was amplified by PCR using the primers and conditions described previously in Brabec et al. (2012). All products were verified on a 1% agarose gel and purified with the QIAquick PCR Purification Kit (Qia- gen). BigDye® Terminator v3.1 cycle sequencing reagents and a PRISM 3130xl automatic sequencer (Applied Bio- systems) were used for bidirectional sequencing of the PCR products using the set of PCR and internal sequencing primers described previously (e.g. Brabec et al. 2012). Sequences were assembled and inspected for errors using Geneious Pro 5.1.6 (Drummond et al. 2010) and aligned using the E-INS-i algorithm in the program MAFFT (Katoh et al. 2005). The resulting alignments were checked in MacClade 4.08 (Maddison & Maddison 2000) and ambigu- ously aligned positions were manually excluded prior to phylogenetic analyses. The phylogenetic relationships were evaluated by maximum likelihood (ML) and Bayesian inference (BI) methods. ML analyses were conducted using the program RAxML ver. 7.2.8-ALPHA (Stamatakis 2006; Stama- Γ takis et al. 2008), employing the GTR+ substitution model. All model parameters and bootstrap nodal support values (1000 repetitions) were estimated using RAxML. BI trees were constructed using MrBayes ver. 3.1.2 (Ron- quist & Huelsenbeck 2003), running 4 independent MCMCMC runs of 4 chains for 5 million generations and sam- Γ pling tree topologies every 1000th generation. As in ML analyses, the GTR+ model was employed. Burn-in period was set to 2.5 million generations according to the standard deviation of split frequencies values (<0.01) and by checking for MCMC convergence using AWTY (Nylander et al. 2008). Results The present study, based on the evaluation of all type material available, voucher specimens and extensive quantity of newly collected tapeworms, has shown that only seven species of bothriocephalidean cestodes, all of the family Bothriocephalidae, occur in freshwater fish in Africa. Adult bothriocephalideans were found in 6 families of fish, BOTHRIOCEPHALIDEAN TAPEWORMS FROM FESHWATER FISH Zootaxa 3309 © 2012 Magnolia Press · 3 namely Alestidae, Clariidae, Claroteidae, Cyprinidae, Distichodontidae and Polypteridae (Appendix 1). A compre- hensive survey of cestodes considered to be valid is provided below, including redescriptions of all but one (Both- riocephalus acheilognathi) species and data on their synonyms, definitive hosts and geographical distribution. A new genus is proposed to accommodate Polyonchobothrium gordoni Woodland, 1937, and a new tapeworm spe- cies is described from Bagrus meridionalis from Malawi. The species are listed in alphabetical order. Bothriocephalus acheilognathi Yamaguti, 1934 (Figs. 1, 2, 4, 12) Syns. (only taxa reported from Africa; for other synonyms – see Pool & Chubb 1985; Pool 1987; Kuchta & Scholz 2007): Both- riocephalus (Clestobothrium) kivuensis Baer & Fain, 1958; Bothriocephalus aegyptiacus Ryšavý & Moravec, 1975; Both- riocephalus barbus Fahmy, Mandour & El-Naffar, 1978; Bothriocephalus sp. of Yimer (2000), Al-Bassel (2003), and Zekarias & Yimer (2007). Type host:Acheilognathus rhombeus (Temminck & Schlegel) (Cypriniformes: Cyprinidae). Other definitive hosts in Africa (for extensive list of hosts from other continents, see Dove & Fletcher 2000): Barbus altianalis Boulenger, Barbus argenteus Günther, Barbus brevipinnis Jubb, Barbus bynni (Forsskål), Bar- bus mattozi Guimarães, Barbus paludinosus Peters, Barbustrimaculatus Peters, Cyprinus carpio Linnaeus, Labeo- barbus aeneus (Burchell), Labeobarbus kimberleyensis (Gilchrist & Thompson), Labeobarbus marequensis (Smith), Labeobarbus nedgia Rüppell (Cypriniformes: Cyprinidae). Type locality: Lake Ogura, Kyoto Prefecture, Honshu, Japan (35°2'N, 135°53'E). Distribution in Africa: Congo basin – Democratic Republic of the Congo; Incomati basin – South Africa; Limpopo basin – South Africa; Maputo basin – South Africa; Nile basin – Egypt, Ethiopia; Orange basin – South Africa. Prevalence and intensity of infection: Precise data are not available, but this cestode seems to be rather infre- quent. Despite extensive sampling, the present authors found only a single barbel infected with B. acheilognathi in Ethiopia (out of 56 barbels examined; see Appendix 1). However, Bertasso & Avenant-Oldewage (2005) found prevalence up to 90% in L. kimberleyensis from South Africa. Type material: MPM 23780 (holotype). Material studied:Type material: holotype (one slide of whole mount and one slide of histological sections) of B. acheilognathi; holotype (one worm on 20 slides) of B. aegyptiacus ex B. bynni from Cairo, Egypt (IPCAS C– 14); three syntypes (3 mounted scoleces and histological sections) of B. kivuensis ex B.altianalis from Lake Kivu, Democratic Republic of the Congo (MHNG 40332) (Fig. 12); vouchers: one specimen ex L. kimberleyensis from Vall Dam, South Africa, collected by M. Barson (MHNG 36429); one specimen of Bothriocephalus sp. ex Hydro- cynus sp. (Craciformes: Alestidae) from Bagata, Kwilu, Democratic Republic of the Congo (MHNG 55308); new material: three specimens ex L. nedgia from Beshelo River, near Old Bridge, Ethiopia, collected by Moges Beletew (MHNG 55310) (Figs. 1, 2, 4). Published records from Africa: Baer & Fain (1958, 1960); Ryšavý & Moravec (1975); Amin (1978); Fahmy et al. (1978); Boomker et al. (1980); Brandt et al. (1981); van As et al. (1981); El-Naffar et al. (1984); Basson & van As (1993); Paperna (1996); Schulz & Schoonbee (1999); Al-Bassel (2003); Bertasso & Avenant-Oldewage (2005); Retief et al. (2006, 2007, 2009); Zekarias & Yimer (2007); Degger & Avenant-Oldewage (2009); Degger et al. (2009); Stadtlander et al. (2011). Remarks: Up to date, six species of Bothriocephalus Rudolphi, 1808 have been reported to occur in freshwa- ter fish in Africa, especially in barbels (Cyprinidae: Barbinae) (Baer & Fain 1958; Tadros 1967; Ryšavý & Moravec 1975; Fahmy et al. 1978; Bertasso & Avenant-Oldewage 2005; Stadtlander et al. 2011). However, the present study has shown that apart from marginal occurrence of B. claviceps (see below), only one species, the Asian fish tapeworm, Bothriocephalus acheilognathi, actually parasitizes African fish. Bothriocephalus prudhoei Tadros, 1967 is invalidated (synonym of Kirstenella gordoni – see below) and three remaining species of Bothrio- cephalus, namely B.(Clestobothrium) kivuensis Baer & Fain, 1958; B. aegyptiacus Ryšavý & Moravec, 1975; and B. barbus Fahmy, Manour & El-Naffar, 1978, are synonyms of B. acheilognathi (see Pool 1987; Kuchta & Scholz 2007). 4 · Zootaxa 3309 © 2012 Magnolia Press KUCHTA ET AL. FIGURES 1–11. Scanning electron micrographs of the scoleces and the body surface.1, 2,4,Bothriocephalus acheilognathi Yamaguti, 1934 ex Labeobarbus nedgia from Ethiopia (MHNG 55310). 1,2, Scolex, dorsoventral and sublateral views. 4, Sur- face of apical region of scolex. 3,6,9,Ichthybothrium ichthybori Khalil, 1971 ex Ichthyborus besse from Sudan (IPCAS C- 455). 3, Scolex, dorsoventral view. 6, Surface of apical region of scolex. 9, Detail of operculum of egg. 5,7, 8,10,11,Kirst- enella gordoni (Woodland, 1937) ex Heterobranchus bidorsalis from Kenya (IPCAS C-609). 5, Surface of apical region of scolex. 7, Scolex, dorsoventral view. 8, Scolex, apical view. 10, Mature segment, ventral view. 11, Detail of operculum of egg. BOTHRIOCEPHALIDEAN TAPEWORMS FROM FESHWATER FISH Zootaxa 3309 © 2012 Magnolia Press · 5 Bothriocephalus acheilognathi has been introduced from its original distribution area in East Asia throughout the world (Scholz et al. 2011b). It is a pathogen of fry of cultured fish, especially carp and other cyprinids, and has been reported to cause mortalities (Williams & Jones 1994). Numerous bothriocephalidean tapeworms have been synonymized with B. acheilognathi, including African taxa (see Kuchta & Scholz 2007 for list of synonyms). Pool (1987) synonymized B. aegyptiacus and B. kivuensis with B. acheilognathi, whereas B. barbus was considered to be a synonym of B. acheilognathi byKuchta & Scholz (2007). Molecular data confirm these synonymies (Fig. 62). The host spectrum of B. acheilognathi is extraordinarily wide and includes more than 200 species of unrelated fish (Scholz et al. 2011b), with cyprinids representing the most suitable definitive hosts. In Africa, it has been recorded in 12 cyprinid species of two native genera, Barbus (7 spp.) and Labeobarbus (4 spp.), as well as in intro- duced Cyprinus caprio (Stadtlander et al. 2011). Records from the Nile basin, South Africa, as well as the Congo basin suggest that B. acheilognathi is present throughout the continent wherever suitable cyprinid hosts are availa- ble. Surprisingly, there are no records from Labeo spp., suggesting that members of this speciose pan-African cyprinid genus constituting a significant component of ichthyofauna of many African basins are not suitable hosts. Such presumed unsuitability of Labeo spp. as hosts for B. acheilognathi might be the reason of its apparent absence in basins in which cyprinids are dominated by Labeo spp. This assumption is supported by the absence of B. achei- lognathi in Lake Turkana (this study, Appendix 1), where none of the examined cyprinids, i.e. 42 Labeo spp. and 11 Barbus bynni, was infected (ratio of examined fish does not reflect real ratio in the lake in which two Barbus spp. in very low densities share the lake with huge populations of two Labeo spp.). Findings of B. acheilognathi in clariid catfish in Ethiopia, Nigeria and Zimbabwe (Anosike et al. 1992; Yimer 2000; Moyo et al. 2009; Bichi & Yelwa 2010) may represent accidental findings due to predation, but extraordi- narily high values of prevalence (up to 60%; Moyo et al. 2009) indicate that Clarias Scopoli catfish may harbour this parasite more frequently. Unfortunately, no voucher specimens of these remarkable, but suspicious findings have been preserved, which casts doubts upon correct identification of the worms found. It thus cannot be excluded that the authors misidentified Tetracampos ciliotheca Wedl, 1861, which is a very frequent and abundant intestinal parasite of Clarias catfish in Africa (see below) and the scolex of which somewhat resembles that of B. acheilog- nathi, especially when hooklets on the apical disc are detached after death of worms. Tapeworms from Tilapia sp. (Characiformes: Cichlidae) and Hydrocynus sp. identified as Bothriocephalus sp. (Khalil & Thurston 1973; present study), may represent incidental infection of B. acheilognathi, but species identi- fication could not be confirmed, because the specimen from Tilapia sp. is not available and the worm from Hydro- cynus sp. is without scolex. Bothriocephalus acheilognathi is distributed throughout Africa, but its abundance is considerably lower than in newly colonized regions, especially in Europe and North America (García-Prieto & Osorio-Sarabia 1991; Wil- liams & Jones 1994; Salgado-Maldonado & Pineda-López 2003). The absence of records from West Africa is prob- ably artificial, because great majority of parasitological studies on cyprinids in this region were obviously focused solely on ectoparasitic monogeneans (see Khalil & Polling 1997). Ichthybothrium ichthybori Khalil, 1971 (Figs. 3, 6, 9, 13, 14) Syns: “immature Bothriocephalus” of Woodland (1936); “unidentified ptychobothriid cestode” of Khalil (1969, 1971a). Type host:Ichthyborus besse (Joannis)(Characiformes: Ichthyboridae). Other definitive host: Ichthyborus quadrilineatus (Pellegrin) (new host). Type locality:White Nile near Khartoum, the Sudan. Distribution: Upper Guinea – Sierra Leone (Moa River in Mano or Pujehun) ; Nile basin – the Sudan (Khar- toum and Kostí). Prevalence and intensity of infection: Nile, Kostí, the Sudan; 67%, n = 6, intensity 15–35 (Khalil 1971b); 50%, n = 4, intensity 1–8 (present study). Type material: Holotype – incomplete worm on 2 slides (USNPC 71666) and 2 paratypes – 3 worms on 3 slides (USNPC 71667). 6 · Zootaxa 3309 © 2012 Magnolia Press KUCHTA ET AL. Material studied: Type material(USNPC 71666, 71667); vouchers: several slides collected by W.N.F. Wood- land from I. quadrilineatus in Sierra Leone (BMNH 1977.3.28.147–179); new material: 9 specimens from Kostí, the Sudan (IPCAS C-455). Published records: Woodland (1936); Khalil (1969, 1971a, b); Kuchta et al. (2008b). Re-description (based on 3 mature and 6 immature worms from Kostí; measurements of holotype in brack- ets): Bothriocephalidea, Bothriocephalidae. Strobila up to 10.0 [4.2] cm long; maximum width 3.7 [1.5] mm. External and internal segmentation weakly developed; segments variable in shape from wider than long to longer than wide, acraspedote (Fig. 13). Two pairs of longitudinal osmoregulatory canals; dorsal canals narrow (diameter up to 10); ventral canals wide (diameter up to 15), connected by transverse anastomoses. Inner longitudinal muscu- lature weakly developed, formed by few muscle fibres. Surface of strobila uniformly covered with capilliform fil- itriches. Scolex oval to lanceolate, much narrower than anterior part of strobila (Figs. 3, 13), 490–630 [495] long by 340–530 wide [413]. Apical disc absent. Bothria shallow, elongate, 240–360 [360] long by 140–200 [186] wide (Fig. 3). Surface of scolex covered with capilliform filitriches; tumuliform globular structures not observed (Fig. 6). Neck absent, first segments appear immediately posterior to scolex (Fig. 13). Strobila with immature segments very long, representing up to 93% [58%] of total length. Mature segments, i.e. with spermatozoa in vas deferens, of two markedly different forms (morphotypes): longer than wide (486–614 by 374–529) [749–1,187 by 564–700] with one genital complex per segment or much wider than long (315–557 by 3,370–3,745) with two or three genital complexes per segment; segment length/width ratio 1.08–1.09 [1.19–1.78] or 0.08–0.16 (Figs. 13, 14). Gravid segments longer than wide (736–831 by 374–523) [1,080–1,280] or wider than long (425–621 by 2,550–3,360); segment length/width ratio 1.59–1.96 [1.50–1.74] or 0.13–0.24. FIGURES 12–14. Line drawings. 12,Detail of genital complex of gravid segment, dorsal view, eggs not illustrated, type spec- imen of Bothriocephalus kivuensis Baer & Fain, 1958 (= B. acheilognathi Yamaguti, 1934) ex Barbusaltianalis from Demo- cratic Republic of the Congo (MHNG 40332). 13,14,Ichthybothrium ichthybori Khalil, 1971 ex Ichthyborus besse from the Sudan (IPCAS C-455). 13, Complete worm with single genital complex per segment. 14, Mature segment with double genital complexes. BOTHRIOCEPHALIDEAN TAPEWORMS FROM FESHWATER FISH Zootaxa 3309 © 2012 Magnolia Press · 7 Testes medullary, oval, 37–69 [48–57] in number per proglottid (110–120 testes in segments with duplicate or triplicate proglottides), 36–50 long by 22–43 wide [77–96 by 54–69], forming 2 narrow longitudinal bands (20–36 [22–29] testes per band), confluent between segments, absent medially and near lateral margins (Fig. 14). Cirrus- sac large, thin-walled (thickness of sac wall up to 32 [35]), spherical, 107–143 in diameter [106–135]; equatorial to pre-equatorial (at 42–50% [41–48%] of length of mature segment from anterior margin). Vas deferens forms numerous loops anterolateral to cirrus-sac, cirrus unarmed, opening into genital atrium (Fig. 14). Genital pore dor- sal, sub-median, pre-equatorial (at 8–12% [4–10%] of length of mature segment from anterior margin), enlarging in gravid segments (Fig. 14). Ovary asymmetrical, elongated or V-shaped, lobulate, 41–143 [131–213] long by 135–394 [186–251] wide (Fig. 14). Vagina a straight, thin-walled tube, with sac-like extension in proximal part, 64–83 [44–60] long by 23– 52 [26–27] wide, opens posterior to cirrus-sac into genital atrium; vaginal sphincter absent (Fig. 14). Vitelline fol- licles numerous, small, spherical, 10–25 [30–45] long by 17–35 [25–30] wide, cortical, form 2 wide longitudinal bands confluent between segments, separated medially, rarely connected by several follicles in postovarian region (Fig. 14). Uterine duct forms numerous tightly coiled loops, filled with eggs, enlarged in gravid segments. Uterine pore thick-walled, opens slightly posterior to midlength of uterus. Eggs oval, thin-walled, operculate, unembryonated, 38–46 [45–50] long by 27–30 [32–33] wide (Fig. 9). Remarks: This tapeworm was described by Khalil (1971b) from the Sudan and placed in a newly proposed genus Ichthybothrium Khalil, 1971. It is characterized by a lanceolate scolex, which is narrower than the anterior end of the strobila (Figs. 3, 13) and possesses shallow bothria, but no apical disc. The strobila has weakly devel- oped segmentation and its large proportion (more than 90% in some specimens) consists of immature proglottides without genital complexes (called “pregenital region” by Khalil 1971b). Specimens recently collected near the type locality differ markedly from the type specimens in the shape of the segments and number of genital complexes per segment, with duplication or even triplication of genital complexes (proglottides) within mature and gravid segments (Kuchta et al. 2008b). Marked differences in the size of the worms may be influenced by crowding effect, because worms in more heavily infected hosts (intensity of infection 8–35 worms) are smaller (maximum length 6.2 cm) and consist of just a few mature and gravid segments, which are longer than wide (length/width ratio 1.08–1.96: 1) and usually contain only one genital complex (Khalil 1971b) (Fig. 13). In contrast, tapeworms from mono- or light infections are up to 10 cm long, their strobila is formed by numerous mature and gravid segments, which are much wider than long (length/width ratio 0.08–0.24: 1), and each of them contains two or even three genital complexes (Kuchta et al. 2008b; present study) (Fig. 14). Khalil (1971b) correctly reported the cortical position of vitelline follicles in the diagnosis of I. ichthybori, but erroneously described vitelline follicles to be medullary in the diagnosis of the genus. The surface of the scolex and strobila was studied using SEM for the first time and a uniform pattern of capilliform filitriches was observed. Khalil (1971b) also reported the presence of “minute hair-like processes” on the scolex and “long hair-like pro- cesses” around the uterine pore, which may also represent capilliform filitriches (Fig. 6). However, tumuliform globular structures commonly observed in many bothriocephalideans (Kuchta et al. 2008b), including all other studied species, are missing in I. ichthybori (Fig. 6). Ichthybothrium ichthybori is a rather frequent parasite of Ichthyborus besse, with high prevalence of infection (50–67%) and a relatively high intensity of infection. However, its fish host is rare and thus only a limited number of records of this tapeworm exist in the literature. Examination of several slides with immature cestodes designated as “immature Bothriocephalus” from Neoborusquadrilineatus (= Ichthyborus quadrilineatus) collected by W.N.F. Woodland in Sierra Leone (BMNH 1977.3.28.147–179) (Woodland 1936) has shown that they belong to I. ichthy- bori. Kirstenella Kuchta n. gen. Diagnosis: Bothriocephalidea, Bothriocephalidae. Medium-sized worms. Segmentation present. Segments trapezi- form, craspedote. Scolex elongate. Bothria elongate, shallow. Apical disc present, as wide as, or slightly wider, than bothrial part of scolex, armed with large, simple hooks arranged in two lateral semicircles. Neck absent. Testes in two lateral fields, continuous between segments. Cirrus-sac thick-walled, spherical, large, representing as much 8 · Zootaxa 3309 © 2012 Magnolia Press KUCHTA ET AL. as 1/6–2/5 of segment width; internal seminal vesicle present; cirrus unarmed. Genital pore dorsal, median. Ovary compact, bilobed, transversely elongate. Vagina posterior to cirrus-sac, enlarged distally. Vitelline follicles cortical, in two lateral fields continuous between segments. Uterine duct sinuous, S-shaped, enlarging in gravid segments. Uterus small, oval. Uterine pore median. Eggs operculate, unembryonated. In freshwater siluriforms. Africa. Etymology: The new genus is named in honour of Kirsten Jensen, University of Kansas at Lawrence, USA, for her outstanding contribution to cestode systematics. Type and only species: Kirstenella gordoni (Woodland, 1937) Kuchta, n. comb. Remarks: The new genus is placed in Bothriocephalidae Lönnberg, 1889 because it possesses the median gen- ital pore, follicular vitellaria, and a ventral uterine pore (Kuchta et al. 2008a, b). It is typified by combination of the following characteristics: (1) scolex is elongate with a small apical disc armed with two lateral semicircles of hooks; (2) cirrus-sac is large (its width represents as much as 1/6–2/5 of the width of segments) and it is situated perpendicularly; (3) a low number of testes (less than 75 per segment); (4) vagina is enlarged distally; (5) vitellaria are cortical. Kirstenella differs from all but three bothriocephalid genera (Polyonchobothrium Diesing, 1854, Senga Doll- fus, 1934 and Tetracampos Wedl, 1861) in the presence ofan apical disc armed with two lateral semicircles of large hooks (Kuchta et al. 2008b). The new genus can be distinguished from Tetracampos by the presence of cortical (instead of medullary) vitelline follicles and presence of operculate, thin-walled (vs. unoperculated, with hyaline membrane) eggs. The new genus can be distinguished from Polyonchobothrium and Senga mainly by the relative size of the cirrus-sac, which is markedly larger (ratio of its width to the width of segments 16–39%) than that of species of Senga (ratio 4–10%) and Polyonchobothrium (ratio 5–10%), and is situated more perpendicularly in Kirstenella, whereas longitudinally or obliquely in the two remaining genera (Figs. 19, 22, 38, 39). Moreover, μ Polyonchobothrium has a markedly wider apical disc (350–510 μm) and much larger hooks (up to 190 m) mostly μ distributed in 4 quadrants, whereas Kirstenella has a slender apical disc (230–300 m) with smaller apical hooks μ (shorter than 90 m) distributed in 2 semicircles. Hooks of Senga have an indistinct blade and handle (data not shown), whereas those of Kirstenella have the blade well differentiated from the handle (Figs. 16, 18). Kirstenella gordoni (Woodland, 1937) Kuchta, n. comb. (Figs. 7, 8, 10, 11, 15–22) Syns: Polyonchobothrium gordoni Woodland, 1937; Polyoncobothrium gordoni (Woodland, 1937) Yamaguti, 1959; Bothrio- cephalus prudhoei Tadros, 1967; Polyonchobothrium sp. of Fagbenro et al. (1993); Senga gordoni (Woodland, 1937) Kuchta & Scholz, 2007. Type host:Heterobranchus bidorsalis Geoffroy Saint-Hilaire (Siluriformes: Clariidae). Other definitive hosts: Clarias anguillaris (Linnaeus) (Siluriformes: Clariidae), Schilbe mystus (Linnaeus) (Siluriformes: Schilbeidae). Both species are considered accidental/atypical hosts – see the Remarks section. Type locality: Waanje River near Pujehun, Sierra Leone. Distribution: Gambia basin – Senegal (Lampsar River, Taoué River, Gambia River); Turkana basin – Ethiopia (lower Omo River) and Kenya (Lake Turkana – restricted to the northernmost freshwater region part of the lake; see Remarks); Upper Guinea – Sierra Leone (Moa River); Niger basin – Mali (Lake Debo), Nigeria (River Ogbase); Nile basin – the Sudan (White Nile, Kostí) and Uganda (Lake Victoria). Prevalence and intensity of infection: Lake Turkana – Omo River delta, Todonyang, Kenya, 60%, n = 10; intensity 1–6 (present study); Ethiopia – Omo River, Omorate, 50%, n = 2, intensity 1 (present study); Nile basin – the Sudan, 20%, n = 5, intensity 1 (present study); Niger basin – Nigeria, 22%, n = 185 (Fagbenro et al. 1993). Type material: Syntypes (several slides with fragments of worms and their histological sections) (BMNH 1965.2.24.36–45). Material studied: Type material: several syntypes; holotype and paratypes of Polyonchobothrium gordoni from Waanje River, Sierra Leone (BMNH 1965.2.24.36–45); Bothriocephalus prudhoei ex Clarias anguillaris, Malakal, the Sudan (BMNH 1998.10.15.6.7; RVC C. 1262); vouchers: Ptychobothriidae gen. sp. ex H. bidorsalis, Mali, Lake Dabo (MNHNP bd10); P. clarias ex H. bidorsalis, Taoué River, Senegal (RMCA 34695; D.T.F. Puy- laert; 15.iii.1966); Polyonchobothrium sp. ex H. bidorsalis,Waanje River, Sierra Leone(BMNH 1965.2.24.62–65; collected by W.N.F. Woodland); Polyonchobothrium sp. ex Schilbe mystus, Lake Victoria, Uganda (BMNH BOTHRIOCEPHALIDEAN TAPEWORMS FROM FESHWATER FISH Zootaxa 3309 © 2012 Magnolia Press · 9 1957.12.30.34–38); new material: 1 specimen ex H. bidorsalis, White Nile in Kostí, the Sudan (MHNG 49379; 26.iii.2006); 1 specimen from Omo River, Omorate, Ethiopia (MHNG 63067; 15.iv.2006); 15 specimens from Omo River delta, Todonyang, Lake Turkana, Kenya (MHNG 69955); 3 specimens from Niokolo-Koba National Park, River Gambia, Senegal collected by B. Koubková (2006; Sen 272-1). The new material is deposited in BMNH (Nos. 2012.3.20.26–28), IPCAS (No. C-609), MHNG (Nos. 55339, 63254, 82036, 82038) USNPC (No. 105390) and ZMB (No. 7524). Published records: Woodland (1937); Tadros (1968); Khalil (1973); Fagbenro et al. (1993). Re-description (based on 10 whole mounts and 1 scolex observed by SEM; measurements from Woodland, 1937 in brackets): Bothriocephalidea, Bothriocephalidae. Strobila up to 10 [1.0–1.5] cm long; maximum width 875 [1,350]. External and internal segmentation present; segments wider than long, markedly craspedote, several ridges on surface of segments (Figs. 10, 15). Two pairs of longitudinal osmoregulatory canals; dorsal canals narrow (diameter up to 7); ventral canals wide (diameter up to 14), connected by transverse anastomoses. Inner longitudinal musculature well developed, muscle fibres diffused (Fig. 21). Surface of strobila covered with capilliform filitriches. Scolex oval, narrow (Figs. 7, 15, 17), 926–1,480 (1,160 ± 168) [280–380] long by 224–419 (317 ± 65) [310– 516] wide (n = 15). Apical disc weakly developed, 230–300 (265 ± 18) wide by 74–102 (87 ± 10) long, armed with 40–42 (41 ± 1) [36–42] small hooks 16–88 (67 ± 19; n = 103) long. Hooks variable in size, arranged in two semi- circles, smallest on periphery above each bothrium and increasing into middle of semicircle with largest hook 83– 88 (86 ± 3; 3) [64] long (Figs. 8, 16, 18). Bothria elongate, shallow, 757–1,355 (1,025 ± 230) long by 73–98 (84 ± 11) wide (n = 7) (Figs. 7, 17). Surface of scolex covered with capilliform filitriches and tumuliform globular struc- tures (diameter around 1) (Fig. 5). Neck absent, first segments appearing immediately posterior to scolex (Figs. 7, 15, 17). Immature segments 112–257 (191 ± 43) long by 217–531 (340 ± 101) wide; segment length/width ratio 0.27– 0.94: 1 (n = 15) (Fig. 15). Mature segments wider than long, 134–371 (242 ± 74) long by 283–998 (667 ± 201) wide; segment length/width ratio 0.23–0.66: 1 (n = 24) (Fig. 22). Gravid segments wider than long, 188–573 (326 ± 97) long by 704–1,104 (869 ± 114) wide; segment length/width ratio 0.24–0.79: 1 (n = 20) (Fig. 15). Testes medullary, oval, 70–133 (86 ± 20; 10) [under 30] in number per segment, 23–51 (35 ± 9) long by 26–45 (36 ± 6) wide [69 × 44 in sections] (n = 10) in diameter, forming 2 narrow longitudinal bands, 34–72 (47 ± 11) tes- tes per band, confluent between segments, absent medially and near lateral margins (Fig. 22). Cirrus-sac large, thick-walled (thickness of sac wall 7–14), spherical, 92–161 (121 ± 21) long by 91–171 (132 ± 20) wide, length/ width ratio 0.66–1.15: 1 (n = 24), pre-equatorial to postequatorial (at 30–69% of length of mature segment from anterior margin; n = 10). Internal seminal vesicle present, Vas deferens forms numerous loops posterolateral to cirrus-sac; internal sperm ducts strongly coiled, cirrus unarmed, opening into genital atrium (Figs. 19, 21, 22). Gen- ital pore dorsal, median, pre-equatorial (Fig. 22). Ovary asymmetrical, compact, bilobed, 44–95 (70 ± 16) long by 224–455 (332 ± 7) wide (n = 16) (Figs. 19, 22). Vagina a straight, thin-walled tube, 22–52 in diameter, opens posterior to cirrus-sac into genital atrium; vaginal sphincter absent (Fig. 19). Vitelline follicles numerous, small, spherical, 11–17 (14 ± 2; 8) [51 × 22] in diameter, cortical, form 2 wide longitudinal bands confluent between segments, separated medially, rarely connected by sev- eral follicles in postovarian region (Fig. 22). Uterine duct forms numerous tightly coiled loops, filled with eggs, enlarged in gravid segments (Fig. 15). Uterus thick-walled, median, spherical, enlarged in gravid segments, occupies 4–30% of segment surface (Fig. 15). Uterine pore thick-walled, opens slightly posterior to midlength of uterus (Figs. 19, 22). Eggs oval, thin-walled, operculate, unembryonated, with abopercular knob, 30–43 (38 ± 4) lo ng by 21–33 (28 ± 4) wide [47 × 31] (n = 11) (Figs. 11, 20). Remarks:Polyonchobothrium gordoni Woodland, 1937 was described from Heterobranchus bidorsalis from Sierra Leone. Woodland (1937) mentioned similarity of his new species with Tetracampos ciliotheca Wedl, 1861 in the shape of the body (“great part of strobila tightly coiled”) and differentiated both taxa by the number and size of apical hooks. Tadros (1967) described a new species Bothriocephalus prudhoei Tadros, 1967 based on material collected by K.N. Soliman from Clarias anguillaris in the Sudan in 1958. Only one complete specimen with several pieces of strobila was available, but the apical part of its scolex is missing. Kuchta and Scholz (2007) considered B. prudhoei to be a junior synonym of T. ciliotheca, but re-examination of the type material did not support this synonymy. In fact, strobilar morphology, including the presence of a large cirrus-sac, is identical to that of K. gordoni. 10 · Zootaxa 3309 © 2012 Magnolia Press KUCHTA ET AL.