Version Date: 31 July 2014 Contributors to this version: Ricardo Betancur-R, Ed Wiley, Nicolas Bailly, Masaki Miya, Guillaume Lecointre, Guillermo Ortí Citing this classification: Betancur-R, R., E. Wiley, N. Bailly, M. Miya, G. Lecointre, and G. Ortí. 2014. Phylogenetic Classification of Bony Fishes --Version 3 (http://www.deepfin.org/Classification_v3.htm). Betancur-R., R., R.E. Broughton, E.O. Wiley, K. Carpenter, J.A. Lopez, C. Li, N.I. Holcroft, D. Arcila, M. Sanciangco, J. Cureton, F. Zhang, T. Buser, M. Campbell, T. Rowley, J.A. Ballesteros, G. Lu, T. Grande, G. Arratia & G. Ortí. 2013. The tree of life and a new classification of bony fishes. PLoS Currents Tree of Life. 2013 Apr 18. Download useful resources: Go to the bottom of this page to download: Spreadsheet with complete classification scheme; molecular phylogeny used as basis for classification (from version 2, needs to be updated to reflect the current classification scheme); and figures with summary phylogenetic hypothesis displaying all major groups. This classification is an update of version 1 (18 April 2013) and version 2 (27 Nov 2013), originally published by Betancur-R. et al. (2013a). This version fixes involuntary errors and omissions (thanks to numerous colleagues who provided feedback!) and incorporates new findings from the recent literature. It is based on the same phylogenetic results used for version 2, from analysis of a molecular data set with 1591 taxa (the data set combines evidence published by Betancur-R. et al. (2013a) and Near et al. (2013)). Families in the classification are linked to FishBase family pages (Froese and Pauly, 2013) except for cases where there are discrepancies in the recognition of valid families, noted and justified in each case in the classification. Version 3 presents new names for series included in subdivision Percomorphaceae, to avoid confusion between this taxon and series Percomorpharia, and hence disambiguate the meaning of "percomorph". Percomorpharia is now referred to as series Eupercaria, and the other series names also have simpler endings (dropping the term "morph" but keeping a consistent ending with "aria"). Under this scheme, the word "percomorph" refers unambiguously to subdivision Percomorphaceae, retaining its original meaning as implied by Wiley and Johnson (2010), following Johnson and Patterson (1993). The common name "eupercarians" now is available for taxa in the clade containing Perciformes, Labriformes, Tetraodontiformes and others (i.e., the "new bush at the top" sensu Betancur-R. et al. (2013a)). Figure 1 below shows the new names. A complete list of changes from version 2 can be found here. General comments on the classification are here.  Figure 1. Phylogeny of the nine major percomorph clades included in Subdivision Percomorphaceae. Each major clade is recognized as a series. Classification for Bony Fishes (version 3)
Megaclass Osteichthyes (=Euteleostomi, =Euosteichthyes) Superclass Actinopterygii (100%) Class Cladistia (100%) Order Polypteriformes Class Actinopteri (100%) Subclass Chondrostei, (100%) Order Acipenseriformes Subclass Neopterygii (100%) Infraclass Holostei (100%) Order Amiiformes Order Lepisosteiformes (100%) Infraclass Teleostei (100%) Megacohort Elopocephalai sensu Arratia (1999) (100%) Supercohort Elopocephala (100%) Cohort Elopomorpha (100%) Order Elopiformes (100%) Order Albuliformes (100%) Order Notacanthiformes (100%) Order Anguilliformes (100%) Not examined: Chlopsidae, Colocongridae, Cyematidae, Derichthyidae, Heterenchelyidae, Monognathidae, Moringuidae, Myrocongridae, Nettastomatidae, Protanguillidae, Synaphobranchidae. Comment: Suborders recognized in Wiley and Johnson (2010) based on previous work cited therein are significantly incongruent with the clades obtained in this analysis; thus, no subordinal classification is proposed. Megacohort Osteoglossocephalai sensu Arratia (1999) (38%) Supercohort Osteoglossocephala sensu Arratia (1999) (99%) Cohort Osteoglossomorpha Order Hiodontiformes (100%) Order Osteoglossiformes (100%) Osteoglossidae (includes Arapaima and Heterotis, formerly in Arapaimidae) Supercohort Clupeocephala sensu Arratia (2010) (100%) Cohort Otomorpha (= Otocephala, Ostarioclupeomorpha) (100%) Subcohort Clupei (87%) Order Clupeiformes Suborder Denticipitoidei Suborder Clupeoidei (100%) Clupeidae (not monophyletic; includes Sundasalanx according to Lavoué et al. (2013), formerly in Sundalangidae) Not examined: Dussumieriidae. Comment: family-level groupings may require major revision; Pristigasteridae, Chirocentridae and Engraulidae are supported by other molecular studies, but not Clupeidae (Li and Ortí, 2007; Lavoué et al., 2013); five well-supported lineages identified by Lavoué et al. (2013) could become new families. Subcohort Alepocephali (100%) Order Alepocephaliformes Alepocephalidae (not monophyletic) Not examined: Leptochilichthyidae. Comment: the position of alepocephaloids as the sister group to Ostariophyisi also was reported by Lavoué et al. (2008) and by Paulsen et al. (2009). Subcohort Ostariophysi (100%) Section Anotophysa (97%) Order Gonorynchiformes Suborder Gonorynchoidei Suborder Chanoidei Suborder Knerioidei (100%) Kneriidae (includes the former Phractolaemidae, now a subfamily in Kneriidae, following Davis et al. (2013)) Section Otophysa (100%) Superorder Cypriniphysae (100%) Order Cypriniformes Botiidae (following Chen et al. (2009)) Not examined: Balitoridae, Barbuccidae, Ellopostomatidae, Psilorhynchidae, Serpenticobitidae, Vaillantellidae. Superorder Characiphysae (100%) Order Gymnotiformes (100%) Suborder Gymnotoidei (not monophyletic) Gymnotidae (not monophyletic here, but see Alves-Gomes (2010) and Nakatani et al. (2011)) Suborder Sternopygoidei (not monophyletic) Sternopygidae (not monophyletic) Not examined: Hypopomidae. Comment: Although not monophyletic here, the monophyly of gymnotiform suborders is corroborated by Albert and Crampton (2005). Order Characiformes (100%) Suborder Citharinoidei (not monophyletic) Suborder Characoidei (not monophyletic) Not examined: Anostomidae, Curimatidae, Iguanodectidae. Comment: Although not monophyletic in this analysis, the monophyly of characiform suborders has been corroborated by other molecular studies (Calcagnotto et al., 2005). In other molecular studies, the monophyly of Characiformes was not obtained, since Characoidei was more closely related to Siluriformes than to Citharinoidei (Nakatani et al., 2011; Chen et al., 2014a). Order Siluriformes (100%) Suborder Loricarioidei (93%) Not examined: Scoloplacidae Suborder Diplomystoidei Suborder Siluroidei (100%) Bagridae (includes taxa formerly in Olyridae, following Sullivan et al. (2006)) Horabagridae (following Sullivan et al. (2006)) Not examined: Aspredinidae, Austroglanididae, Erethistidae, Lacantuniidae. Supercohort Clupeocephala (cont.) Cohort Euteleosteomorpha (100%) Subcohort Lepidogalaxii Order Lepidogalaxiiformes Subcohort Protacanthopterygii sedis mutabilis (37%) Order Galaxiiformes (100%) Order Argentiniformes (100%) Order Salmoniformes (100%) Order Esociformes (100%) Subcohort Stomiati (73%) Order Stomiatiformes (=Stomiiformes) (100%) Diplophidae (following Nelson (2006); apparently ommited or considerd a junior synonym of Gonostomidae by Eschmeyer and Fong (2014)) Phosichthyidae (not monophyletic) Order Osmeriformes (100%) Subcohort Neoteleostei (100%) Infracohort Ateleopodia (100%) Order Ateleopodiformes Infracohort Eurypterygia (96%) Section Aulopa (100%) Order Aulopiformes Suborder Aulopoidei (not monophyletic) Synodontidae (not monophyletic) Suborder Paraulopoidei Suborder Alepisauroidei (not monophyletic) Alepisauridae (includes taxa previously listed in Omosudidae and Anotopteridae, following Davis (2010)) Chlorophthalmidae (not monophyletic) Ipnopidae (not monophyletic) Paralepididae (not monophyletic) Scopelarchidae (not monophyletic) Sudidae (following Davis (2010); synonym of Paralepididae according to Eschmeyer (2014)) Not examined: Bathysauroididae, Bathysauropsidae sensu Davis (2010). Comment: Anotopteridae is not recognized; aulopiform families listed follow Davis (2010). Although not monophyletic herein, the monophyly of aulopiform suborders listed is supported by Davis (2010). Section Ctenosquamata (97%) Subsection Myctophata (100%) Order Myctophiformes Subsection Acanthomorphata (97%) Division Lampridacea (100%) Order Lampridiformes Not examined: Radiicephalidae, Veliferidae. Division Paracanthomorphacea sensu Grande et al. (2013) (93%) Series Percopsaria (100%) Order Percopsiformes Series Zeiogadaria (= Zeiogadiformes sensu Li et al. (2009)) (98%) Subseries Zeariae (100%) Order Zeiformes Not examined: Cyttidae, Grammicolepididae, Oreosomatidae. Subseries Gadariae (100%) Order Stylephoriformes (sensu Miya et al. (2007)) Order Gadiformes (100%) Suborder Macrouroidei Macrouridae (not monophyletic here, but see Roa-Varon and Ortí (2009)) Steindachneriidae (following Roa-Varon and Ortí (2009); synonym of Merluciidae according to Eschmeyer (2014)) Suborder Gadoidei (not monophyletic) Gaidropsaridae (formerly a subfamily of Lotidae, raised to family level) Not examined: Bregmacerotidae, Euclichthyidae, Melanonidae. Suborder Muraenolepidoidei Comment: The subordinal classification follows Roa-Varón and Ortí (2009): fig. 6. Division Polymixiacea (100%) Order Polymixiiformes Division Euacanthomorphacea sensu Johnson and Patterson (1993) (99%) Subdivision Berycimorphaceae (87%) Order Beryciformes (similar to Trachichthyiformes sensu Moore (1993)) Trachichthyidae (not monophyletic) Not examined: Gibberichthyidae, Hispidoberycidae (expected affinity following Moore (1993)). Subdivision Holocentrimorphaceae (100%) Order Holocentriformes Comment: Moore (1993) and Stiassny and Moore (1992) provide morphological evidence supporting a sister-group relationship between holocentrids and percomorphs, which further guarantees placement of this family in its own order. Subdivision Percomorphaceae (="Percomorpha" sensu Miya et al. (2003) and Miya et al. (2005)). Subdivision Percomorphaceae (cont.) Series Ophidiaria (100%) Order Ophidiiformes Suborder Ophidioidei Ophidiidae (not monophyletic as traditionally recognized, but monophyletic if Carapidae is included) Suborder Bythitoidei Bythitidae (not monophyletic) Not examined: Parabrotulidae Series Batrachoidaria (100%) Order Batrachoidiformes Series Gobiaria (see also Thacker (2009) and Chakrabarty et al. (2012)) (100%) Order Kurtiformes (98%) Suborder Kurtoidei Suborder Apogonoidei Comment: Johnson (1993) noted that the configuration of dorsal gill-arch elements may be homologous in Kurtus and apogonids. Order Gobiiformes (100%) Suborder Odontobutoidei (100%) Suborder Eleotroidei (97%) Suborder Gobioidei (100%) Gobiidae (includes Microdesmus, previously listed in Microdesmidae; see Thacker (2009)) Not examined: Kraemeriidae, Rhyacichthyidae, Schindleriidae, Thalasseleotrididae, Xenisthmidae. Comment: In addition to the well-supported molecular circumscription of Gobiaria, kurtids, apogonids and gobioids are characterized by the presence of sensory papillae rows on the head and body (Thacker, 2009). Series Syngnatharia (96%) Order Syngnathiformes Suborder-level incertae sedis in Syngnathiformes Suborder Syngnathoidei (92%) Not examined: Solenostomidae (assumed affinity with Syngnathidae) Suborder Dactylopteroidei (100%) Suborder Callionymoidei (100%) Not examined: Draconettidae (assumed affinity with Callionymidae). Suborder Mulloidei (92%) Series Pelagiaria (= Stromateoidei sensu Li et al. (2009); = Pelagia sensu Miya et al. (2013)) (100%) Order Scombriformes Arripidae (not examined, included here following Yagishita et al. (2009) and Miya et al. (2013)) Gempylidae (not monophyletic) Scombridae (not monophyletic here, but see Miya et al. (2013)) Families not examined: Amarsipidae, Scombropidae (Doiuchi et al. (2004); Miya et al. (2013)), Tetragonuridae (not examined, included here following Miya et al. (2013)) Comment: interfamilial resolution in Scombriformes is tenuous; circumscription of scombriform families into suborders (e.g., Scombroidei, Stromateoidei, Icostoidei) or new orders requires further work. Series Anabantaria (= Anabantiformes sensu Li et al. (2009)) (99%) Order Synbranchiformes (100%) Suborder Mastacembeloidei Not examined: Chaudhuriidae. Suborder Indostomoidei Suborder Synbranchoidei Order Anabantiformes (= Labyrinthici) (100%) Suborder Anabantoidei (95%), new circumscription Suborder Channoidei (100%), new circumscription Suborder Nandoidei (94%), new anabantiform suborder Comment: suborders of Anabantiformes now reflect well-supported monophyletic groups, correcting an error in Version 2 and defining a third suborder (Channoidei) for the family Channidae. Affinities of Channidae with other anabantiform families varies among studies (e.g., Near et al. (2013), Betancur-R. et al. (2013a), and our new results). The new scheme with three subroders is robust to this ambiguity. Series Carangaria (= Carangimorpha sensu Li et al. (2009)) (100%) Order-level incertae sedis in Carangiaria Centropomidae (includes taxa formelry listed in Latidae, following Greenwood (1976) and Li et al. (2011)) Not examined: Lactariidae (included in Carangimorphariae according to Campbell et al. (2013)) Order Istiophoriformes (100%) Order Carangiformes sedis mutabilis (not monophyletic) Comment: Monophyly of Carangiformes is not significantly rejected by the data (Betancur-R. et al., 2013b). Order Pleuronectiformes sedis mutabilis (not monophyletic) Suborder Psettodoidei (100%) Suborder Pleuronectoidei (100%) Paralichthyidae (not monophyletic as traditionally recognized, but monophyletic if the Cyclopsetta group is included in its own family; see Betancur-R. et al. (2013b)) Poecilopsettidae (following Chapleau (1993) and Munroe (2005)) Rhombosoleidae (following Chapleau (1993) and Munroe (2005); includes taxa listed in Achiropsettidae according to Betancur-R. et al. (2013b)) Not examined: Paralichthodidae (following Chapleau (1993) and Munroe (2005)). Comment: Although Psettodidae is not recovered as the sister group of pleuronectoids in the present analysis, the order was resolved as monophyletic by recent studies that address this issue specifically (Betancur-R. et al. (2013b), Betancur-R and Ortí (2014)). Series Ovalentaria (sensu Smith and Near in Wainwright et al. (2012); = Stiassnyiformes sensu Li et al. (2009)) (100%) Order-level incertae sedis in Ovalentaria Congrogadidae sensu Godkin and Winterbottom (1985) and Wainwright et al. (2012), formerly a subfamily of Pseudochromidae Grammatidae (not monophyletic) Superorder Cichlomorphae (93%) Order Cichliformes Order Pholidichthyiformes Superoder Atherinomorphae (100%) Order Atheriniformes (100%) Not examined: Dentatherinidae, Notocheiridae. Order Beloniformes (32%) Suborder Adrianichthyoidei Suborder Belonoidei (100%) (corrected name, listed as Exocoetoidei in version 2) Belonidae (including the former Scomberesocidae; see Lovejoy and Collette (2001) and Lovejoy et al. (2004)) Hemiramphidae (not monophyletic) Zenarchopteridae (not monophyletic here, but see Lovejoy et al. (2004)) Order Cyprinodontiformes (57%) Suborder Aplocheiloidei Not examined: Nothobranchiidae, Rivulidae. Suborder Cyprinodontoidei (100%) Not examined: Anablepidae, Goodeidae, Profundulidae, Valenciidae. Superorder Mugilomorphae (100%) Order Mugiliformes Superorder Blenniimorphae (80%) Order Blenniiformes (100%) sensu Li et al. (2009) Suborder Gobiesocoidei Suborder Blennioidei (not monophyletic; = Blenniiformes sensu Lin and Hastings (2013)) Labrisomidae (includes Stathmonotus; not monophyletic here, but see Lin and Hastings (2013)) Comment: While blennioids are not monophyletic in our results, we note that preliminary analyses resulted in the reciprocal monophyly of gobiesocoids and blennioids, which is congruent with molecular (Wainwright et al., 2012; Lin and Hastings, 2013) and morphological (Springer and Orrell, 2004) evidence. Monophyly of gobiesocoids and blennioids (as separate orders/suborders) is further supported by both morphological (Wiley and Johnson, 2010) and molecular evidence. Chaenopsidae is monophyletic if Stathmonotus is included in Labrisomidae, following Lin and Hastings (2013). Series Eupercaria (99%) Order-level incertae sedis in Eupercaria Lutjanidae (includes the former Caesionidae, following Johnson (1993), Miller and Cribb (2007), and others) Not examined: Dinolestidae, Dinopercidae (see Smith and Craig (2007)). Six families traditionally placed in “Perciformes” are also provisionally listed here are: Bathyclupeidae, Dichistiidae, Hapalogenyidae, Parascorpididae, Symphysanodontidae, Trichonotidae; these are not placed in the recently circumscribed Perciformes given the long history of phylogenetic indistinctiveness between Percoidei, Perciformes, and Percomorpha (e.g., Smith and Craig (2007)). Order Uranoscopiformes (= Paratrachinoidei sensu Li et al. (2009)) (95%) Possibly included: Centrogenyidae (bootstrap support for Uranoscopiformes plus Centrogenyidae is only 41%, as opposed to 99% for Uranoscopiformes s.s. but placement is congruent with results of Near et al. (2013). Order Labriformes sensu stricto (100%) Labridae (includes taxa previoulsy listed in Scaridae; see Price et al. (2010, 2011) and Wainwright et al. (2012)) Order Lobotiformes (100%) Order Ephippiformes (100%) Comment: Greenwood et al. (1966) hypothesized a close affinity between Drepane and ephippids. Order Spariformes sensu Akazaki (1962) and Johnson (1981) (94%) Possibly included (examined): Nemipteridae Not examined but assumed affinity (Johnson, 1981): Centracanthidae Comment: Akazaki (1962) proposed that Lethrinidae, Sparidae, and Nemipteridae were closely related based on specializations of the suspensorium and other features (Johnson (1993)). Johnson (1981) supported the monophyly of Akazaki's spariforms with the addition of Centracanthidae. Order Chaetodontiformes, new eupercarian order (50%) Comment: Although support for this clade is weak (50%) in our analysis, it has been consistently obtained by previous studies with higher nodal support (90-99% in Near et al. (2012a) and 70-89% in Near et al. (2013)). Order Lophiiformes (100%). This order is the sister group of Tetraodontiformes (55% bootstrap); also supported by anatomical evidence (Chanet et al., 2013), larval characters (Baldwin, 2013), and previous molecular studies (e.g. Dettaï and Lecointre, 2008; Miya et al., 2003; Miya et al., 2010). Suborder Lophioidei (100%) Suborder Antennarioidei (100%). Not examined: Brachionichthyidae, Lophichthyidae, Tetrabrachiidae. Suborder Chaunacoidei (100%) Suborder Ogcocephaloidei (100%) Suborder Ceratioidei Not examined: Caulophrynidae, Centrophrynidae, Diceratiidae, Linophrynidae, Neoceratiidae, Thaumatichthyidae. Order Tetraodontiformes (100%). This order is the sister group of Lophiiformes (55% bootstrap); also supported by anatomical evidence (Chanet et al., 2013), larval characters (Baldwin, 2013), and previous molecular studies (e.g. Dettaï and Lecointre 2008; Miya et al., 2003; Miya et al., 2010). Suborder Triacanthodoidei (100%) Suborder Triacanthoidei Suborder Tetraodontoidei (100%) Suborder Moloidei (100%) Suborder Balistoidei (100%) Suborder Ostracioidei (100%) Suborder Triodontoidei Comment: This subordinal classification differs from that proposed by Santini and Tyler (2003). Order Acanthuriformes, restricted circumscription (see also Holcroft and Wiley (2008)) (100%) Order Pempheriformes sedis mutabilis (44%) Acropomatidae (not monophyletic) Polyprionidae (not monophyletic) Not examined: Leptoscopidae (assumed affinity with Creediidae according to Odani and Imamura (2011)) Comment: Tominaga (1986) suggested that features of the cranium and swimbladder may be homologous in Pempheris and Glaucosoma. Although support for Pempheriformes is only 44%, this clade is often recovered in different analyses. Order Centrarchiformes (77%), new circumscription Suborder Centrarchoidei (89%) Sinipercidae (following Li et al. (2010); synonym of Percichthyidae according to Eschmeyer (2014)) Comment: inclusion of Enoploside in this suboorder differs from results obtained by Lavoué et al. (2014). Suborder Cirrhitioidei (similar to Cirrhithoidea sensu Greenwood (1995) and Burridge and Smolenski (2004); treated as Cirrhitiformes in previous versions of the classification) (81%) Not examined but expected affinity (Burridge and Smolenski, 2004; Greenwood, 1995): Aplodactylidae, Chironemidae, Latridae. Suborder Percichthyoidei Percichthyidae (not monophyletic, a new family for Percalates is necessary; includes Percilia) Comment: percichthyoids and Percichthyidae sensu Johnson (1984) are not monophyletic: the Australian species Percalates colonorum and Percalates novemaculeata are not closely related to other members of Percichthyidae (Betancur-R. et al. (2013a); Chen et al. (2014b); Lavoué et al. (2014)), so these species are herein placed in their own suborder (Peter Unmack pers. comm.; Lavoué et al. (2014)). Percalates is listed as a junior synonym of Macquaria by Eschmeyer (2014), but the type species of Macquaria (M. australasica) is closely related to other species of Macquaria (M. ambigua) within Percichthyidae sensu stricto, hence both names are valid genera (Peter Unmack et al., pers. comm.; Lavoué et al. (2014)). Percichthyidae sensu stricto includes Percilia (formerly placed in its own family Perciliidae). Suborder Percalatoidei, new "Percalatidae" (to be described) Comment: Formal description of a new family for Percalates is required to comply with the ICZN. Suborder Terapontoidei (= Clade "h2" of Yagishita et al. (2009); treated as Terapontiformes in previous versions of the classification) (95%) Girellidae (following Carpenter 2001; listed as a subfamily of Kiphosidae by Eschmeyer (2014)) Comment: Although the family name Cirrithidae Macleay 1841 is older than Centrarchidae Bleeker 1859, we retain the name Centrarchiformes for this order in agreement with previous usage but expand its membership following recent proposals by Near et al. (2013) and Chen et al. (2014b) and Lavoué et al. (2014). Order Perciformes (= Serraniformes sensu Lautredou et al. (2013); Li et al. (2009) (99%) Not examined (10 families traditionally placed in Scorpaeniformes): Apistidae, Aploactinidae, Congiopodidae, Eschmeyeridae, Gnathanacanthidae, Neosebastidae, Pataecidae, Perryenidae, Plectrogeniidae, Zanclorhynchidae. Suborder-level incertae sedis in Perciformes Bembropidae (following Smith and Craig (2007); synonym of Percophidae according to Eschmeyer (2014)) Suborder Serranoidei sedis mutabilis (49%) Suborder Percoidei, restricted circumscription (99%) Niphonidae (following Smith and Craig (2007); synonym of Serranidae according to Eschmeyer (2014)) Not examined: Trachinidae. Comment: Lautredou et al. (2013) using seven nuclear markers obtained a clade uniting Percidae and Trachinidae with full support. Suborder Notothenioidei (100%) Bathydraconidae (not monophyletic) Nototheniidae (not monophyletic) Suborder Scorpaenoidei (83%) Scorpaenidae (not monophyletic) Suborder Bembroidei (96%) Suborder Triglioidei sensu Jordan (1923) (100%) Suborder Cottoidei (= Cottimorpha sensu Li et al. (2009)) (100%) Comment: We have chosen to recognize clades within this suborder as infraorders, adopting the ending "–ales" for this rank. Gasterosteales and Zoarcales are probably sister-groups (although not in our results); they have been grouped as Zoarciformes by Li et al. (2009). Infraorder Anoplopomatales (= Anoplopomatoidei in previous classifications) Infraorder Gasterosteales (= Gasterosteoidei in previous classifications--in part, excluding Indostomidae) (100%) Infraorder Zoarcales (= Zoarcoidei in previous classifications) (100%) Bathymasteridae (not monophyletic) Stichaeidae (not monophyletic) Not examined: Ptilichthyidae, Scytalinidae. Infraorder Hexagrammales (100%) (= Hexagrammoidei in previous classifications) Hexagrammidae (sensu stricto, following Smith and Busby (2014)) Comment: Hexagrammidae as formerly defined is not monophyletic. We now split it into two families (formerly subfamilies): Hexagrammidae (sensu stricto) and Zaniolepidoales following Washington et al. (1984), Shinohara (1994), and Smith and Busby (2014). As in previous cottoid classifications, these families are placed in their own infraorders (note that previous classifications use suborders instead of infraorders). Infraorder Zaniolepidoales (= Zaniolepidoidei sensu Smith and Busby (2014)) Zaniolepididae (sensu Smith and Busby (2014), formerly a subfamily in Hexagrammidae; not listed by Eschmeyer (2014)) Infraorder Cottales (96%) (= Cottoidei sensu Smith and Busby (2014)) Agonidae (not monophyletic here but see Smith and Busby (2014); includes the former Hemitripteridae) Cottidae (not monophyletic here but see Smith and Busby (2014); includes the former Abyssocottidae, Comephoridae, and Cottocomephoridae.) Psychrolutidae (includes the former Bathylutichthyidae) Scorpaenichthyidae (see Smith and Busby (2014)) Not examined: Jordaniidae (not listed by Eschmeyer (2014); newly defined by Smith and Busby (2014)), Ereuniidae, Normanichthyidae, and Rhamphocottidae. Comment: Smith and Busby (2014) changed the membership of Cottidae and Psycholutridae to achieve reciprocal monopyhyly of these families. Our phylogenetic results do not resolve a monophyletic Cottidae even under the new circumscription proposed by these authors. We refer the readers to that study for more details. Superclass Sarcopterygii (96%) Class Coelacanthimorpha (= Actinistia) Order Coelacanthiformes Class Dipnotetrapodomorpha sedis mutabilis (65%) Subclass Dipnomorpha (100%) Superorder Ceratodontae Order Ceratodontiformes Suborder Ceratodontoidei Suborder Lepidosirenoidei (100%) Subclass Tetrapodomorpha (100%)
General comments on the classification This classification (version 3), as its previous versions, also buids on Wiley and Johnson (2010) and Betancur-R. et al. (2013a), intending to preserve names and taxonomic composition of groups as much as possible. However, adjustments have been made to recognize well-supported molecular clades, many of which also have been obtained by previous molecular studies (several examples discussed below--references are cited when decisions are based on other publications). Order-level or supraordinal taxa have been erected (listed as new) or resurrected on the basis of well-supported clades only (>90% bootstrap values). Current taxon names supported by previous molecular or morphological studies have been retained if congruent with our results, even if bootstrap support is low (e.g., Osteoglossocephalai sensu Arratia (1999) with only 38% bootstrap). In some cases, ordinal or subordinal taxa that were not monophyletic in our analysis are also validated, as long as the incongruence is not supported by strong bootstrap values. Examples include the suborder Blennioidei (not monophyletic here but monophyletic in Wainwright et al., 2012) and the order Pleuronectiformes (not monophyletic here but monophyletic in Betancur-R. et al., 2013b). The classification is presented in phylogenetic order up to the subordinal rank (following the branching order in our results), but families within orders (or suborders) are listed alphabetically. A total of 67 orders are classified, of which one is new (Chaetodontiformes) and two were sinked (Terapontiformes and Cirrhitiformes) in version 3. Three orders were new in the version 1 (Holocentriformes, Istiophoriformes, and Pempheriformes) and two were new in version 2 (Lobotiformes and Terapontiformes). The ordinal status of several percomorph families examined (as well as many others unexamined) belonging to the Series Carangimorpharia, Ovalentaria, and Percomorpharia remains uncertain (incertae sedis) due to poor phylogenetic resolution. We therefore list these families as incertae sedis within each of these groups (Carangimorpharia, Ovalentaria, and Percomorpharia) awaiting new phylogenetic evidence to clarify their ordinal status. Family names for bony fishes are based on Eschmeyer and Fong (2014) and van der Laan et al. (2013), with minor modifications, indicted in each case in the classification. Van der Laan et al. (2013) and Eschmeyer (2014) should be consulted for authorship of family names and Wiley and Johnson (2010) for authorship of ordinal and subordinal names. Our list is not intended as a comprehensive revision of valid family names. Instead, it is simply an adaptation of their lists based on published studies that we know validate or synonymize taxa using explicit phylogenetic evidence. In order to minimize the number of non-monophyletic taxa, we have changed the membership of many families traditionally recognized in ichthyology whose validity is strongly challenged by molecular/morphological phylogenetic analyses. For instance, we no longer recognize families such as Scaridae, Caesionidae, and Microdesmidae (lumped with Labridae, Lutjanidae, and Gobiidae, respectively). Accordingly, the number of non-monophyletic families decreased from 41 in version 2 to 30 in this version (indicated in each case). A total of 496 families of bony fishes are now recognized (excluding tetrapods), of which 394 (79.4%) were examined. For each order/suborder we list all families examined as well as the unexamined families whose taxonomic affinity is expected on the basis of traditional taxonomy or phylogenetic evidence. The list of 102 unexamined families can be easily obtained from this spreadsheet that also contains the complete classification, and is intended as a resource to help fish systematists direct future sequencing efforts.
The new classification scheme presented here (version 3) should be considered work in progress as any other hypothesis. It is likely to include involuntary errors and omissions in addition to the many unexamined, sedis mutabilis, and incertae sedis taxa. Updates should be forthcoming as new evidence becomes available and feedback from experts help refine it. Please send comments or concerns to classification@deepfin.org. For the most updated version always visit DeepFin.
List of all changes from version 2 to version 3
References Eschmeyer, W.N. 2014. Catalog of Fishes: Genera, Species, References. Electronic Version accessed 16 July 2014. Lavoue, S., K. Nakayama, D.R. Jerry, Y. Yamanoue, N. Yagishita, N. Suzuki, M. Nishida, and M. Miya. 2014. Mitogenomic phylogeny of the Percichthyidae and Centrarchiformes (Percomorphaceae): comparison with recent nuclear gene-based studies and simultaneous analysis. Gene (in press). Nelson, J.S. 2006. Fishes of the world. John Wiley & Sons, Hoboken. van der Laan, R., R. Fricke, and W.N. Eschmeyer. 2013. Family-Group Names (version 11/17/2013). California Academy of Sciences http://research.calacademy.org/ichthyology/catalog/family. |
