Taxonomic recommendations for British birds: Fourth report

Abstract

This paper is the fourth report of the Taxonomic Sub-Committee of the BOU Records Committee relating to the British List. Species-level decisions are based on criteria outlined by Helbig et al. (2002, Guidelines for assigning species rank; Ibis 144: 518–525). The third report of the Sub-Committee was published in Ibis 147: 821–826. In our previous report we referred to the name mariloides as being unavailable (Ibis 147: 821–826). This is incorrect; the name is inapplicable. Phylogenetic analyses based on DNA-DNA hybridization data (van Tuinen et al. 2001. Proc. R. Soc. Lond. B 268: 1345–1350), mitochondrial and nuclear DNA sequences (van Tuinen et al. 2001; Chubb 2004. Mol. Phylogen. Evol. 30: 140–151; Cracraft et al. 2004. In: Cracraft & Donoghue, Reconstructing the Tree of Life, pp. 468–489; Ericson et al. 2006. Biol. Lett. 2: 543–547) and morphology (Mayr & Clarke 2003. Cladistics 19: 527–553; Mayr 2004. Zool. J. Linn. Soc. 140: 157–169; Manegold 2006. Acta Ornithol. 41: 79–82) provide congruent support for a sister-group relationship of flamingos Phoenicopteriformes and grebes Podicipediformes. Storer (2006. Auk 123: 1183–1184) challenged some of the morphological evidence for the flamingo-grebe grouping, and a recent comprehensive phylogeny based on morphological characters (Livezey & Zusi 2007. Zool. J. Linn. Soc. 149: 1–95) supported a sister relationship between divers and grebes. This latter paper contained several hypotheses about higher order avian relationships that are not supported by other highly congruent studies, and it seems likely that any apparent close relationship between grebes and divers is due to convergence. Recognition of the flamingo-grebe clade is therefore recommended. This clade was recently named Mirandornithes (Sangster 2005. Ibis 147: 612–615) and is placed between Ciconiiformes and Falconiformes, with the flamingos listed before the grebes. The author of the scientific name for the American Bittern was changed without explanation in the 6th Checklist (BOU 1992. Checklist of the Birds of Britain and Ireland, 6th edn.) from Montagu to Rackett. For clarification, both descriptions were based on the same specimen, but Rackett's publication pre-dated that of Montagu by several weeks and thereby takes precedence (Macdonald & Grant 1951. Bull. Brit. Orn. Club 71: 30). The scientific name of this species is as given above (cf. Ibis 147: 821–826). Allen's Gallinule Porphyrula alleni Purple Gallinule Porphyrula martinica The purple gallinules Porphyrula, Porphyrio, and Notornis constitute a monophyletic group. Porphyrula differs from Porphyrio in its smaller size, less massive bill and more oval nostril but the two genera share a number of characters, particularly of the hindlimb (Olson 1973. Wilson Bull. 85: 381–416). Porphyrula (i.e. P. alleni, P. martinica, P. flavirostris) forms a monophyletic group which is sister to Porphyrio/Notornis, which also forms a monophyletic group (Livezey 1998. Philos. Tr. R. Soc. Lond. B 353: 2077–2151). Both the continued recognition of Porphyrula and Porphyrio as separate genera and the inclusion of Porphyrula in Porphyrio are potentially consistent with their evolutionary relationships. However in light of the striking morphological similarities between the species in these genera there is growing international support for the inclusion of Porphyrula in Porphyrio. This treatment has been adopted in major checklists and handbooks (e.g. Urban et al. 1986. The Birds of Africa 2; del Hoyo et al. 1996. Handbook of the Birds of the World 3; Taylor & van Perlo 1998. Rails; Banks et al. 2002. Auk 119: 897–906; Dickinson 2003. The Howard and Moore Complete Checklist of the Birds of the World). It is recommended that Porphyrula be treated as a subgenus within Porphyrio. Allen's Gallinule and Purple Gallinule are on the British List and become Porphyrio alleni and Porphyrio martinica, respectively. Molecular phylogenetic analysis (Pereira & Baker 2005. Condor 107: 514–526) groups two species traditionally placed in the genus Heteroscelus (i.e. Grey-tailed Tattler H. brevipes and Wandering Tattler H. incanus) and Willet Catoptrophorus semipalmatus in the Tringa clade. These data suggest that the tattlers and Willet should be reclassified in the genus Tringa. Grey-tailed Tattler is on the British List and becomes Tringa brevipes. Phylogenetic analyses of mitochondrial and nuclear DNA sequences and morphology (Pereira & Baker 2005. Condor 107: 514–526) have clarified the evolutionary relationships among the shanks. The species in the genera Xenus, Actitis and Tringa should be listed in the following sequence: Terek Sandpiper Xenus cinereus Common Sandpiper Actitis hypoleucos Spotted Sandpiper Actitis macularius Green Sandpiper Tringa ochropus Solitary Sandpiper Tringa solitaria Grey-tailed Tattler Tringa brevipes Spotted Redshank Tringa erythropus Greater Yellowlegs Tringa melanoleuca Common Greenshank Tringa nebularia Lesser Yellowlegs Tringa flavipes Marsh Sandpiper Tringa stagnatilis Wood Sandpiper Tringa glareola Common Redshank Tringa totanus The recommendation that Yellow-legged Gull L. michahellis and Armenian Gull L. armenicus should be treated as separate species from Herring Gull L. argentatus was communicated previously (Sangster et al. 2005. Ibis 147: 821–826). Phylogenetic evidence based on analysis of mitochondrial DNA (mtDNA) sequences indicates that the large white-headed gull complex consists of two main clades: (1) an ‘Atlantic’ clade consisting of Yellow-legged Gull, Armenian Gull, most individuals of Herring Gull, and including Great Black-backed Gull L. marinus and Palearctic individuals of Glaucous Gull L. hyperboreus; (2) an ‘Aralo-Caspian’ clade consisting of L. a. cachinnans, L. a. barabensis, L. a. heuglini, L. a. taimyrensis, Lesser Black-backed Gull L. fuscus, Kelp Gull L. dominicanus, some individuals of L. a. argentatus, and including an ‘Arctic/Pacific’ grouping of L. a. vegae, L. a. smithsonianus, L. a. mongolicus, Slaty-backed Gull L. schistisagus, Iceland Gull L. glaucoides, Glaucous-winged Gull L. glaucescens and Nearctic individuals of Glaucous Gull (Crochet et al. 2002. Auk 119: 603–620; Crochet et al. 2003. Evolution 57: 2865–2878; Liebers et al. 2004. Proc. R. Soc. Lond. B 271: 893–901). Western Gull L. occidentalis is not included in these clades and forms an outgroup. Genetic structure between the taxa is generally well-defined, except within the Arctic/Pacific grouping of the Aralo-Caspian clade, for which the taxa are poorly separated with some shared haplotypes (Gay et al. 2005. Auk 122: 684–688). Morphological and genetic variation in Lesser Black-backed Gulls L. fuscus fuscus, L. f. intermedius and L. f. graellsii is clinal (Liebers & Helbig 2002. J. Evol. Biol. 15: 1021–1033). These taxa are closely related to the West Siberian taxa L. a. heuglini, L. a. taimyrensis and L. a. barabensis with evidence of continuing low levels of gene flow in spite of apparent ecological separation of L. f. fuscus and L. a. heuglini in parapatry (Filchagov et al. 1992a. Zool. Zh. 71: 148–152; Rauste 1999. Limicola 13: 105–128; 153–188; Liebers & Helbig 2002; Liebers et al. 2004). It is recommended that these taxa be treated as a single species L. fuscus. Variation in heuglini is slight (Buzun 2002. Br. Birds 95: 216–232) and the validity of taimyrensis has been questioned (Filchagov et al. 1992b. L’Oiseau 62: 128–148; Yésou 2002. Dutch Birding 64: 271–298). Caspian Gull L. a. cachinnans is diagnosably distinct from all other taxa on the basis of plumage and vocalisations (Panov et al. 1991a. Zool. Zh. 70/1: 76–90; Panov et al. 1991b. Zool. Zh. 70/3: 73–89; Garner & Quinn 1997. Br. Birds 90: 25–62; Klein & Gruber 1997. Limicola 11: 49-75; Liebers & Dierschke 1997. Dutch Birding 19: 277–280; Jonsson 1998. Alula 3: 74–100; Yésou 2002). It forms a discrete genetic grouping and is probably most closely related to the West Siberian gulls heuglini and barabensis. However, introgression between cachinnans and the West Siberian taxa is restricted and probably unidirectional (Panov & Monzikov 2000. Br. Birds 93: 227–241; Liebers et al. 2004). Recent range expansion has led to widespread hybridization with argentatus Herring Gulls in Central and Eastern Europe which may partly underlie the apparently large intrataxon variation in cachinnans (Klein & Gruber 1997. Limicola 11: 49–75; Liebers & Dierschke 1997; Panov & Monzikov 1999. Russ. J. Zool. 3: 129–141). However the hybrid zone is reported to be narrow in relation to the potential dispersal distance of the parental taxa (Neubauer et al. 2006. Vogelwelt 127: 11–22; Yakovets 2006. Vogelwelt 127: 23–30). Based on this evidence, it is recommended that Caspian Gull be treated as a monotypic species L. cachinnans. Genetic evidence suggests that East Siberian and American Herring Gulls, L. a. vegae, L. a. mongolicus and L. a. smithsonianus fall within the Arctic/Pacific species group of the Aralo-Caspian clade (Liebers et al. 2004; de Knijff et al. 2005. Birding 37: 402–411; Gay et al. 2005). There are sharp genetic and morphological boundaries between vegae and the parapatric or partially sympatric West Siberian taxon heuglini (Yésou 1994. Alauda 62: 247-252; Liebers et al. 2004; Yésou 2002). Many individuals of vegae, mongolicus and smithsonianus can be identified to subspecies (Lonergan & Mullarney 2004. Dutch Birding 26: 1–35), but diagnosability with respect to each other has not been demonstrated. Although many individuals can be identified, smithsonianus is not fully diagnosable from European Herring Gulls on the basis of plumage alone (Lonergan & Mullarney 2004; Adriaens & Mactavish 2004. Dutch Birding 26: 151–179) but is clearly differentiated on the basis of mtDNA (Crochet et al. 2002, 2003, Liebers et al. 2004, Gay et al. 2005). In addition, European Herring Gulls are reported to respond poorly to calls of smithsonianus (Frings et al. 1958. Ecology 39: 126–131). Based on current evidence, American Herring Gull L. smithsonianus is best treated as a separate species. Given their lack of diagnostic differences, vegae and mongolicus are considered conspecific with L. smithsonianus. We recommend recognition of the following species to better reflect recent advances in knowledge of the evolution and systematics of large gulls: Caspian Gull Larus cachinnans (monotypic) Lesser Black-backed Gull Larus fuscus (polytypic, including fuscus, intermedius, graellsii, heuglini, taimyrensis, barabensis) American Herring Gull Larus smithsonianus (polytypic, including smithsonianus, vegae, mongolicus) Yellow-legged Gull Larus michahellis (polytypic, including michahellis, atlantis) Armenian Gull Larus armenicus (monotypic) Herring Gull Larus argentatus (polytypic, including argentatus, argenteus) A manuscript explaining these decisions has been submitted to British Birds. Two research programmes, one based on morphology (Chu 1998. Cladistics 14: 1–43) and another based primarily on mitochondrial DNA sequences (Crochet et al. 2000. J. Evol. Biol. 13: 47–57; Pons et al. 2005. Mol. Phylogen. Evol. 37: 686–699) have examined phylogenetic relationships of the entire group of gulls. Both studies indicate that the genus Larus, as currently defined [e.g. Voous 1977. List of Recent Holarctic Bird Species. Br. Ornithol. Union, London; Cramp & Simmons 1983. The Birds of the Western Palearctic. Vol. 3. Oxford UP, Oxford; Burger & Gochfield 1996. Gulls. In: del Hoyo, J., Elliot, A. & Sargatal, J. (eds.) Handbook of the Birds of the World. Vol. 3. Hoatzin to Auks. Lynx Edicions, Barcelona], is not monophyletic. These studies indicate that the generic limits of the gulls need revision. Both studies indicated a separate position of Creagrus, Rissa, Xema, Pagophila from all other gulls, supporting the continued recognition of these genera, but also supported another well-defined clade (which includes Slender-billed Gull Larus genei, Bonaparte's Gull L. philadelphia and Black-headed Gull L. ridibundus) which is not the sister group to other gulls in Larus as currently defined. Both studies also resolved a sister-group relationship of Ross's Gull and Little Gull and their separate position from the main clade of gulls. A taxonomic revision that would maintain monophyly of gull genera with minimum change to the British List would be to include all gulls in Larus, except those currently placed in Creagrus, Rissa, Xema and Pagophila. This arrangement however does not reflect the taxonomic information derived from the cited studies. Pons et al. (2005) in contrast suggested recognition of ten genera, requiring adoption of five new generic names: Chroicocephalus (which includes Slender-billed Gull, Bonaparte's Gull and Black-headed Gull); Saundersilarus (Saunder's Gull); Hydrocoloeus (Ross's and Little Gull); Leucophaeus (some of the New World ‘hooded’ gulls); Ichthyaetus (southern Palaearctic ‘black-headed gulls’). Four of these putative genera have been recorded in Britain. Retention of separate genera for Little and Ross's Gull would be justifiable on the basis of the long branch lengths separating the two, which are comparable with those separating Xema and Pagophila. For the purposes of the British List, the TSC recommends recognition of an intermediate taxonomy that adopts Chroicocephalus for the clade which includes Slender-billed Gull, Bonaparte's Gull and Black-headed Gull and adopts Hydrocoloeus for Little Gull, but does not change the generic status of other gull species, as follows: Ivory Gull Pagophila eburnea Sabine's Gull Xema sabini Black-legged Kittiwake Rissa tridactyla Slender-billed Gull Chroicocephalus genei Bonaparte's Gull Chroicocephalus philadelphia Black-headed Gull Chroicocephalus ridibundus Little Gull Hydrocoloeus minutus Ross's Gull Rhodostethia rosea Laughing Gull Larus atricilla Franklin's Gull Larus pipixcan Mediterranean Gull Larus melanocephalus Audouin's Gull Larus audouinii Pallas's Gull Larus ichthyaetus Mew Gull Larus canus Ring-billed Gull Larus delawarensis Lesser Black-backed Gull Larus fuscus Herring Gull Larus argentatus Yellow-legged Gull Larus michahellis Caspian Gull Larus cachinnans American Herring Gull Larus smithsonianus Iceland Gull Larus glaucoides Glaucous Gull Larus hyperboreus Great Black-backed Gull Larus marinus The Chroicocephalus gull clade is strongly supported by congruency between Pons et al. (2005) and the results of Chu (1998). The groups designated as ‘hooded’ and ‘black-headed’ gulls by Pons et al. (2005) and which may merit generic status were not found by Chu (1998), so further division within Larus s.s. is not recommended at this stage. The gender of the name Onychoprion is masculine and the correct names for these taxa are: Aleutian Tern Onychoprion aleuticus Sooty Tern Onychoprion fuscatus Brachyramphus m. marmoratus and B. m. perdix show differences in plumage [Konyukhov & Kitaysky 1995. In: Ralph et al. (eds), USDA Forest Service General Technical Report PSW-GTR-152, pp. 23–29; Gaston & Jones 1998. The Auks], mtDNA restriction fragment lengths (Zink et al. 1995. Condor 97: 639–649) and allozymes and mtDNA sequences (Friesen et al. 1996. Condor 98: 681–690); allozymes and mtDNA sequences suggest that marmoratus is more closely related to Kittlitz's Murrelet B. brevirostris than to perdix. B. marmoratus is best treated as two species: Marbled Murrelet B. marmoratus (monotypic) Long-billed Murrelet B. perdix (monotypic) A recently-published phylogenetic analysis of the kingfishers indicates that Ceryle rudis is the sister of Chloroceryle and is not closely related to Ceryle alcyon (Moyle 2006. Auk 123: 487–499). This indicates that the current treatment of Megaceryle as a subgenus of Ceryle does not accurately reflect their phylogenetic relationships. In view of their distinctive morphology and to avoid paraphyly of Ceryle, three genera of ceryline kingfishers are recognized: Megaceryle, Ceryle and Chloroceryle. Both Miller (1912. Bull. Am. Mus. Nat. Hist. 31: 239–311; 1920. Auk 37: 422–429) and Fry (1980. Living Bird 18: 113–160) emphasized anatomical differences among the three groups in support for treatment as three genera. These data suggest that Belted Kingfisher should be reclassified in the genus Megaceryle. Belted Kingfisher is on the British List and becomes Megaceryle alcyon. The author of the scientific name is Laxmann (Dickinson & Ericson 2002. Zool. Verh. 340: 205–206). Phylogenetic analyses of mitochondrial DNA sequences (Klicka et al. 2005. Mol. Phylogen. Evol. 34: 486–500) indicate that Varied Thrush does not group with African and Asian Zoothera thrushes but is part of a New World radiation of thrushes (which also includes Catharus, Hylocichla, Cichlopsis, Entomodestes and Ridgwayia). On the basis of this evidence, Varied Thrush is reinstated in a monotypic genus and becomes Ixoreus naevius. The gender of the name Poecile is feminine (not as given in Ibis 147: 821–826). Therefore, the correct name of Willow Tit is Poecile montana. We thank Dick Banks, Edward Dickinson and Laurent Raty for their help and apologize to any others that we have omitted to mention.