THE BREAK-UP OF THE
OLD WORLD WARBLERS
A DISCUSSION OF THE 'NEW' TREE
Let's look at our nice oak tree a little closer to establish where we are. We might think of the trunk of the tree branching in two: to Non-passerines and to Passerines. It is actually more complicated that this with many more major 'trunks' in the Non-passerines but a simple passerine/non-passerine division does split the ~10,000 or so bird species in the world in halves (more or less). Let's go up the Passerine trunk. First (in the yellow arrows) is a split between suboscine and oscine passerine birds. The Suboscines include such bird families as pittas, broadbills, New World flycatchers, cotingas and others. We will not consider them further in this discussion. The blue arrows show two big important limbs: the corvoids, which arose in Australasia and include everything from lyrebirds to crows, and the non-corvoids. The corvoids include a gradation of groups not yet well understood, including a "basal corvida" and a "crown corvida" (to use Jønsson & Fjelså (2006) terms), and some other offshoots not closely related (e.g., New Zealand Wrens) but, again, we will not follow this route. Rather, following that non-corvoid path, we come to a major three-way division of the remaining Passerines: Muscicapoidea, Sylvioidea, and Passeroidea (from Jønsson & Fjeldså 2006, who summarized 99 important genetic papers). There appears to be consensus of all these major branches.

The Old World Warblers were an important part of the Sylvioidea background color of light blue here and our 'new' tree and the discussion following will focus on this major group of birds. Recall that we 'read' the tree from the bottom up. Also you should know that when two or more 'branches' take off from the same spot in the tree, the left to right reading of the branches is entirely arbitrary. The 'real' tree is three-dimensional. I listed the 3 major limbs of non-corvoid passerines as "Muscicapoidea, Sylvioidea, and Passeroidea," but they could be rearranged as "Passeroidea, Muscicapoidea, and Sylvioidea" or any other order, as long as the families and species within each group is kept together. In the taxonomic tree below, it is important to keep this feature in mind. If 2 or more branches [shown in orange on the 'new' tree] start at the same spot, the order in which those branches are arranged is arbitrary. But if the branches start at different spots on the tree, that is important information. The lower branches represent evolutionary lineages that began earlier in evolutionary history.

A. Reading from the bottom up, and recognizing that we are looking only at the Oscine Passerine trunk of the tree, we see the first major branch goes to the corvoid assemblage, including a "crown" Corvida a huge set of bird families that include crows,, birds-of-paradise, shrikes, and vireos. Here is our first major surprise. The bird know as White-bellied Yuhina and even listed among the genus Yuhina within the Babblers is not even close to babblers! It should be called White-bellied Erpornis  Erpornis zantholeuca and assigned to the Corvida (Cibois 2003). It is most closely related to Vireo;, I've assigned it to its own family on my family listing for the time being. Whether other, unstudied "babblers" or other birds belong here is yet to be determined.

B.  Next thing up the tree is a side-branch to the Rockjumpers, endemic to South Africa. [Cape Rockjumper is shown; hereafter, please refer to the discussion of the "old" tree for a key to photographs]. The Rockjumpers were usually listed as Babblers or Thrushes but they are not close to either. There are their own distinct group that diverged from the main Passerine tree about 45 million years ago (Barker et al. 2004, Cracraft et al. 2004). The Picathartes Rockfowl (2 species in tropical Africa; not pictured) also diverged at this time, and the split from the Rockjumpers about 31 million years ago (Beresford et al. 2005). It is obvious to me that each deserve their own family ranking. Two other strange 'babblers' or 'thrushes' Spot-throat Modulatrix stictigula and Dapple-throat Arcanator orostruthus, from the Tanzania mountains [not pictured] remain enigmatic. Some studies suggest they are most closely related to the Sugarbirds of South Africa [Promeropidae] but others found that "although it was not statistically rejected ... additional data ... are clearly needed in order to evaluate the validity of this 'clade' of very different-looking birds" (Fuchs et al. 2006).

C. Slightly up the Passerida tree are two small branches, to Kinglets and to Hyliotas. Kinglets seem to "jump around" in the taxonomic tree, depending on the methodology used, and so are just tentatively placed here, near the base of the Passerids, pending further research (Jønsson & Fjeldså 2006). The Hyliotas of Africa have recently been shown to be a very isolated deep branch of the Passerida (Fuchs et al. 2006); it seems likely they will achieve family rank if these data are confirmed.

D. We now come to the major three-way division of the Passerines into Muscicapoidea (green background), Sylvioidea (blue background), and Passeroidea (yellow background). There are many new research findings of interest in each of these three groups but, for this discussion, we focus only on those that impact the 'old' Sylviidae family. The Gnatcatchers are not Sylvids but are closer to Wrens and Nuthatches, and these families (along with Dippers) are on the Muscicapid branch, and thus not close to the Tits & Chickadees (Paridae) as they were thought to be before.

E. Finally our tree climbs into the Sylvioidea, the group of birds once dominated by the Old World Warblers. Jønsson & Fjeldså (2006) recognized 13 branches (="clades") in this group of birds, and the numbers on this tree are the same as the numbers of the Sylvioidea "clades" in their paper. They recognize that many of these are poorly resolved, and especially the uppermost branches, clades 9-13 [note that I have clades 12 & 13 on the same orange-colored branch; that will be explained later in this discussion]. Some of the clades represent familiar families: Larks (clade 3), Swallows (clade 6), and Bulbuls (clade 9, but now missing the two special groups pictured among the Bulbuls in the "old" tree). Some of these clades are well supported but others (e.g., clades 2, 7, 11) are not.

However, some of our familiar families are now "sub-clades". For example, the branch leading to Tits & Chickadees (clade 2) now also has the Penduline-Tits (including Verdin of North America) and a third group, the Stenostirids. Within clade 2, each of these three groups forms a monophyletic group of birds on their own evolutionary branch. Taxonomists often like to think of themselves as "rank-free." They are interested in the finding that these three groups are monophyletic, but don't care whether each of the three groups is called a Family or not.  It is equally correct to list the three groups as 3 families Paridae, Remizidae, and Stenostiridae or to lump the entire clade together as just one family. If the latter, the Paridae would become a much bigger family that includes the penduline-tits, Verdin, and all the Stenostirid 'flycatchers' that are not flycatchers at all.
   I much prefer the former approach for both practical and historical reasons. Stating that there are three families in clade 2 violates no rule of taxonomy but does preserve the historic and well-entrenched, familiar families (Tits & Chickadees on the one hand, Penduline-Tits on the other hand) and allows us to recognize the previously overlooked group of African & Asian 'flycatchers' in their own right, undiluted by titmice and Verdins. This is exactly what Beresford et al. (2005) proposed: that the Stenostiridae be accepted as its own family.

In my layout, I have picked out those monophyletic groups that seem to me to warrant recognition as Families, whether they are "clades" or "sub-clades" in Jønsson & Fjeldså (2006). These prior members of the Old World Warbler family now appear in ten different clades or subclades, representing 7-10 or more different families, not even counting the two (Kinglets, Hyliotas) that now fall outside this group completely! The 13 clades of the Sylvioidea are discussed next. However, the fact that Jønsson & Fjeldså (2006) recognized 13 clades and not 10 or 16 or whatever was entirely arbitrary and was as much influenced by layout of the paper as anything (Knud Jønsson, in litt.). So not too much emphasis should be placed on any apparent different between "clades" and "subclades."

  • Clade 1 is just one species, the Bearded Reedling Panurus biarmicus, of the Palearctic. Its true position is still uncertain (Alström et al. 2006, Jønsson & Fjeldså 2006) but it is not a traditional parrotbill, as had previously been thought. Whether it deserves family rank is not yet known [I do so in my family listing provisionally, because it 'must' appear somewhere for world birding purposes] and further research may show it is actually closer to the other parrotbills than this assignment might suggest. 
  • Clade 2 is discussed just above, and includes the Paridae and Remizidae. The most interesting finding is that some African & Asian 'flycatchers' are not flycatchers at all but within this sylvioid assemblage. The Stenostirids include Fairy Flycatcher Stenostira scita (shown), various crested-flycatchers & blue-flycatchers assigned to Elminia or Trochocercus, and two species of canary-flycatcher [Culicicapa] from Asia. Support for including these among clade 2 is not strong, and further research may alter the tentative arrangement here.
  • Clade 3 is the Larks [Alaudidae]; nothing new here except than some genera (e.g., Mirafra, Eremalauda) prove to be polyphyletic and need revision
  • Clade 4 is a newly discovered lineage of African warblers & crombecs [Beresford et al. 2005; these could be called the Sylviettidae] which include the Sylvietta crombecs, Damara Rockrunner Achaetops pychopygius, Cape Grassbird Sphenoeascus afer, Victorin's Scrub-Warbler "Bradypterus" victorini, Yellow Longbill Macrosphenus flavicans, Moustached Grass-Warbler Melocichla mentalis, and probably others as yet unreviewed. 
  • Clade 5 is composed the three Nicators of Africa (although only Yellow-spotted Nicator N. chloris has been reviewed genetically)
  • Clade 6 is the Swallows; genetics confirm that the two river-martins (one in s.e. Asia, one in central Africa) are quite remote from the rest of the world's swallows
  • Clade 7 is an interesting group of 2 or 3 subclades.
    • The most remote best suited to remain as a Family, in my view are the Long-tailed Tits [Aegithalidae], which include the Bushtit of the New World. The two Leptopoecile tit-warblers of China are in this Family (having observed both in China, this makes good sense to me). Alström et al. (2006) call this "clade G" in their study and, while confirming it is near the next group to be discussed, propose that the family name remain as Aegithalidae. This is an example of how different taxonomists, using more or less the same information, can differ is what is a "clade" and what is a "subclade" within a clade.
    • The other main subclade here are the Cettid Warblers & allies. These include the 15 or so species of Cettia bush-warblers [Cettia proves to be polyphyletic] and an eclectic group of odd warblers: Mountain Tailorbird "Orthotomus" cucullatus [most of the other Orthotomus tailorbirds are in a different clade], the almost tail-less Tesia warblers and Asian Stubtail Urosphena squameiceps, and Broad-billed Warbler Tickella hodgsoni. Close to this subclade are Asian warblers in genus Abroscopus, the African 'flycatcher' Erythrocercus, Green Hylia Hylia prasina, and Tit-Hylia Pholidornis rushlae. Alström et al. (2006) propose that this group ("clade H" in their study, which did not include all the taxa in Jønsson & Fjeldså 2006) be formally named as the family Cettiidae. The complete parameters of this family have yet to be worked out.
  • Clade 8 is composed of the 68 or so leaf-warblers assigned to the genera Phylloscopus and Seicercus (Seicercus is 'embedded' within Phylloscopus; there will be generic name changes in due course). Alström et al. (2006) formally propose that this clade [termed "clade F" in their study but composed of the same species] be named as the family Phylloscopidae. So far, all the members are currently assigned to Phylloscopus or Seicercus.
  • Clade 9 is the Bulbuls, a nice monophyletic group once the two non-bulbul groups are removed [Nicators to clade 5 and Madagascar species to clade 11]. A major study on the genetics of this family will be published in the near future (Moyle & Marks in press) that confirms this is a monophyletic family but shows that some of the relationships within the family are different than previously assumed.
  • Clade 10 is the family Cisticolidae [Cisticolas & allies]. This family was recognized by some recent world checklists (e.g., Dickinson 2003); some 110 species were among them, including 50+ species of cisticolas, genus Cisticola, and ~60 species of Prinia, Apalis, & allies. To this set can now be added most of the Orthotomus tailorbirds. Mountain Tailorbird proved to be in clade 7, among the Cettiidae, and possible a couple of other tailorbirds go there as well. The remaining dozen or so species now go to the Cisticolidae (to date the species known to be here are Common Tailorbird O. sutorius and Dark-necked Tailorbird O. atrogularis; Alström et al. 2006). Also added to this family are 3 of the Neomixis Jerys from Madagascar (Cibois et al. 1999), and three odd warblers from Africa: Oriole Warbler Hypergerus atriceps, Gray-capped Warbler Eminia lepida, and Kopje Warbler Euryptila subcinnamomea (Beresford et al. 2005)
  • Clade 11 is a newly discovered group that may include up to 4 subclades. It is possible that each of these subclades should be treated as a family; it is also possible that additional research will suggest other arrangements. Two of the subclades have already been proposed to be families by Alström et al. (2006): the Megaluridae [Grassbirds & allies] and the Acrocephalidae [Acrocephalid warblers; these were clades "C" and "D" of their study]. This leaves unresolved two other groups. The full set are:
    • the 40+ species of Acrocephalus & Hippolais warblers, plus a couple of Chloropeta yellow warblers in Africa, and the Streaked Scrub-Warbler Scotocerca inquieta of the s. Palearctic (but not yet well-supported). Excepting for some questions on the latter species, this is the well-defined group proposed to be the Acrocephalidae.
    • a largish group of grass or marsh-loving skulkers, which will be known as the Megaluridae. Among them are Megalurus grassbirds, the Schoenicola broad-tailed & fan-tailed warblers, the 9 species of Locustella grasshopper-warblers, the Cincloramphus songlarks of Australia, and most of the 20+ species of Bradypterus bush-warblers & scrub-warblers.
    • related to the Megaluridae is Black-capped Donacobius of South America. Barker (2004) showed it was not a wren or related to any other New World but was, instead, a sylvoid an ice-age remnant, if you will, long isolated in the Amazon Basin. Its exact position is uncertain as studies are incongruent (Alström et al. 2006). The three other sylvoids to reach the New World do so only near the once-extant Bering landbridge. Arctic Warbler Phylloscopus borealis breeds in both Siberia & Alaska, but the Wrentit Chamaea fusciata (in clade 12) and Bushtit Psaltriparus minimus (in clade 7) are restricted primarily to the western United States, where isolated after the Pleistocene ice age. The Wrentit diverged from its babbler ancestors only 6.58 million years ago (Burns & Barhoum 2006). More data are needed, but it seems likely the Donacobius diverged and was isolated from its Old World relatives much longer ago than that, and therefore likely warrants Family rank [more on this topic below; it is more philosophy than taxonomy]
    • finally, and perhaps most fascinating of all, there is a radiation of birds in Madagascar that fit in here, either as a family (perhaps called Bernieriidae?) or a subfamily. It makes much better sense to me to rank them as a family. In this group are birds originally assigned to Old World Warblers Wedge-tailed Jery Neomixis flavovirdis, Thamnornis Warbler T. chloropetoides, and the recently described Cryptic Warbler Cryptosylvicola randrianasoloi plus birds assigned to Babblers White-throated Oxylabes Oxylabes madagascarienses, Yellow-browed Oxylabes Crossleyia xanthoprys plus birds once considered Bulbuls Long-billed "Greenbul" Bernieria madagascariensis, and 4 species of Xanthomixis "greenbuls," all now known as Tetrakas.
  • Clade 12 is an interesting group that includes the 18 species of Sylvia & Parisoma warblers [Parisoma should likely be subsumed into Sylvia], the Wrentit Chamaea fasciata of North America, and the 18 species of Paradoxornis parrotbills plus Conostoma [Great Parrotbill]. This wipes out the Parrotbill family for good (and remember that Bearded Reedling apparently turns out not to be a parrotbill, but an odd lineage in clade 1). In addition, various babblers go here: Golden-breasted Fulvetta "Alcippe" chrysotis, Streak-throated Fulvetta "Alcippe" cinereiceps, Yellow-eyed Babbler Chrysomma sinense, African Hill-Babbler "Illadopsis" abyssinica, and the White-browed Chinese Warbler Rhopophilius pekinensis, usually listed in the Old World Warblers. This clade can be thought of as the Sylvia babblers, parrotbills, and allies.  If the clade is given Family rank, the name "Sylviidae" is available to represent this group, now much reduced from the previous Sylviidae.
    • There is a problem here. Although Jønsson & Fjeldså (2006) assign this group to their own clade, it is a sister group to the final clade, the remaining Babblers [Timaliidae]. Alström et al. (2006) combined these clades 12 & 13 into just one clade (called "clade E" of their study, although they did recognize 12 & 13 as clades as "E1" and "E2" in their work) and proposed that the single, enlarged group be called the Timaliidae. They said "Cibois (2003) suggested that if the Timaliidae and several groups of warblers are recognized at the same family level, then Sylviidae (Leach, 1820) should be suppressed and the name Timaliidae (Vigors and HorsWeld, 1827) kept for the babblers and Sylvia. We agree with this, and consequently propose that the name Timaliidae refers to clade E of the present study. However, formal suppression of Sylviidae can only be sanctioned by the International Commission on Zoological Nomenclature."
    • It may be too early to suggest suppression of Sylviidae, at least while it looks like Jønsson & Fjeldså's "clade 12" is a useful, monophyletic group that could use that name. This will all be worked out in time. But it is possible that Babblers/Sylvia warblers could become two families: Typical Babblers [Timaliidae] and Sylvid Babblers [Sylviidae or some other name, if Sylviidae were suppressed].
  • Clade 13 is the final sylvioid group, and it is the remaining Babblers [Timaliidae]. As discussed just above, it may or may not include the "Sylvid Babblers" of clade 12. Standing alone as clade 13, it is composed of 50+ species of Garrulax laughingthrushes, all the Zosterops white-eyes that have been studied [the family Zosteropidae bites the dust], plus hundreds of "typical" babblers of many genera, which are known to include some of the Alcippe babblers (others in clade 12), the Yuhina babblers (minus White-bellied Erpornis), and at least these genera: Stachryris, Babax, Turdoides, Cutia, Actinodura, Minla, Liocichla, Heterophasia, Leiothrix, Pomatorhinus, Illadopsis (minus African Hill-Babbler), Xiphirhynchus, Macronous, Timalia, Spelaeornis, Graminicola, Malacopteron Gampsorhynchus, Pellorneum, Jabouileia, Napothera, Malacocincla, and Kenopia [some of these genera are polyphyletic and new assignments among the genera will occur in due course]. There are quite a number of other genera that have yet to be genetically analyzed, including Malia, Neolestes, and Horizorhinus.
A couple of comments:
  • This is all very new and while some of the new information is well supported, other implications are less supportable on current knowledge. It is beyond doubt that the Old World Warbler family as we knew it the huge Sylviidae assemblage is gone forever, splintered into perhaps 10 or more new families, and with bits and pieces landing far outside. But our 'new' tree is tentative, subject to significant change, and needs more research.
  • I'm not an ornithologist ... but I play one one the Internet. :-)  Seriously, folks, when the topic becomes taxonomic philosophy ("art") and not systematics (science), then I'm happy to add my opinion to the debate. If you've read all the stuff above, you've seen where this can happen: whether a lineage is a "clade" or a "subclade" is often a philosophical call, and (as we have seen) different scientists can call it differently. The same goes for what is a "Family" or just a "subfamily." There is a lot of interest in birds at the Family level by amateur birders (not as much interest as in the Species level decisions, but still considerable interest).
  • In making Family level decisions, and if all things are equal*, I favor more Families over fewer Families it adds excitement and interest to world birding and I favor retaining Families that have historic meaning. In other words, keep the old families alive unless the genetic evidence shows that they are embedded within another family [the white-eyes and parrotbills are clearly embedded within other groups, so 'good-bye' to them]. Likewise, newly discovered evolutionary lineages should be given Family level rank when they have been long isolated and are comparable to current Family level decisions. Examples include the recently discovered radiation in Madagascar what a fabulous discovery! Let's not 'bury' it within a newly minted Megaluridae, and undercut its significance, when (as is true) the evidence shows that elevating it to Family rank breaks no taxonomic rules. Without going into more detail, I favor retaining families such as the Hypocolius, Palmchat, Bristlehead, and Ibisbill (all of which can be retained, as I understand it, without breaking any rules) and elevating to Family rank long isolated birds on the 'other side of the world' from their closest relatives, such as Donacobius, Olive Warbler, and Erpornis (assuming that subsequent research confirms current implications). We happen to live at a time of great discoveries in evolutionary history coincident with the ability to travel to the places these new lineages have been found. What a fine opportunity!
* I am aware that some naturalists are of the opinion there are already too many bird families, compared to the level of diversity at the Family level in, say, fish or beetles. I'm not convinced that there needs to be a 'universal' definition for such a man-made concept as "Families" across the spectrum of life on earth, nor am I convinced that the approach of, say, ichthyologists is necessarily the preferred one. The evolution of birds spans "only" 80 million years while some beetle families have existed for over 100 million years. Comparing families of birds to families of beetles is comparing "apples to oranges," as the old saying goes.
BACK TO
THE
INTRODUCTION
Click Here
BACK TO DISCUSSION
OF THE
OLD TREE
Click Here
Literature cited:
Alström P., P.G.P. Ericson, U. Olsson, and P. Sundberg. 2006. Phylogeny and classification of the avian superfamily Sylvoidea. Molecular Phylogenetics & Evolution 38: 381-397.

Barker, F.K. 2004. Monophyly and relationship of wrens (Aves: Troglodytidae): a congruence analysis of hererogeneous mitochondrial and nuclear DNA sequence data. Molecular Phylogenetics & Evolution 31: 486-504.

Barker, F.K., A. Cibois, P. Schikler, J. Feinstein, and J. Cracraft. 2004. Phylogeny and diversification of the largest avian radiation. Proc. Nat. Acad. Sci. 101: 11040-11045.

Beresford, P., F.K. Barker, P.G. Ryan, and T.M. Crowe. 2005. African endemics span the tree of songbirds (Passeri): molecular systematics of several evolutionary 'enigmas'. Proc. R. Soc. B 272: 849-858.

Burns, K.J., and D.N. Barhoum. 2006. Population-level history of the wrentit (Chamaea fasciata): implications for comparative phylgeography in the California Floristic Province. Molecular Phylogenetics & Evolution 38: 117-129.

Cibois, A., E. Pasquet, and T.S. Schulenberg. 1999. Molecular systematics of the Malagasy babblers (Timaliidae) and Warblers (Sylviidae), based on cytochrome b and 16S rRNA sequences. Molecular Phylogenetics & Evolution 3: 581-595.

Cibois, A., B. Slikas, T.S. Schulenberg, and E. Pasquet. 2001. An endemic radiation of Malagasy songbirds is revealed by mitochondrial DNA sequence data. Evolution 55: 1198-1206.

Cibois, A. 2003. Mitochondrial DNA phylogeny of babblers (Timaliidae). Auk 120: 35-54.

Cracraft, J., F.K. Barker, M. Braun, J. Harshman, G.J. Dyke, J. Feinstein, S. Stanley, A. Cibois, P. Schikler, P. Beresford, J. García-Moreno, M.D. Sorenson, T. Yuri and D.P. Mindell. 2004. Phylogenetic relationships among modern birds (Neornithes): toward an avian tree of life. Pp. 468-489 in J. Cracraft & M. J. Donoghue, eds. Assembling the Tree of Life. Oxford University Press, New York.

Dickinson, E.C., ed. 2003. The Howard & Moore Complete Checklist of the Birds of the World. 3d ed. Princeton Univ. Press, Princeton, N.J.

Ericson, P.G.P., M. Irestedt, and J.S. Johansson. 2003. Evolution, biogeography, and patterns of diversidfication in passerine birds. Jour. Avian Biol. 34:3-15.

Fuchs, J., J. Fjeldsa, R.C.K. Bowie, G. Voelker, and E. Pasquet. 2006. The African warbler genus Hyliota as a lost lineage in the Oscine songbird tree: molecular support for an African origin of the Passerida. Molecular Phyolgenetics & Evolution 39: 39: 186-197.

Jønsson, K.A., and J. Fjeldså. 2006. A phylogenetic supertree of oscine passerine birds. Zoologica Scripta 35: 149-186.

Moyle, R.G., and B.D. Marks. In press. Phyogenetic relationships of the Bulbuls (Aves: Pycnonotidae) based on mitochrondrial and nuclear DNA sequence data. , Molecular Phylogenetics & Evolution

Sibley, C.G., and J.E. Ahlquist. 1990. Phylogeny and Classification of Birds: A Study in Molecular Evolution. Yale Univ. Press, New Haven, CT.

Sibley, C. G., and B.L. Monroe, Jr. 1990. Distribution and Taxonomy of Birds of the World. Yale Univ. Press, New Haven, CT.

TOP
TO LIST OF FAMILIES OF THE WORLD

TO HOME PAGE

Page created 6 June 2006, revised 26 June 2006.