Corvida II

Passerines

Tyranni: Suboscines

Passeri: Oscines

Passerida

Sylvioidea
Muscicapoidea and allies
Passeroidea

The 45 Orders

Paleognaths

Galloanserae

Metaves

Pelecanae

Charadriae

Passerae

Corvoidea

The Corvoidea form the inner core of the corvids. They are more closely related to each other than to anything else. Except perhaps for an isolated species or two, there's not any real question about which families, or even genera, belong to this core group. However, the ordering of the families has been rather obscure. This has improved with the comprehensive six-gene analysis of Jøonsson et al. (2011b), especially figure S1-B. This mostly resolves the Corvoidea, but support for some of the groupings is still a bit shaky.

Jøonsson et al. (2008b) put the Crested Jay near the shrikes. In contrast, Jøonsson et al. (2011b) show two alternative possibilities. They show on its own basal branch in figure S1-B, but nest it further inside the Corvoidea in S1-A. Due to this conflict, I've placed the Crested Jay incertae sedis at the base of the Corvoidea (with little confidence).

Putting the Crested Jay to the side, the drongos (Dicruridae) are the basal group in the Corvoidea, followed by the fantails (Rhipiduridae). The remaining splits into two parts: with the mudnesters (Corcoracidae), melampittas (Melampitta), and birds-of-paradise (Paradisaeidae) on one side and the monarchs (Monarchidae), shrikes (Laniidae), and jays and crows (Corvidae) on the other.

There is another interesting small change in the Corvidae besides the removal of the Crested Jay. The other ringer is Hume's Groundpecker (or Ground Jay), Pseudopodoces humilis. It was thought to be a corvid, the smallest of them. It is so listed even in Howard and Moore, but recent evidence (James et al., 2003; Gill et al., 2005) suggests it is actually a parid!

Incertae Sedis: Crested Jay

1 genus, 1 species HBW-14, in Corvidae

Dicruridae: Drongos Vigors, 1825 (1824)

1 genus, 25 species HBW-14

The ordering of the drongos is based on Pasquet et al. (2007).

Rhipiduridae: Fantails Sundevall, 1872

3 genera, 46 species HBW-11

Click for Rhipiduridae tree
Click for Rhipiduridae tree

Irestedt et al. (2008) found that the Silktail is not a monarch (or anything else it had been considered in the past). It is really a basal member of the fantails, most closely related to the Pygmy Drongo (which is not a Drongo).

The Yellow-bellied Fantail, formerly Rhipidura hypoxantha, has been moved as it is not a fantail (Nyári et al., 2009). It now takes the scientific name Chelidorhynx hypoxantha and is one of the Stenostiridae. The overall arrangement of the fantails is based on the Bayesian analysis in Nyári et al. (2009).

Corcoracidae: Australian Mudnesters Mathews, 1925

2 genera, 2 species HBW-14, as Struthideidae

Incertae Sedis: Melampittas

1 genus, 2 species HBW-12, in Eupetidae

Jønsson et al. (2011b) did not place the melampittas securely with either the mudnesters or birds-of-paradise. Their figure S1-A shows the latter while S1-B shows the former. In such a case, listing it incertae sedis right between the two appears to be the best choice.

Paradisaeidae: Birds-of-paradise Vigors, 1825

16 genera, 41 species HBW-14

Paradisaeidae tree The phylogeny here follows Irestedt et al. (2009a). Note that two species formerly placed in Ptiloris have been moved to Lophorina.

Monarchidae: Monarchs Bonaparte, 1854

16 genera, 98 species HBW-11

Monarchidae As mentioned earlier, Ifrita seems to belong with the monarchs. Norman et al. (2009a) found Ifrita sister to a clade containing Monarcha and Myiagra. Jønsson et al. (2011b), which has better taxon sampling and provides further support for it as a basal monarch flycatcher.

The genetics of the Monarchidae have been studied by Pasquet et al. (2002), Filardi and Moyle (2005), and Filardi and Smith (2005). It is clear that the traditional Monarcha itself is paraphyletic. Christidis and Boles (2008) discuss appropriate generic names. They have been applied here, breaking Monarcha into three genera: Carterornis, Monarcha, and Symposiachrus. Note that Metabolus has been absorbed into into the narrower Monarcha based on Filardi and Moyle (2005.

Since the papers above did not test all of the species in Monarcha, information from the species accounts in HBW-11 (del Hoyo et al., 2006) and family account (Coates et al., 2006) have been used to divide its species among the three genera. I'm a bit more uncertain about pileatus than the rest, so it is flagged in the list below.

As none of the papers took a comprehensive look at Monarchidae, the HBW-11 family account (Coates et al., 2006) was helpful preparing the tree. Ifrita and Melampitta are tentatively put first. Then there is a division between the Paradise-Flycatchers (plus Hypothymis) and the other monarchs. The position of Eutrichomyias is somewhat uncertain. Is it closer to Trochocercus or Terpsiphone?

The Monarcha branch splits into three main groups: Arses and Myiagra; Grallina; and Monarcha and allies. The taxonomic position of Grallina has been controversial, but Filardi and Moyle (2005) have resolved its position as sister to the Monarcha group.

Cibois et al. (2004) addressed the species status of the Pomarea Monarchs of the central Pacific. VanderWerf (2007) showed that the forms of Elepaio on different islands were relatively unresponsive to each other's songs, while VanderWerf et al. (2010) shows they are as genetically distinct as the species in the closely related genus Pomarea. The combination of reasonable genetic distance and evidence of biological separation results in a split of the Elepaio, Chasiempis sandwichensis, into 3 species: Kauai Elepaio, Chasiempis sclateri, Oahu Elepaio, Chasiempis ibidis, and Hawaii Elepaio, Chasiempis sandwichensis.

Laniidae: Shrikes Rafinesque, 1815

4 genera, 36 species HBW-13

Laniidae tree The gray shrikes have been the subject of several genetic studies, including those of Mundy and Helbig (2004), Gonzalez et al. (2008), Klassert et al. (2008), and Olsson et al. (2010). Although Olsson et al.'s analysis is the most comprehensive, it still leaves a fair amount of uncertainty.

Mundy and Helbig (2004), analyzing DNA from several gray shrike taxa, found that the North American form invictus, conventionally considered part of excubitor, actually seemed to be more closely related to ludovicianus. Both Gonzalez et al. (2008) and Klassert et al. (2008) provided support for this, and added the subspecies meridionalis to the invictus-ludovicianus clade. This rendered the specis meridionalis polyphyletic. Olsson et al. (2010) analyzed many more of the gray shrike subspecies. Two of the fiscals, the Somali Fiscal and Taita Fiscal, seem to belong to the gray shrike group. The Taita Fiscal is most distant from the others, which divide into two clades as shown in the table below.

Exactly how to divide these groups into species is less than clear, and Olsson et al. suggest several possibilities. The one I've taken divides the first clade into 3 species, the monotypic Socotra Shrike, Lanius uncinatus (split from L. meridionalis); the Desert Shrike, Lanius elegans (also split from L. meridionalis); and the Great Gray Shrike, Lanius excubitor. The latter has gained the remainder of meridionalis (except meridionalis itself). I have also merged the Steppe Gray Shrike back into excubitor. The DNA differences between the various subspecies of Desert and Great Gray Shrike are large enough that they could represent more species, but small enough that all of them could be lumped into Great Gray Shrike. There is limited other evidence available. There are plumage differences, but one of the Desert Shrike subspecies is thought to interbreed some with one of the Great Gray Shrike subspecies.

The other clade includes six species. The Chinese Gray Shrike has been split into two monotypic species: Chinese Gray Shrike, Lanius sphenocercus, and Tibetan Shrike, Lanius giganteus. The two not only differ in plumage, but the DNA differences are consistent with species status. The Northern Shrike, Lanius borealis, includes not only North American subspecies, but also the eastern Palearctic subspecies of the former Northern/Great Gray complex. Finally, the Southern Gray Shrike, which is closely related to the Northern Shrike, retains only one subspecies, meridionalis.

The Gray Shrikes
Species Subspecies
Taita Fiscal, L. dorsalis *dorsalis
Socotra Shrike, L. uncinatus *uncinatus
Desert Shrike, L. elegans *koenigi, *algeriensis, *elegans, *leucopygos
Great Gray Shrike, L. excubitor *excubitor, *homeyeri (inc. *leucopterus), *aucheri, theresae, *buryi, jebelmarrae, *pallidirostris, *lathora
Somali Fiscal, L. somalicus *somalicus
Loggerhead Shrike, L. ludovicianus *excubitorides (inc. gambeli & sonoriensis), migrans, ludovicianus, miamensis, anthonyi, mearnsi, grinnelli, *mexicanus (inc. nelsoni)
Chinese Gray Shrike, L. sphenocercus *sphenocercus
Tibetan Shrike, L. giganteus *giganteus
Northern Shrike, L. borealis *borealis (inc. invictus), *sibiricus, *bianchii, *mollis, *funereus
Southern Gray Shrike, L. meridionalis *meridionalis
Subspecies based on Dickinson et al. (2003).
*An asterisk indicates taxa sampled by Olsson et al. (2010).

More recently, Fuchs et al. (2011c) have studied some of the fiscals. Their results suggest (1) the Uhehe Fiscal, Lanius marwitzi, is embedded in the Common Fiscal and should be merged (2) the Common Fiscal contains two species. The new alignment results in Northern Fiscal, Lanius humeralis (smithii, humeralis, capelli) and Southern Fiscal, Lanius collaris (aridicolus, pyrrhostictus, marwitzi, subcoronatus, collaris). Although the genetics appear to support this, it would be nice to see additional work backing it up, especially in the area where the two groups meet.

At this point, there is finally enough information for a preliminary reorganization of Lanius. The Masked Shrike seems relatively basal, as does the Gray-backed/Long-tailed Fiscal pair. The Woodland/Lesser Gray Shrike may group together. There's a core fiscal group, tentatively newtoni through gubernator, a brown shirke group, speculatively including tigrinus through validirostris, and the well-supported gray shrike group running from dorsalis on.

Corvidae: Crows, Jays Leach, 1820

25 genera, 128 species HBW-14

Corvidae tree The corvid genera are arranged based on Ericson et al. (2005) and, for the New World jays, Bonaccorso and Peterson (2007). I'm treating the natural division into 5 clades as subfamilies. A bit different arrangement is used by Ekman and Ericson (2006), but it includes fewer taxa. There is some ambiguity about the position of Perisoreinae, Cyanocoracinae, and Corvinae, so I have left that unresolved in the tree. This ambiguity doesn't seem to affect the ordering of species.

Based on van Balen et al. (2012), the two allopatric forms of Short-tailed Green-Magpie are split into Bornean Green-Magpie, Cissa jefferyi, and Javan Green-Magpie, Cissa thalassina. Based on plumage, morphology, and voice, they argue that the differences between them are at a level typical of distinct species.

I've split the Western Scrub-Jay into Woodhouse's Scrub-Jay, Aphelocoma woodhouseii, and California Scrub-Jay, Aphelocoma californica, based on Rice et al. (2003) and Delaney et al. (2008). This split failed to get the necessary 2/3's majority from AOU's NACC in 2009 (6-5 in favor). A further split of sumichrasti lost 4-7. I've also split the Mexican Jay into Transvolcanic Jay, Aphelocoma ultramarina, and Mexican Jay, Aphelocoma wollweberi, as in McCormack et al. (2008). It seems likely from their results that A. wollweberi will need at least one further split.

Bonaccorso (2009) was the primary source concerning the arrangment of Cyanolyca. The Calocitta to Cyanocorax clade is based on maximum parisomy and mitochondrial trees in Bonaccorso et al. (2010). I've separated several of the South American jays into Uroleuca (Bonaparte, 1850). This allows retention of Calocitta for the distinctive magpie-jays and lets me restrict Psilorhinus to the Brown Jay. This is consistent with the maximum parsimony tree they found based on 5 genes. (The MP tree is described but not shown in the paper.)

Only one gene was used for the Azure Jay, Uroleuca caerulea. It ended up sister to the Brown Jay, but this was only weakly supported. Bonaccorso et al. describe the situation as a “virtual polytomy”. It doesn't make much sense that the Azure Jay of extreme southern Brazil should be sister to the distinctive Brown Jay of Middle America, especially when there are similar jays right next door to the Azure Jay. Since the mitrochondrial tree yields a virtual tie, I've used geography and plumage to break it, placing the Azure Jay with the Uroleuca jays. I should also note here that as part of the polytomy, Bonaccorso et al. found apparent paraphyly in the Violacenous Jay, with the Ecuadorian birds basal relative to those from Peru.

Since Bonaccorso et al. found substantial genetic distance between South American and Middle American green jays. These two groups are separated by Panama, Costa Rica, and Nicaragua; they look obviously different; vocalizations are different; there is substantial genetic distance. While they could be the same species, I think the balance of evidence favors splitting the 8 Middle American races as Green Jay, Cyanocorax luxuosus, and the South American races as Inca Jay, Cyanocorax yncas. This is how Hilty (2003) and Ridgely and Greenfield (2001) treat them, even though the AOU has not yet done so.

Four of the Cyanocorax jays, melanocyaneus through beecheii are sometimes separated as the genus Cissilopha (Bonaparte, 1850). Although they form a clade, they are nested within Cyanocorax.

The genus Corvus remains confusing as no broad sampling has been done. However, Kryukov and Suzuki (2000) and Haring et al. (2007a) found that the jackdaws were rather distant from the other Corvus species. Accordingly, they are separated in the genus Coloeus, as recommended by Rasmussen and Anderton (2005). Although I had presumed that Collared Crow took the name Corvus torquatus because of this, thinking that otherwise Corvus monedula torquatus (Bechstein 1791) would have first claim on torquatus, this is apparently incorrect. Rather, Bechstein did not use the term to designate a species or subspecies. This did not make the name available as a species-group name, and so Lesson's use of Corvus torquatus for the Collared Crow in 1831 takes precendence over Corvus pectoralis for the Collared Crow. See BirdForum for more.

There is also evidence that the Chihuahuan Raven, Corvus cryptoleucus, and possibly the Pied Crow, Corvus albus, are embedded in the Common Raven complex. Moreover, the Chihuahuan Raven seems more closely related to a mostly Californian clade of Common Ravens than either is to the other Common Ravens. See Feldman and Omland (2004) and Omland et al. (2000, 2006).

What seems to have happened is that the California population became separated from the other Common Ravens. During that time, the Chihuahuan Ravens separated from the California Common Ravens. Barriers to interbreeding developed between them, making Chihuahuan Raven a separate species. Later, the California Common Ravens came back into contact with the northern Common Ravens. However, they were still able to interbreed, and do so freely to this day (Webb et al., 2011), leaving only a mitochondrial signature of their divergent ancestry.

The Crested Jay, Platylophus galericulatus, has been placed incertae sedis in Corvoidea.

Pyrrhocoracinae: Choughs and Treepies G.R. Gray, 1846 (1831)

Cissinae: Green and Blue Magpies Kaup, 1855

Perisoreinae: Northern Jays Informal?

Cyanocoracinae: American Jays Kaup, 1855

Corvinae: Crows and Palearctic Jays Leach, 1820

Previous Page Next Page