Sylvioidea I

Passerines

Tyranni: Suboscines

Passeri: Oscines

Passerida

Sylvioidea
Muscicapoidea and allies
Passeroidea

The 45 Orders

Paleognaths

Galloanserae

Metaves

Pelecanae

Charadriae

Passerae

Sylvioidea

We have reached the heart of the Passerida, the 3-way division between Sylvioidea, the Muscicapoidea clade, and Passeroidea. Barker et al., (2004) found that Sylvoidea was sister to the rest, but Irestedt and Ohlson (2008) found that Passeroidea was sister to the rest. I'm treating Certhioidea, Reguloidea, Bombycilloidea, and Muscicapoidea as more closely related to each other than to Sylvioidea or Passeroidea. I'm also treating Sylvoidea as sister to the rest and so will consider Sylvoidea first.

Sylvioidea has been a mess for a long time. Both the Sylviidae and Timaliidae have been treated at taxonomic trashcans. If a species is hard to classify, just dump it in the Sylviidae or Timaliidae. Then we don't have to worry about it. This has been a particular problem because many of the species look very similar. The lack of a distinct juvenile plumage helps separate them from the Muscicapoidea, but there's really not much to go on in terms of morphology.

Just how much of a mess the Sylvioidea were is exemplified by the Sibley-Monroe checklist. They removed the cisticolas, kinglets, parids, and white-eyes and others, and still had a mass of 560 species lumped together in the family Sylviidae with the Timaliidae nested inside as a tribe (Timaliini). This was actually progress as it was an admission about how little we knew.

Recently, Sylvioidea has been undergoning major restructuring, starting with the formation of a new superfamily---Certhioidea (Cracraft et al., 2004). The Certhioidea are comprised of several families of small birds that the modified SAM list has at the beginning of the Sylvioidea. In fact, they may be more closely related to the Muscicapoidea. One could include them in the Muscicapoidea, but I follow the Tree of Life and place in a separate superfamily. The families in question are the Sittidae (nuthatches), Tichodromadidae (wallcreeper), Certhiidae (treecreepers), Polioptilidae (gnatcatchers and gnatwrens), and Troglodytidae (wrens).

Although these results cleared the ground, what was really needed was to sample a large number of sylvioid taxa. Alström, Beresford, Ericson and others attacked the problem on a wide scale, while Cibois and company focused on Timaliidae. The result is that the Sylvioidea have been sliced and diced. You may want to consult the tree to see the changes.

Much work is still to be done on sylvioid taxonomy, but Alström, Cibois, and their associates have managed to build a framework of sylvioid families. The framework will probably need some adjustment and rearrangement, but I think it creates a coherent set of families. The framework involves the creation of fourteen new families: Panuridae, Stenostiridae, Nicatoridae, Macrosphenidae, Acrocephalidae, Donacobiidae, Bernieridae, Locustellidae, Pnoepygidae, Hyliidae, Cettiidae, Phylloscopidae, Pellorneidae, and Leiothrichidae. The papers by Alström et al. (2006), Barker et al. (2004), Beresford et al. (2005), Jønsson and Fjeldså (2006a), Johansson et al. (2008b), and Gelang et al. (2009) lay out the big picture, and Cibois (2003), Cibois et al. (1999, 2001, 2002), Fuchs et al. (2006a, b), Nguembock et al. (2007), Pasquet et al. (2006), and Zhang et al. (2007) fill in many of the details. I also recommend Don Roberson's web page, The Break-up of the Old World Warblers, which covers the changes to the Sylviidae.

Stenostiridae, Remizidae, and Paridae

The Sylvioidea start with a small clade containing three small families: Stenostiridae (fairy flycatchers), Remizidae (penduline-tits), and Paridae (chickadees and titmice). There is some controversy about whether this clade belongs in the Sylvioidea. Barker et al. (2004), Beresford et al. (2005), and Treplin et al. (2008) consider it the sister group to the rest of the Sylvioidea. Treplin et al. suggest they are a separate superfamily, Paroidea. In contrast, Ericson and Johansson (2003) and Alström et al. (2006) place the parid group outside Sylvioidea.

Stenostiridae: Fairy Flycatchers Beresford et al., 2005

3 genera, 9 species Not HBW Family

The Stenostiridae are one of the new families. They are a group of small African and Asian flycatchers that were previously scattered across three superfamilies. They include the Fairy Flycatcher Stenostira (Sylvioidea: Sylviidae), the Elminia (Corvoidea: Monarchidae) crested-flycatchers, and the Culicicapa (Muscicapoidea: Muscicapidae) canary-flycatchers. The last three Elminia were previously considered part of the genus Trochocercus in the corvid family Monarchidae.

Nguembock et al. (2008b) was the first paper to analyze the 8 then-known species in the Stenostiridae. They confirm the arrangement below, and like the other analyses, they find that the Stenostiridae are only distantly related to the other Passerida.

More recently, Nyári et al. (2009a) and Fuchs et. al (2009) found a ninth member of the Stenostiridae. It is the Yellow-bellied Fantail, Rhipidura hypoxantha, which had been considered unusual. Some authors had previously separated it as the monotypic genus Chelidorhynx. Accordingly, it becomes Chelidorhynx hypoxantha. Fuch et al. (2009) show that it is sister to the Fairy Flycatcher.

Interestingly, both clades of the Stenostiridae (Chelidorhynx/Stenostira and Culicicapa/Elminia) consist of an African genus and an Oriental genus. Fuchs et al. (2009) find that both splits seem to have occurred at about the time. Nguembock et al. (2008b) note that there was a biological interchange between Africa and Asian at about that time. Interestingly, the Stenostiridae are not the only family to have such a distribution. Fuchs et al. (2009) note the parallel with the clade containing the South African Chaetops and the Oriental Eupetes. There may also be a similar case with the African family Platysteiridae, which seems to also contain the New Guinean Rhagologus.

Remizidae: Penduline-Tits Olphe-Galliard, 1891

4 genera, 12 species HBW-13

Remizidae tree Whether Cephalopyrus belongs here is uncertain. There hasn't been a genetic study of the entire group. The arrangment is based on Gill et al. (2005).

Paridae: Tits, Chickadees, Titmice Vigors, 1825

7 genera, 59 species HBW-12

The main change in the membership of the Paridae was mentioned earlier, the Paridae gain Hume's Groundpecker (formerly Hume's Ground-Jay), Pseudopodoces humilis from the Corvidae (James et al., 2003; Gill et al., 2005). The arrangement of genera and species is based on Gill et al. (2005) and James et al. (2003).

Paridae tree

Panuridae and Alaudidae

Here the Sylvioidea splits into two clades. The smaller one contains the Panuridae and Alaudidae (larks). Panuridae currently consists of just one species, the Bearded Reedling (Panurus biarmicus), although others may still be hidden within Sylviodea. There have been many opinions about how to classify the Reedling, and it is sometimes considered a parid or parrotbill, and usually named to match (Bearded Tit or Bearded Parrotbill). It had recently been considered a parrotbill, in the now-defunct Parrotbill family (Paradoxornithidae). The modified SAM list had already rolled the Parrotbills into the Sylviidae. Using overlapping but distinct data sets, Alström et al. (2006), Ericson and Johansson (2003), and Fuchs et al. (2006a) concur that Panurus is sister to the larks.

Panuridae: Bearded Reedling des Murs, 1860

1 genus, 1 species Not HBW Family

Alaudidae: Larks Vigors, 1825

21 genera, 98 species HBW-9

Alaudidae The larks are the major piece of this clade. Although they were considered part of Passeroidea in the modified SAM list, this is not correct. Rather, they belong in the Sylvioidea. Although the placement of the larks may have changed, their composition has not. This reflects the fact that they are one of the two easily identifiable passerine families (the swallows are the other).

Tieleman et al. (2003) found that some of the generic limits need adjusting (including Mirafra) and that some general reorganization is called for. Some of the blanks were filled in as Ryan et al. (1998) separated Barlow's Lark from Karoo Lark, while Ryan and Bloomer (1999) studied the long-billed larks. The result of all this is the tree. Whether Alaemon (with Chersophilus?) or Heteromirafra is basal has not been settled. Gray's Lark is now in a different genus (Ammomanopsis), as are the long-billed larks (Certhilauda). Although the placement of Ramphocoris and Melanocorypha is not completely arbitrary, it could use more supporting evidence.

Guillaumet et al. (2005, 2006, 2008) have been studying the Galerida larks. This has lead to the split of Crested Lark, Galerida cristata, with the long-billed subspecies in the Maghreb becoming Maghreb Lark, Galerida macrorhyncha, while the short-billed subpsecies remain in Galerida cristata. The subspecies of Thekla Lark, Galerida theklae, around the horn of Africa seem to be candidates for one or more future splits (the subspecies involved include ellioti, praetermissa, and hueii).

Nicatoridae

The nicators form are a small African clade. They were formerly considered to be in the bulbul family, but seem not closely related to anything else (Beresford et al., 2005; Johansson et al., 2008b). They seem to require their own family.

Nicatoridae: Nicators Informal?

1 genus, 3 species Not HBW Family

Macrosphenidae

Macrosphenidae: Crombecs, African Warblers Wolters, 1983

7 genera, 19 species Not HBW Family

Macrosphenidae tree The Macrosphenidae are another one of the new families. This is formed entirely from the Sylviidae and consists of crombecs and African warblers (see Beresford et al., 2005; Nguembock et al., 2007). Johansson et al. (2008b) put it next in order.

Victorin's Warbler was formerly considered part of the genus Bradypterus. The rest of Bradypterus is now in Locustellidae.

Although Alström placed Macrosphenus kretschmeri in the bulbuls, this turned out to be a mistake based on a mislabelled specimen (Johansson et al., 2007b). However, Macrosphenus kretschmeri is different enough from the other Macrosphenus to sometimes get its own genus (Suaheliornis). It has yet to be sequenced and it is possible it does not belong here. I've also put Grauer's Warblers here, as it is thought to be close to Macrosphenus. The rest of the family has been rearranged using Johansson et al. (2007b).

Core Sylvioidea: Hirundinidae and the rest

The rest of the Sylvioids, over a thousand species, are likely more closely related to each other than to the preceeding families. It's unclear exactly how they relate to one another as several groups appear to have split in quick succession. Various analyses group them various ways (as have I), but there is an unresolved 5-way basal polytomy. The 5 groups are the swallows (Hirundinidae), the Acrocephalus group, the cisticolas (Cisticolidae), the bulbuls (Pycnonotidae), and a large group containing the rest of the Sylvioidea (9 families).

Since these five groups are treated as a polytomy, they are ordered by size of the group, the swallows having the fewest species and the big group containing the babblers having the most, almost 600 species.

Hirundinidae: Martins, Swallows Rafinesque, 1815

18 genera, 88 species HBW-9

The organization of the swallows is based on Sheldon et al. (2005), with Whittingham et al. (2002) and Dor et al. (2010) providing additional information. The arrangement within Tachycineta is based on Cerasale et al. (2012). I've inserted subfamiles and tribes to make the organization a bit clearer. There is a basal division between the river martins (Pseudochelidoninae) and everything else (Hirundinae). Hirundinae divides into three groups, with the Psalidoprocnini likely closer to the Hirundinini. The tree shows this in detail:

Hirundinidae tree

The Cave Swallow, Petrochelidon fulva, which includes two geographically-based distinct color morphs, is sometimes considered a candidate for a split. However, Kirchman et al. (2000) found substantial gene flow between the various populations. It was also striking how genetically distinct the Florida population had become in just a few years.

Pseudochelidoninae: River Martins Shelley, 1896

Hirundininae Rafinesque, 1815

Prognini Cassin, 1853

Psalidoprocnini Sharpe, 1870

Hirundinini Rafinesque, 1815

Acrocephalus Group

The next five families are relatively closely related.

Pnoepygidae: Wren-babblers Gelang et al., 2009

2 genera, 7 species Not HBW Family

Gelang et al. (2009) found that the Pnoepyga wren-babblers do not belong to any of the extant Sylvioidea families. They recommend placing it in its own family. Their analysis suggested they might be close to the cisticolas or bulbuls, but with low support. The overlapping analyses of Alström et al. (2011a) and Irestedt et al. (2011) included any family that might be related. The Pnoepygidae ended up as a basal member of this group. It is possible that the cisticolas also belong here.

Gelang et al. (2009) only sequenced two of the Pnoepyga. Elachura has been thought to be closely related, and some or all of the Spelaeornis have previously been placed in Pnoepyga, so it is possible some of them belong near Pnoepyga. Although they have not been sequenced, I'm guessing that Elachura formosa and Spelaeornis caudata and badeigularis belong near Pnoepyga based on plumage and structure. I don't have an alternative name for the two Spelaeornis, but given their appearance, it seems plausible to move them into Elachura, at least for now. Spelaeornis chocolatinus definitely belongs in Timaliidae, and I've left the other Spelaeornis wren-babblers there.

Acrocephalidae: Acrocephalid Warblers Salvin, 1882 (1838)

7 genera, 62 species Not HBW Family

The reed-warblers, Acrocephalidae, are another new family formed out of the Sylviidae. The arrangement here is primarily based on Fleischer et al. (2007), Cibois et al. (2007, 2008, 2011) and Fregin et al. (2009). The multigene analysis of the last supercedes Leisler et al. (1997) and Helbig and Seibold (1999). Cibois et al. (2007, 2011) give insight into how the reed-warblers expanded into Polynesia. The exact branching pattern differs some between Cibois et al. and Fleischer et al., and I have used a compromise version here which presumes each did best where taxon sampling was densest. I follow this in preference to Fregin et al. for the Polynesian species, as Fregin et al. only sampled one gene from that group. For the other species, Fregin et al. is given preference.

Although there had been some question about which family it belongs to, Nesillas seems basal in the Acrocephalidae. Johansson et al. (2008b) included one member of Nesillas in their analysis. They found that it too was part of Acrocephalidae. Johansson et al. (2008b) and Fregin et al. (2009) concur that it is also more basal among the acrocephalids sampled.

The single species in Calamonastides was formerly included in Chloropeta. Following Fregin et al. (2009), the other two former Chloropeta have been submerged in Iduna, which otherwise contains former Hippolais warblers. Earlier work suggested that the Thick-billed Warbler, Phragamaticola aedon, may not be closely related to the remaining Acrocephalus. In fact, Fregin et al. place it sister to Iduna. Recent practice (e.g., IOC, BOU-TSC) is to merge it Iduna, and that is done here.

Click for Acrocephalidae tree
Click for Acrocephalidae tree

Both Leisler et al. and Helbig and Seibold endorsed some of the traditional subgenera of Acrocephalus. Fregin et al. (2009) is consistent with this arrangement. The small streaked Acrocephalus become Calamodus while most of the small unstreaked Acrocephalus become Notiocichla. Although often placed elsewhere, the Large-billed Reed-Warbler belongs in Notiocichla (Bensch and Pearson, 2002). Iduna consists of former Hippolias warblers.

The Millerbird is sometimes considered as two species, Nihoa Millerbird, Acrocephalus kingi, and the extinct Laysan Millerbird, Acrocephalus familiaris. The genetic distance is in the species/subspecies borderline area (Fleischer et al., 2007). AOU currently treats them as one species, as I do here.

Cibois et al. (2011) were the first to include A. luscinius, syrinx, and rehsei in their analysis. They also examined a number of the extinct island taxa (see also Cibois et al., 2008). As a result, The Saipan Reed-Warbler, Acrocephalus hiwae, Pagan Reed-Warbler, Acrocephalus yamashinae, and Mangareva Reed-Warbler, Acrocephalus astrolabii are split from Nightingale Reed-Warbler, Acrocephalus luscinius. Also, the Leeward Islands Reed-Warbler, Acrocephalus musae and Moorea Reed-Warbler, Acrocephalus longirostris are split from Tahiti Reed-Warbler, Acrocephalus caffer. Note that most of these taxa are now extinct.

Donacobiidae: Donacobius Aleixo & Pacheco, 2006

1 genus, 1 species Not HBW Family

The monotypic Donacobius has variously been considered a wren, thrush, or mockingbird. It is none of these. It is a sylvioid of some sort, probably fairly closely related to the following two families. We place it next in its own family.

Bernieridae: Malagasy Warblers Cibois et al., 2010

8 genera, 11 species Not HBW Family

Bernieridae tree The new Malagasy warbler family, Bernieridae mostly comes from the Sylviidae, except for Bernieria itself, which comes from the Pycnonotidae (Cibois et al., 1999, 2001). Whether Rand's Warbler belongs in the group is uncertain.

Locustellidae: Grassbirds Bonaparte, 1854

13 genera, 57 species Not HBW Family

Another new family follows: the grassbirds. As with several of the other new families, the Locustellidae are comprised entirely of former Sylviidae. Indeed, many were members of the Megalurinae subfamily, so the name Megaluridae has been used for this family. This fails to properly account for the addition of Locustella to the group. The name Locustellidae, which is used by BOU, has priority over Megaluridae. In particular, Locustellidae dates to Bonaparte in 1854, while Megaluridae dates to Blyth (1875, p.117; see also Bock, 1994). Blyth's Megaluridae included Locustella, and may be the only such use for over a century.

Click for Locustellidae species tree
Click for Locustellidae
species tree

For a long time, both Megalurus and Locustella have been included in families with more senior names (e.g. Sylviidae, Turdidae, Muscicapidae). Megaluridae appears to have next seen the light of day (as Megalurinae) in Sibley and Ahlquist (1985a). As their Megalurinae did not contain Locustella, their use has no bearing on the correct name of this family as currently constituted. Alström et al. (2006) added Locustella to the Megaluridae, but it apparently did not occur to them that Locustella might bring its own family name with it. This was subsequently followed by a few authors (and this website), but Sangster et al. (2009) listed the family as Locustellidae. A little checking showed it had priority, and this website now follows their lead.

The Asian Bradypterus species were previously to Locustella using a combination of Drovetski et al (2004), guesswork based on geography and rumors of a paper by Alstöm and others (which I had not then seen). Alström et al. (2011b) confirms the previous restructuring of Locustella and Bradypterus. The current arrangement of species and genera is based on their study, which included about 2/3 of the species in Locustellidae. The species and genera not included are marked with a question mark or blue on the tree. Dromaeocercus has been merged into Bradypterus.

As Alström et al. remind us, some of the genera not tested may not belong to this family. They also note that several species show an excessive amount of genetic variation and likely contain two or more species. This is particularly true of Little Rush Warbler, Bradypterus baboecala. Certain other species pairs are very closely related, and may not truly be distinct biological species. These cases still need further sorting out, and potential changes are not reflected in the current species list.

Alström et al. (2011b) also found evidence that Megalurus is polyphyletic. However, the taxonomic changes they propose suggest they doubt this. The taxonomic changes implemented here take this evidence more seriously. The troublesome species is the Striated Grassbird, Megalurus palustris, which doesn't seem to group with the rest of Megalurus. Since it is the type species, that means the others need to change their genus.

The Tawny Grassbird is close to the songlarks, and I've moved it and the New Guinean grassbirds into Cincloramphus. It seemed appropriate to resurrect the genus Bowdleria (Rothschild 1896, punctata) for the New Zealand fernbirds. That and the phylogeny mean that the Littel Grassbird needs its own genus. The name Poodytes (Cabanis 1850) is available.

The position of the thicketbirds (Buettikoferella and Megalurulus) is speculative, and the position of the other untested genera more so. This is particularly true of Chaetornis.

The Marsh Grassbird, now Locustella pryeri, was originally moved from Megalurus based on Drovetski et al. (2004b). This was confirmed by Alström et al. (2011b).

Drovetski et al. (2004) and Alström et al. (2011b) found substantial genetic differences between the Sakhalin Grasshopper-Warbler, Locustella amnicola, and Gray's Grasshopper-Warbler, Locustella fasciolata. These had been suspected to be separate species, and so are split here.

Previous Page Next Page