Heliornithidae: FinfootsG.R. Gray, 1840

3 genera, 3 speciesHBW-3

Heliornithidae (Finfoot) tree

• African Finfoot,Podica senegalensis
• Masked Finfoot,Heliopais personatus
• Sungrebe,Heliornis fulica

Sarothruridae: Flufftails

Sarothruridae (Flufftail) tree

There have been suggestions that this group deserves recognition as a family since atleast Sibley and Ahlquist (1985). Hackett et al. (2008) found thatSarothrura is more closely related to the finfoots than to the rails. Garcia-R. et al. (2014a) found that same is true ofCanirallus, andthat it is more closely related toSarothrura than to the finfoots.Accordingly, they are placed in a separate family.

Canirallus /Mentocrex Split:The genusCanirallus has been split as it was found that part ofthe genus belonged to the Sarothruridae, and part to the Raillidae. SeeBoast et al. (2019), Kirchman et al. (2021), and Oswald et al. (2021).According, one species is retained inCanirallus, the Gray-throatedRail,Canirallus oculeus. The other two move toMentocrex(JL Peters 1932), typekioloides.Canirallus moves back tothe Rallidae, whileMentocrex remains in Sarothruridae.

The Tsingy Wood Rail,Mentocrex beankaensis, has been newlydiscovered within the Madagascan Wood Rail complex. See Goodman et al.(2011).

Livezey (1998) suggested thatRallicula (formerly part ofRallina) may also belong with the flufftails based on aphylogenetic analysis of osteological, myological, and integumentarycharacters.

Finally, recently discovered ancient DNA evidence indicates that boththe extinct cave-rails of the Caribbean (Nesotrochis) and the extinct adzebills of New Zealand (Aptornis) are in the flufftail clade. Given their ancientness, I've ranked themboth as subfamilies in Sarothruridae. See Oswald et al. (2021) and Boast et al.(2019). Although these lineages are ancient, the final extinctions happenedfairly recently.

3 genera, 15 speciesNot HBW Family

• Madagascan Wood Rail,Mentocrex kioloides
• Tsingy Wood Rail,Mentocrex beankaensis
• Chestnut Forest-Rail,Rallicula rubra
• White-striped Forest-Rail,Rallicula leucospila
• Forbes's Forest-Rail,Rallicula forbesi
• Mayr's Forest-Rail,Rallicula mayri
• White-spotted Flufftail,Sarothrura pulchra
• Buff-spotted Flufftail,Sarothrura elegans
• Red-chested Flufftail,Sarothrura rufa
• White-winged Flufftail,Sarothrura ayresi
• Slender-billed Flufftail,Sarothrura watersi
• Streaky-breasted Flufftail,Sarothrura boehmi
• Chestnut-headed Flufftail,Sarothrura lugens
• Striped Flufftail,Sarothrura affinis
• Madagascan Flufftail,Sarothrura insularis

Rallidae: Rails, Gallinules, CootsRafinesque, 1815

56 genera, 152 speciesHBW-3

Click for Rallidae species tree

As I write, the most comprehensive published phylogeny of the rails isGarcia-R. and Matzke (2021). The papers by Kirchman et al. (2021) andBoast et al. (2019) are also quite helpful. Garcia-R. and Matzke (2021)combine molecular data with morphological data from Livezey (1998) toprovide a nearly complete phylogeny of the rails.

However, that is not quite what I'm using for TiF. There are a numberof species where DNA was available, but not used by Garcia-R. and Matzke. In fact, some of that DNA is even from another Garcia-R. paper (Garcia-R.and Trewick, 2015). There are other examples of unused DNA, and I'vemodified the Garcia-R./Matzke phylogeny in an attempt to incorporate thisextra DNA data. This is made considerably easier by the fact that some ofthe extra taxa involve entire clades. In particular, Maley and Brumfield(2013) studied the Clapper/King Rail clade; Garcia-R. and Trewick (2015)examined the swamphens; Groenenberg et al. (2008) focused on the moorhensand Slikas et al. (2002) focused on a clade in the Zapornini (their clade3). In these cases, I just pulled out the version of each clade from theGarcia-R. and Matzke phylogeny and dropped in a replacement.

Other additional DNA involved an isolated species or species pair, andwas a little more difficult to weave in. In some of these cases it wasimpossible to get a consensus. For this, I consulted the papers by Trewick(1997), Kirchman (2012), Boast et al. (2019), Stervander et al. (2019),Chaves et al. (2020), Garcia-R. et al. (2020), and Kirchman et al. (2021).Besides published papers, I also used the 2014 reanalysis of archived databy Raty on BirdForum, for the Speckled Rail,Creciscus notatus.

To help see what DNA was used,, I've added some special markings to thespecies tree. If Garcia-R. and Matzke used DNA,the species name has a trailingblack asterisk. Species where I only hadother DNA available have a trailingred asterisk. Species inorangeindicate extinct species that are not currently on the TiF list. Some ofthem have “ancient” DNA available, in which case they have atrailing asterisk. There are other annotations on the tree, so I'veincluded a key at the end of the tree.

Needless to say, the new phylogeny has involved some big changes. Thebasic structure still involves three subfamilies: Rallinae (which isbasal), and the sister subfamilies Fulicinae (has priority overthe previously used Gallinulinae) and Porphyrioninae. But the compositionof these has changed somewhat.

Rallinae now contains two tribes, Pardirallini (moved from Fulicinae)and Rallini. Fulicinae is also split into two tribes, the new tribeHimantornithini and the old Fulicinini (previously called Gallinulini). The last subfamily, Porphyrioninae now has three tribes: Porphyrionini(basal) followed by Laterallini and Zapornini.

The most noticeable changes involve the genera, which have balloonedfrom 43 to 56! One driving force behind this is that Garcia-R. and Matzke(2021) provide time estimates for every branch in the supplementarymaterial, which is useful when trying to decide whether a separate genus isneeded.

So how old is the rail family? Garcia-R. and Matzke (2021) don'taddress this, but Garcia-R. et al. (2014) did. They put the split betweenthe rails and flufftail/finfoot clade at about 61 mya ±9 millionyears. The high side of that range is probably too high as it is on theother side of the KT boundary. Others have also tried to address thisquestion. Figure 4 in Kirchman et al. (2021) puts it about 40 mya, witha range of roughly 32-53 mya. In Boast et al. (2019, Figure 2) it's at54 mya, with a range of about 41-70 mya. As you can see, these estimatesare not very precise!

The three subfamilies date to 25-30 mya according to Garcia-R. and Matzke,or 10 million older according to Boast et al. (2019), while Kirchman et al. (2021)prefer a date in-between those two. Again, we have no precise number,but the relative scales are still helpful in deciding which clades should be subfamilies.

Canirallus returns:Before considering the subfamilies, we first return one species tothe Rallidae. The genusCanirallus, previously placed inSarothruridae, has been split. Boast et al. (2019), Kirchman et al.(2021), and Oswald et al. (2021) found that part was in Rallidae(Canirallus), and the other part (Mentocrax) inSarothruridae. As a result,Canirallus now consists of a singlespecies, the Gray-throated Rail,Canirallus oculeus.

Rallinae: Long-billed Rails and alliesRafinesque, 1815

There is single deep division in Rallinae, that is about 24.4 millionyears old according to Garcia-R. and Matzke (2021). It divides the Rallinaeinto two tribes: Pardirallini and Rallini. There are also some lesserdivisions within Rallini, with clades, involvingRallus,Crex, andGallirallus. However, those are enough more recentthan the tribal divisions (16.5-18.5 mya) that I do not rank them as tribes(and TiF is allergic to subtribes).

TheCanirallus problem:However, I've glossed over a problem.Canirallus has been reducedto a single species, the Gray-throated Rail,Canirallus oculeus. Weneed to ask, where does it go? There are three relevant papers: Boast etal. (2019), Kirchman et al. (2021), and Oswald et al. (2021). Oswald et al.make clear thatCanirallus belongs in Rallinae, and is a fairlybasal branch, possibly the basal branch. However, they don't include anymembers of Pardirallini, so we can't say much more. Kirchman et al. (2021,Taxon Set C) put it in Rallini. Boast et al. (2019, Fig. 3) put it inPardirallini. In other words, there's a complete lack of consensus.

Can morphology help? Livezey (1998) saysCanirallus is in“a poorly resolved set of rallid genera”, which is not helpful.However, his tree suggests it is more likely in the Rallini. SincePardirallini is confined to the Americas andCanirallus is African,this makes sense (but I could tell a story the other way too). Moreover,the plumage seems more in line with Rallini that Pardirallini. In the end,I'm placing it in Rallini as the basal branch, but my confidence in this atbest medium. With that out of the way, we can now consider the two tribesof Rallinae.

Pardirallini

Compared with the previous treatment in TiF, I've moved the Pardirallinito subfamily Rallinae where they join the Chestnut-headed Crake,Anurolimnas castaneiceps. Otherwise, the composition is unchanged,but has been rearranged a little.

Genus Changes:There are some changes of genus:

• The Colombian Crake,Mustelirallus colombiana and Paint-billed Crake,Mustelirallus erythrops are transferred to genusNeocrex (Sclater and Salvin, 1869), typeerythrops.
• The Zapata Rail,Mustelirallus cerverai returns to themonotypic genusCyanolimnas (Barbour and JL Peters, 1927).

Wood-Rails: Based on Marcondes and Silveira (2015), the Russet-naped Wood-Rail,Aramides albiventris, including subspeciesmexicanus,vanrossemi,pacificus, andplumbeicollis, has been split from Gray-necked Wood-Rail,Aramides cajaneus. Both the AOS NACC and SACC have also changed thename of Gray-necked Wood-Rail to Gray-cowled Wood-Rail.

The phylogeny in Garcia-R. and Matzke (2021) suggests thatmexicana andplumbeicollis are better regarded as subspecies of the Rufous-necked Wood-Rail,Aramides axillaris rather than the Russet-naped Wood-Rail,Aramides albiventris. If this was based on DNA,I would have to take it seriously. However, this is based purelyon morphology and the calls of the birds disagree. The sound likealbiventris. If you have any doubts, consultSACC #797.If you still have doubts, listen to the songs onXeno-canto.org.

Rallini

As mentioned above, Rallini loses two species to Pardirallini. It alsoloses the threeGymnocrex species to Himantornithini, but gainsGray-throated Rail,Canirallus oculeus, from Sarathruridae andRouget's Rail,Rougetius rougetii, previously consideredincertaesedis.

Virginia Rail:There is one more addition to Rallini — a split. Based on Ridgelyand Greenfield (2001) and the fact that the birds in Ecuador soundquite different from those in North America, the Virginia Rail,Ralluslimicola, is split into:

• Virginia Rail,Rallus limicola
• Ecuadorian Rail,Rallus aequatorialis, includingmeyerdeschauenseei

Clapper and King Rails:We also have an older split that's been in TiF for a while.Based on Maley (2012) and Maley and Brumfield (2013),King Rail,Rallus elegans, has been split into:

• King Rail,Rallus elegans (eastern North America, Cuba) and
• Aztec Rail,Rallus tenuirostris (central Mexico).

Further,Clapper Rail,Ralluslongirostris, has been split into:

• Clapper Rail,Rallus crepitans, (Caribbean and eastern North America)
• Ridgway's Rail,Rallus obsoletus, (western North America), and
• Mangrove Rail,Rallus longirostris, (South America).

The table below puts them in their proper order and list the subspecies and describes the ranges.

Mangrove(?) Rails have recently been discovered along the Pacific Coast ofsouthern Central America (esp. the Gulfs of Fonseca and Nicoya). There isalso a report from the Atlantic coast (Panama: Bocos del Toro). I do notknow if any have definitely identified as to subspecies, but Van Dort(2013) suggests the Pacific coast rails are Mangrove Rails. In any event, theycurrently seem to be accepted as Mangrove Rails.

Genus Changes:As of version 3.07 of this page, TiF recognizes a number of new genera in Rallini.

• The African Crake,Crex egregia, is not sister to the Corn Crake, so it moves to the monotypic genusCrecopsis(Sharpe, 1893).
• The Snoring Rail,Lewinia plateni has moved away fromLewinia to become the monotypic genusAramidopsis(Sharpe, 1893).
• The Red Rail,Aphanapteryx bonasia and Rodrigues Rail,Erythromachus leguati will move into theGallirallus group, which is spilt up into various genera, many of them monotypic.
• The Invisible Rail,Gallirallus wallacii is now in the monotypic genusHabroptila (Gray, 1861).
• Hawkins's Rail,Gallirallus hawkinsi has become the monotypic genusDiaphorapteryx (Forbes, 1892).
• The Calayan Rail,Gallirallus calayanensis becomes genusAptenorallus (Kirchman et al., 2021).
• The Chestnut Rail,Gallirallus castaneoventris becomes genusEulabeornis (Gould, 1844).
• The New Caledonian Rail,Gallirallus lafresnayanus becomesTricholimnas (Sharpe, 1893).
• The Chatham Rail,Gallirallus modestus joinsAphanapteryx (Frauenfeld, 1868), typebonasia.
• The Okinawa Rail,Gallirallus okinawae, Barred Rail,Gallirallus torquatus, and Pink-legged Rail,Gallirallusinsignis are all moved toHabropteryx (Stresemann, 1932), typeinsignis.
• Finally, the rest ofGallirallus is transferred to genusHypotaenidia (Reichenbach, 1853), typephilippensis.

Woodford's Rail Complex:While studying Garcia-R and Matzke (2021) and related papers, I noticed asmall problem concerning the Woodford's Rail complex in the Garcia-R andMatzke phylogeny. So let's consider Woodford's Rail,Hypotaenidiawoodfordi. Like IOC, TiF treats as a single species with three subspecies:woodfordi,immaculata andtertia.In contrast, HBW elevates the subspecies to full speciesstatus yielding

• Bougainville Rail,Hypotaenidia tertia
• Santa Isabel Rail,Hypotaenidia immaculata
• Guadalcanal Rail,Hypotaenidia woodfordi

Garcia-R. and Matzke seem to follow this too, but useGallirallusrather thanHypotaenidia. In any event, bothG. woodfordi andG. immaculatus appear on their tree.

So what's the problem? Garcia-R et al. (2014a) had previously sampledDNA from a specimen at the Burke Museum in Seattle, with the resultssubmitted to GenBank in 2013 asNesoclopeus woodfordi. It is also referred to that way in thenex files for Garcia-R. and Matzke(2021). It wasn't noted originally that the DNA sample was actuallyimmaculatus, from Isabel, Solomon Islands. It was just listed aswoodfordi from the Solomons. This is clear in the supplementarydata for the 2014a paper. In fact, the very same specimen together withanother at the Burke was previously sampled by Kirchman (2009), whocorrectly listed them asimmaculatus from Isabel, SolomonIslands.

So what this means is that for Garcia-R and Matzke,G.immaculatus has the means morphological characteristics ofimmaculatus, but no DNA, and theirG. woodfordi is an odd type ofchimera, with the morphological characteristics ofG. woodfordi butDNA ofG. immaculatus. Opps!

Sincewoodfordi seems to be in roughly the right place,genetically speaking, this has an easy fix. I put Woodford's Rail(including the subspecies) whereG. woodfordi was, and supressedG. immaculatus as an error.

There's some question about how old the branches are in the cladecontainingHypotaenidia. Kirchman (2012) argued that many of themare very young, but Garcia-R. and Matzke (2021) gave much older ages thatwould suggest using even more genera than I have used. I have compromisedhere, and consider that the currentHypotaenidia makes a reasonablycoherent group.

Kirchman argued that the genetic distances between all theGallirallus species is fairly small, and indicates a common ancestryas recently as a million or so years ago. This is amazingly recent!

I don't believe this. In particular, I think he has chosen a calibration point unwisely. He based the ages on the fact thatrails cannot have been on Wake Island for more than about 125,000 years.The idea is that Wake Island was completely submerged at that time. Whenhe applied more conventional molecular dating he obtained a much olderdate, about 7.8 million years ago, suggesting that the Wake Island Railrecently arrived from another island. This suggests a very rapid loss ofthe ability to fly. Kirchman (2012) found that the extinct Wake IslandRail's closest relative was the extinct Mangaia Rail,Gallirallusripleyi from the Cook Islands. The Mangiaia Rail is known only fromsubfossil remains that appear to be 700-1000 years old, and its date ofextinction is uncertain. It is not included on the main list, but isincluded in theRallidae tree.

Interestingly, Garcia-R. and Matzke (2021) give a different topology forthis clade and an age of approximately 1.5 million years for split betweenthe Wake Island Rail, Hypotaenidia wakensis, and the Guam Rail,Hypotaenidia owstoni (which is not its closest relative).

The extinct Sharpe's Rail,Gallirallus sharpei, is known fromone specimen from an unknown location. Kirchman (2012) did not obtainDNA from Sharpe's Rail. According to Bird LifeInternational, it is now thought to have been a color morph ofBuff-banded Rail,Gallirallus philippensis. I've not seen anypublished information on this.

Pardirallini: Wood-Rails and alliesLivezey, 1998

1. Chestnut-headed Crake,Anurolimnas castaneiceps
2. Ash-throated Crake,Mustelirallus albicollis
3. Paint-billed Crake,Neocrex erythrops
4. Colombian Crake,Neocrex colombiana
5. Zapata Rail,Cyanolimnas cerverai
6. Plumbeous Rail,Pardirallus sanguinolentus
7. Spotted Rail,Pardirallus maculatus
8. Blackish Rail,Pardirallus nigricans
9. Uniform Crake,Amaurolimnas concolor
10. Little Wood-Rail,Aramides mangle
11. Rufous-necked Wood-Rail,Aramides axillaris
12. Russet-naped Wood-Rail,Aramides albiventris
13. Giant Wood-Rail,Aramides ypecaha
14. Gray-cowled Wood-Rail,Aramides cajaneus
15. Brown Wood-Rail,Aramides wolfi
16. Red-winged Wood-Rail,Aramides calopterus
17. Slaty-breasted Wood-Rail,Aramides saracura

Rallini: Long-billed Rails and alliesRafinesque, 1815

1. Gray-throated Rail,Canirallus oculeus
2. Plain-flanked Rail,Rallus wetmorei
3. Ridgway's Rail,Rallus obsoletus
4. Aztec Rail,Rallus tenuirostris
5. Mangrove Rail,Rallus longirostris
6. King Rail,Rallus elegans
7. Clapper Rail,Rallus crepitans
8. Water Rail,Rallus aquaticus
9. Brown-cheeked Rail,Rallus indicus
10. African Rail,Rallus caerulescens
11. Virginia Rail,Rallus limicola
12. Ecuadorian Rail,Rallus aequatorialis
13. Bogota Rail,Rallus semiplumbeus
14. Austral Rail,Rallus antarcticus
15. African Crake,Crecopsis egregia
16. Rouget's Rail,Rougetius rougetii
17. Snoring Rail,Aramidopsis plateni
18. White-throated Rail,Dryolimnas cuvieri
19. †Reunion Rail,Dryolimnas augusti
20. Madagascan Rail,Biensis madagascariensis
21. Corn Crake,Crex crex
22. Slaty-breasted Rail,Lewinia striata
23. Brown-banded Rail,Lewinia mirifica
24. Lewin's Rail,Lewinia pectoralis
25. Auckland Rail,Lewinia muelleri
26. Invisible Rail,Habroptila wallacii
27. †Rodrigues Rail,Erythromachus leguati
28. †Hawkins's Rail,Diaphorapteryx hawkinsi
29. Calayan Rail,Aptenorallus calayanensis
30. Chestnut Rail,Eulabeornis castaneoventris
31. Weka,Gallirallus australis
32. New Caledonian Rail,Tricholimnas lafresnayanus
33. †Red Rail,Aphanapteryx bonasia
34. †Chatham Rail,Aphanapteryx modesta
35. Pink-legged Rail,Habropteryx insignis
36. Okinawa Rail,Habropteryx okinawae
37. Barred Rail,Habropteryx torquatus
38. Lord Howe Woodhen,Hypotaenidia sylvestris
39. Woodford's Rail,Hypotaenidia woodfordi
40. †Bar-winged Rail,Hypotaenidia poeciloptera
41. †Dieffenbach's Rail,Hypotaenidia dieffenbachii
42. Buff-banded Rail,Hypotaenidia philippensis
43. Roviana Rail,Hypotaenidia rovianae
44. Guam Rail,Hypotaenidia owstoni
45. †Tahiti Rail,Hypotaenidia pacificus
46. †Wake Island Rail,Hypotaenidia wakensis

FulicinaeNitzsch 1829

The next group is the Fulicinae. Obviously, it contains the coots(Fulica). It's split into two tribes: Himantornithiniand Fulicini. Although the coots and moorhens are here, neither theswamphens nor the American Purple Gallinule are part of theFulicinae.

Himantornithini

Himantornithini contains two genera:Himantornis andGymnocrex. Kirchman et al. (2021) is the only paper to include DNAfrom both genera. They found them to be sister genera. Garcia-R. andMatzke found the same thing, basing their conclusion on morphology ratherthan DNA. Unlike Garcia-R. and Matzke (2021), both Garcia-R. et al. (2020)and Kirchman et al. foundHimantornis sister to Fulicini, and thatis how I will treat them. This group is a distant sister of Fulicini, soI've placed them in separate tribes. Since the topology is different fromthat in Garcia-R. and Matzke, we can't get a read on the distance. However, Kirchman et al. found them intermediate in age between thePorphyrionini branch and the Laterallini/Zapornini split. Translated tothe Garcia-R. and Matzke paper, that would be an age of 26 to 28.5 millionyears.

So why do Garcia-R. and Matzke (2021) put Himantornithini sister to therest of the Rallidae? I don't know, perhaps the morphological data theyuse, combined with limited DNA causes it. Certainly, their positionning ofHimantornithini is much the same as Liverzey (1998), based purely onmorphological data. We know that weird things can happen when usingmorphological data. Garcia-R. and Matzke unintentionally demonstrate this with theNesotrochis cave-rails. They place the cave-rails next totheAramides wood-rails. In fact, there is genetic data for thecave-rails (Oswald et al., 2021). They found the cave-rails belong to the Flufftail family, Sarothruridae.

Fulicini

Fulicini mainly contains the moorhens/gallinules and coots, togetherwith the similar woodhens and native-hens. Fulicini also contains theremaining portion ofPorzana, after many species have beentransferred toZapornia and elsewhere. I follow Christidis and Boles(2008) and separatePareudiastes (Hartlaub and Finsch, 1871), typepacificus, andTribonyx (DuBus, 1840), typemortierii,fromGallinula.

Spot-flanked Gallinule:The Spot-flanked Gallinule, formerly inGallinula, is now in themonotypic genusPorphyriops (Pucheran 1845).Raty's reanalysis found that the Spot-flanked Gallinule,Porphyriopsmelanops, belongs nearPorzana. This makes sense as theSpot-flanked Gallinule looks much like a Sora,Porzana carolina.Garcia-R. et al. (2014), using more data, placed it as shown on the diagramand estimated that the common ancestor ofPorzana andPorphyriops lived perhaps 20 million years ago. I believe thisnumber to be somewhat exaggerated, perhaps by 40%. Even so, they wouldstill be separated by 12 million years, more than enough to put them indifferent genera. The placement sister toPorzana is supported byBoast et al. (2019), Garcia-R. et al. (2020), and Kirchman et al. (2021),using partially different gene sets, although Garcia-R. and Matzke (2021)put it in a different position.

Lesser Moorhen:The Lesser Moorhen is nowParagallinula angulata instead ofGallinula (Sangster, Garcia-R., and Trewick, 2015), typeangulata. This change was needed becauseParagallinula isbasal to bothGallinula (true gallinules and moorhens) andFulica (coots).

Common Gallinule/Moorhen:This list treats the Common Gallinule,Gallinula galeata and CommonMoorhen,Gallinula chloropus, as separate species, as does the AOU(both NACC and SACC). Groenenberg et al. (2008) found that the CommonMoorhen is more closely related to the Gough Moorhen,Gallinulacomeri, and the extinct Tristan Moorhen,Gallinula nesiotis,than to the Common Gallinule. The relationships of the rest of the formerGallinula have not been subject to genetic testing.

Caribbean Coot:Following AOU supplement 57 (2016), the Caribbean Coot,Fulicacaribaea, is now treated as a color morph of American Coot,Fulicaamericana. There was never strong evidence these were separatespecies.

Finally, if you look at the tree, you may notice Hodgen's Waterhen,Pyramida hodgenorum. It's an extinct species from New Zealand thatis not on the TiF list. The monotypic genusPyramida (Oliver, 1955)takes its name from the Pyramid Valley where many subfossil remains ofHodgen's Waterhen were found. The most recent date to the 1700's. It appearsto have once been widespread in New Zealand.

Himantornithini: Bush-hens and WaterhensGR Gray 1871

1. Nkulengu Rail,Himantornis haematopus
2. Blue-faced Rail,Gymnocrex rosenbergii
3. Talaud Rail,Gymnocrex talaudensis
4. Bare-eyed Rail,Gymnocrex plumbeiventris

Fulicini: Coots, True Gallinules and MoorhensNitzsch 1829

1. Spot-flanked Gallinule,Porphyriops melanops
2. Sora,Porzana carolina
3. Spotted Crake,Porzana porzana
4. Australian Crake,Porzana fluminea
5. Makira Woodhen,Pareudiastes silvestris
6. †Samoan Woodhen / Samoan Moorhen,Pareudiastes pacificus
7. Black-tailed Native-hen,Tribonyx ventralis
8. Tasmanian Native-hen,Tribonyx mortierii
9. Lesser Moorhen,Paragallinula angulata
10. Dusky Moorhen,Gallinula tenebrosa
11. Common Gallinule,Gallinula galeata
12. †Tristan Moorhen,Gallinula nesiotis
13. Gough Moorhen,Gallinula comeri
14. Common Moorhen,Gallinula chloropus
15. Red-fronted Coot,Fulica rufifrons
16. †Mascarene Coot,Fulica newtonii
17. Giant Coot,Fulica gigantea
18. Horned Coot,Fulica cornuta
19. Red-gartered Coot,Fulica armillata
20. Eurasian Coot,Fulica atra
21. Red-knobbed Coot,Fulica cristata
22. Hawaiian Coot,Fulica alai
23. American Coot,Fulica americana
24. Slate-colored Coot / Andean Coot,Fulica ardesiaca
25. White-winged Coot,Fulica leucoptera

PorphyrioninaeReichenbach, 1849

Although the swamphens are placed next to the gallinules and coots, thegenetic distance between them is quite large. Garci-R. and Matzke (2021)estimate that their most recent common ancestor lived about 28 millionyears ago. This should not now be surprising. Earlier genetic evidencehad shown they were quite separated. Ozaki et al. (2010) placed theswamphens sister to Laterallini and Trewick (1997) grouped them withZapornini. The morphological evidence has also cast doubt on the positionof the swamphens, with Livezey (1998) putting them in a relatively basalposition.

The swamphens are in the final subfamily, Porphyrioninae. In the currentedition of TiF, it contains three tribes: Porphyrionini, Zapornini, andLaterallini. The swamphen tribe Porphyrionini is basal group in subfamily Porphyrioninae.

Porphyrionini

The first tribe of Porphyrioninae, Porphyrionini, is mainly composed ofpurple gallinules (Porphyrula) and swamphens (Porphyrio). Morphological data suggests thatAphanocrex belongs in this tribetoo. Indeed, Livezey's (1998) analysis suggested that the extinct St.Helena Rail,Aphanocrex podarces, might even be embedded inPorphyrula, while Garcia-R. and Matzke's (2021) combinedmolecular/morphological analysis put them as the basal bird inPorphyrionini.

American purple gallinules:Olson (1973) had recommended that the American purple gallinule genus,Porphyrula, be merged intoPorphyrio. Trewick (1997) made agenetic case for this merger, which was adopted by AOU in 2002. However,Garcia-R and Matzke (2021) found that common ancestor of the two generalived over 8 mya. As a result, the purple gallinules

• Allen's Gallinule,Porphyrio alleni
• Purple Gallinule,Porphyrio martinica
• Azure Gallinule,Porphyrio flavirostris

have been restored toPorphyrula (Blyth, 1852), typealleni.

Purple Swamphens:Based on the genetic analysis of Garcia-R. and Trewick (2015) and theearlier papers by Sangster (1998) and Sangster et al. (1999), the PurpleSwamphen,Porphyrio porphyrio has been split into 5 species:

• Western Swamphen,Porphyrio porphyrio,
• Black-backed Swamphen,Porphyrio indicus, includingviridis,
• Australasian Swamphen,Porphyrio melanotus, includingmelanopteruspalliatus,bellus,samoensis,vitiensis, andpelewensis,
• Gray-headed Swamphen,Porphyrio poliocephalus, includingcaspius andseistanicus,
• Philippine Swamphen,Porphyrio pulverulentus.

Laterallini

The next tribe is Laterallini, the tribe of New World crakes. It'scomposition remains pretty much the same as before, except for transferringthe St. Helena Crake,Zapornia astrictocarpus (Zapornini) to genusLaterallus (GR Gray, 1855), typemelanophaius.

Rufirallus:I have resurrectedRufirallus (Bonaparte 1854, typeviridis)for the Russet-crowned Crake,Rufirallus viridis. Based solely onmorphological data, Garcia-R. and Matzke (2021) group it with theRusty-flanked Crake,Laterallus levraudi and Ruddy Crake, asRufirallus ruber. However, genetic data from Kirchman et al. (2021)indicates that the Ruddy Crake belongs elsewhere. Genetic data is almostinvariably superior. Using a sparser data set, Kirchman et al. found itcloser toalbigularis than tomelanophaius, and I haverepositioned it accordingly on the Garcia-R. and Matzke tree. I wasn'tsure whether to bringlevraudi with it. Without a compelling reasonto move it, I left it sister toviridis.

Whether the above is correct or not, the Russet-crowned Crake, typespecies ofRufirallus, is closely related to the Ocellated Crake,Micropygia schomburgkii.Micropygia andRufirallusform the basal claed in Laterallini.

More genus changes:Here are the changes of genus for the rest of Laterallini:

• The Ascension Crake,Mundia elpenor joins the black rail genusCreciscus (Cabanis, 1857), typejamaicensis.
• The White-throated Crake,Limnocrex albigularis, is transferredtoLaterallus (GR Gray, 1855), typemelanophaius.
• The Gray-breasted Crake,Laterallus exilis is sister to the Yellow-breasted Crake,Hapalocrex flaviventer. However, theyare distant relations, so a new genus is needed. I don't know ofany withexilis as type, so a temporary name must be used.Gray-breasted Crake becomes"Hapalocrex" exilis.

The next branch contains Swinhoe's and Yellow Rails (Coturnicops).

After that, the picture becomes much murkier. Stervander et al. (2019),Garcia-R et al. (2020), Garcia-R. and Matzke (2021), and Kirchman et al.(2021) have all weighed in on this, and their results can't really bereconciled. Because Garcia-R. and Matzke use the most data, I havefollowed a lightly modified form of their tree for the rest ofLaterallini.

In Garcia-R and Matzke (2021), the next branch consists of theYellow-breasted Crake,Hapalocrex flaviventer and the Gray-breastedCrake. These seem rather distinct, and are separated by about 8 millionyears (Gracia-R. and Matzke), so I've put them in separate genera.Unusually for the rails, there's no genus name available, so I'm referringto the Gray-breasted Crake as"Hapalocrex" exilis. Note thatStervander et al. (2019) have bothexilis andflaviventer inLaterallus. Further, both Garcia-R and Matzke (with morphology) andGarcia-R. et al. (2020, genes only), have the Rufous-sided Crake,Laterallus melanophaius and Red-and-white Crake,Laterallusleucopyrrhus as sister species. Kirchman et al. (2021) have a verydifferent take on this. The other papers don't consider both of them.

The rest of Laterallini are either inLaterallus orCreciscus. This includes the Black-banded Crake,Laterallusfasciatus, which has been moved fromRufirallus (orPorzana orAnurolimnas, depending on the taxonomy used). I suspect there will be further changes in the clade as the other taxa aresampled, and we may need to recut the genera too.Laterallus inGarcia-R. and Matzke is estimated to have had a common ancestor about 5 mya,so I treat it as a single genus. However, movingruber there mayhave changed matters, but that will have to wait for a more comprehensive studyof the clade.

Raty's reanalysis grouped together a few species that look like Black Rail(jamaicensis). Most importantly, it included the Speckled Rail,Creciscus notatus, which doesn't seem to be in any publishedmolecular phylogeny. The Galapagos Rail belongs here too (Chaves et al.,2020). The old nameCreciscus (Cabanis 1857, typejamaicensis) applies to the group, which also contains the AscensionCrake,Creciscus elpenor. On morphological grounds, theInaccessible Island Rail,Creciscus rogersi, and the St. HelenaCrake,Creciscus astrictocarpus also belong to this group.

Zapornini

The last tribe to consider is Zapornini, which combines the old Zaporniniwith almost all of the old Himantornithini,Himantornis itself excepted.Zapornini contains a number of Old World crakes, some of them once place inPorzana.

There are a number of genus changes that affect Zapornini.

• The Isabelline Bush-hen,Amaurornis isabellina, seems a bit distantfrom the otherAmaurornis and becomes the monotypic genusOenolimnas (Sharpe, 1853).
• The Brown Crake,Zapornia akool, is transferred toLimnocorax (W Peters, 1854), typeflavirostra.
• Finally, a group of formerZapornia (and before thatPorzana) have been separated as the genusPennula (Dole, 1878), typesandwichensis. They are:
  • Sakalava Rail,Zapornia olivieri
  • Black-tailed Crake,Zapornia bicolor
  • Hawaiian Rail,Zapornia sandwichensis
  • Red-eyed Crake,Zapornia atra
  • Spotless Crake,Zapornia tabuensis
  • Kosrae Crake,Zapornia monasa
  • Tahiti Crake,Zapornia nigra

There are two parts to Zapornini. The first is the remains of the old Himantornithini. This clade constains the Watercock, various Bush-hens,and the White-breasted Waterhen,Amaurornis phoenicurus.

The other part starts with what remains ofRallina. The phylogenyfor the rest is based on Slikas et al. (2002), except for the Maui Crakes,which are placed based on morphology. Amazingly, part 3 of Figure 2 formsa nice block that can just be dropped into the tree, and that is what I did.

Many of the other species in this clade were previously classified ingenusPorzana, but were then moved toZapornia (Leach 1816,typeGallinula minuta Montagu 1813 =parva). Given thegenetic distances involved, I've separated the Ruddy-breasted Crake andBand-bellied Crake in genusLimnobaenus (Sundevall 1873, typefuscus); and also the Black Crake asLimnocorax flavirostra(W. Peters, 1854) along with the Brown Crake, nowLimnocorax akool.The remaining species then fall into two groups. Again, there issubstantial distance between them, and they are the remainingZapornia and the birds just transferred toPennula.

Porphyrionini: Purple Gallinules and SwamphensReichenbach, 1849

1. †St. Helena Rail,Aphanocrex podarces
2. Allen's Gallinule,Porphyrula alleni
3. Purple Gallinule,Porphyrula martinica
4. Azure Gallinule,Porphyrula flavirostris
5. Western Swamphen,Porphyrio porphyrio
6. Black-backed Swamphen,Porphyrio indicus
7. African Swamphen,Porphyrio madagascariensis
8. South Island Takahe,Porphyrio hochstetteri
9. †North Island Takahe,Porphyrio mantelli
10. Australasian Swamphen,Porphyrio melanotus
11. Gray-headed Swamphen,Porphyrio poliocephalus
12. Philippine Swamphen,Porphyrio pulverulentus
13. †White Swamphen / Lord Howe Swamphen,Porphyrio albus

Laterallini: New World CrakesKirchner et al., 2021

1. Ocellated Crake,Micropygia schomburgkii
2. Russet-crowned Crake,Rufirallus viridis
3. Rusty-flanked Crake,Rufirallus levraudi
4. Swinhoe's Rail,Coturnicops exquisitus
5. Yellow Rail,Coturnicops noveboracensis
6. Yellow-breasted Crake,Hapalocrex flaviventer
7. Gray-breasted Crake,"Hapalocrex" exilis
8. Ruddy Crake,Laterallus ruber
9. White-throated Crake,Laterallus albigularis
10. Rufous-faced Crake,Laterallus xenopterus
11. Black-banded Crake,Laterallus fasciatus
12. Rufous-sided Crake,Laterallus melanophaius
13. Red-and-white Crake,Laterallus leucopyrrhus
14. Speckled Rail,Creciscus notatus
15. Black Rail,Creciscus jamaicensis
16. Galapagos Rail / Galapagos Crake,Creciscus spilonota
17. Dot-winged Crake,Creciscus spiloptera
18. †St. Helena Crake,Creciscus astrictocarpus
19. †Ascension Crake,Creciscus elpenor
20. Inaccessible Island Rail,Creciscus rogersi

Zapornini: Old World CrakesReichenbach, 1849

1. White-browed Crake,Poliolimnas cinereus
2. New Guinea Flightless Rail,Megacrex inepta
3. Striped Crake,Aenigmatolimnas marginalis
4. Watercock,Gallicrex cinerea
5. Isabelline Bush-hen,Oenolimnas isabellinus
6. White-breasted Waterhen,Amaurornis phoenicurus
7. Plain Bush-hen,Amaurornis olivacea
8. Talaud Bush-hen,Amaurornis magnirostris
9. Pale-vented Bush-hen,Amaurornis moluccana
10. Slaty-legged Crake,Rallina eurizonoides
11. Red-necked Crake,Rallina tricolor
12. Andaman Crake,Rallina canningi
13. Red-legged Crake,Rallina fasciata
14. Ruddy-breasted Crake,Limnobaenus fuscus
15. Band-bellied Crake,Limnobaenus paykullii
16. Black Crake,Limnocorax flavirostra
17. Brown Crake,Limnocorax akool
18. Little Crake,Zapornia parva
19. Baillon's Crake,Zapornia pusilla
20. †Laysan Rail,Zapornia palmeri
21. Black-tailed Crake,Pennula bicolor
22. Sakalava Rail,Pennula olivieri
23. †Hawaiian Rail,Pennula sandwichensis
24. Spotless Crake,Pennula tabuensis
25. †Kosrae Crake,Pennula monasa
26. Red-eyed Crake / Henderson Crake,Pennula atra
27. †Tahiti Crake,Pennula nigra

Previous PageNext Page