The Passerida are the sister group of the Corvida. There are several basalgroups that branch off before we get to the heart of the Passerida. The taxonomyof Passerida has been changing rapidly and most current checklists lag wellbehind the latest research.
Exactly which families are basal Passerida is still being clarified, withfamilies being shuffled between the Passerida and Corvida.The latest on this is the paper by Irestedt and Ohlson (2008), which makes astrong argument that several families thought to be Corvida are actually basalPasserida. Irestedt and Ohlson identified a genetic marker that seemsto separate the Passerida from the Corvida. If this is correct, it means thattaxonomies that rely primarly on the RAG-1 gene (e.g., Baker et al., 2004), arenot entirely correct. Their paper is a reminder of the importance of using several independent genes to test taxonomic hypotheses. There areother portions of the avian tree where this standard has not yet been met;those areas are particularly subject to change.
We start with a group of New Guinea endemics that had formerly been considered Passerida, the berrypeckers (Melanocharitidae). Sibley and Monroe (1993)listed them in the Passeroidea, near the flowerpecker and sunbirds.They include the former honeyeatersOedistoma andToxorhamphus.
The satinbirds (Cnemophilidae) are the second branch moved from theCorvida. These New Guinea endemics were once consideredbirds-of-paradise (Paradisaeidae), but now have their own family.
After two groups of New Guinea endemics, it's time for a change.The Callaeoidea are New Zealand endemics. They include the New Zealandwattlebirds (Callaeidae) and the stitchbird (Notiomystidae). Themonotypic stitchbird was previously considered a honeyeater, but isactually most closely related to the wattlebirds (Ewen et al., 2006;Driskell et al., 2007). The division between the Stitchbird and otherfamilies in its clade seems quite ancient, so it is given its ownfamily.
The three wattlebird genera apparently diverged almost simultaneously(Shepherd and Lambert, 2007), and are treated as a trichotomy.
Following OSNZ, IOC, and HBW-14, the Kokako,Callaeas cinereus, is split into North Island Kokako,Callaeas wilsoni, and the recently extinctSouth Island Kokako,Callaeas cinereus. See also Double and Murphy (2002).Further, the Saddleback,Philesturnus carunculatus is split intoNorth Island Saddleback,Philesturnus rufusater, andSouth Island Saddleback,Philesturnus carunculatus.
The next group is yet another Australasian group, the Australasianrobins (Petroicidae). The genetic studies that have been done have been somewhatequivocal on its placement—whether it goes before or after thePicathartoidea. Jønsson and Fjeldså (2006a) argue thatplacing it first makes the best sense both genetically andbiogeographically. Genetically, the rest of the Passerida share aninsertion in the nuclearc-myc gene that the Petroicidae lack(Ericson et al., 2000). The biogeographic sense is that all of thePetroicidae are Australasian. It seems unlikely the remaining Passerida woulddevelop in Africa (the Picathartoidea), then suddenly jump back toAustralia before returning to Africa. Using a different set of genes,Irestedt and Ohlson (2008) also put the Australasian robins beforethe Picathartoidea.
Differences from standard arrangments are based on Loynes et al. (2009).Some information from Miller and Lambert (2006) has been includedforPetroica itself. Following Loynes et al.,Tregellasia hasbeen merged intoEopsaltria and the Yellow-bellied Robin has moved fromEopsaltria toMicroeca. Surprisingly, they foundPeneoenanthe nested inPeneothello. Although they were relectantto recommend merging them, I've done so here. Several genera,Eugerygone,Monachella, andAmalocichla, were not included in their analysis.In the case ofAmalocichla we are able to turn to the Norman et al. (2009b).It shows that theAmalocichla must be more basal than the other Petroicidae that have been tested. Unfortunately, they somehowforgot to include any members of Passerida outside of Petroicidae. Apparentlythey were too fixated on the corvids. This lapse means that we can't becertain thatAmalocichla belongs in Petroicidae! They didn'trule out the possibility of it being close toPasser domesticusor the Stitchbird. However, all of the other alternatives seem fairlyunlikely, and we retain it in Petroicidae, but mark it as a bituncertain on the tree.
The Picathartoidea are the next group to branch off separately fromthe Passerida tree, and did they ever branch! Somewhere in the gapbetween the Petroicoidea and Picathartoidea the passerids made the leapout of Australasia all the way to Africa. Which way they went, we arenot sure. They may have come across the now-submerged Kerguelen Plateau.The Picathartoidea (rockfowl and rockjumpers) are both African endemics.They were considered Incertae sedis by Sibley and Monroe (1993) andthere has been some discussion about whether they are best considered acorvid relative or part of Passerida. Several recent papers have madethe case for placing them here (Barker et al., 2004; Beresford et al.,2005; Cracraft et al., 2004; Irestedt and Ohlson, 2008). Of course,once we have placed the Petroicidae in the Passerida, the choice isforced, Picathartoidea must also be in the Passerida. Recent research byJønsson et al. (2007) found that the Rail-babbler belongs in thesame clade. Each is distinct enough, and the genetic separation largeenough, that we consider each a separate family in thePicathartoidea.
This small group of African endemics was considered part of Sylviidaeuntil recently. Current evidence suggests they are not particularly closelyrelated to the Sylvioidea at all, nor do they seem particularlyrelated to any clade in Passerida. However, they are Passerida—andmore deeply so than the Australasian Robins (Fuchs et al., 2006a).So we put them here in their own superfamily.
