-
Welcome to BirdForum, the internet's largest birding community with thousands of members from all over the world. The forums are dedicated to wild birds, birding, binoculars and equipment and all that goes with it.
Please register for an account to take part in the discussions in the forum, post your pictures in the gallery and more.
Gruiformes and Charadriiformes (1 Viewer)
- Thread starter Peter Kovalik
- Start date
Snapdragyn
Well-known member
Like the split of Asian Dowitcher from the other 2 dowitchers being older than some passerine families?
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
Or almost 30 mya for the first lineage of Burhinidae, it's huge !!
andrew147
Well-known member
Mostly fits well with other divergence time estimates for Charadriiformes. Thick-knees, plovers, sandpipers, coursers and auks always show up as comprising lots of very old lineages.
I agree that some of the details are odd. Oystercatchers are convincingly shown elsewhere to constitute an Old World and a New World clade, positions of Lymnocryptes, Eupoda asiatica, Vanellus duvaucelii, Glareola ocularis and Stercorarius longicaudus are unexpected, and monophyletic noddies as sibling to terns was definitely a surprise.
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
It could well be that their position in the tree would be due to errors from what I understand
DavidCerny
New member
Hi everyone,
I’m one of the authors of the preprint under discussion. Thank you all for your interest in our work; it’s very exciting to see it provoking thoughtful discussions before it has even been published. At the same time, let me take this opportunity to remind everyone that for the time being, our study is really just that: a preprint, not a published paper. It hasn’t passed peer review yet. That’s a good thing – it means that any potential errors can still be fixed!
I thought I could perhaps clear up potential confusion and respond to some of the comments here:
We didn’t tackle this question because it can’t really be tackled. Unlike clades, families are not objectively real biological entities whose existence can be inferred from data. The number of families is not decided by analysis but by convention. I’m well aware that there have been efforts to make ranking schemes objective by basing them on divergence time cut-offs, but at the end of the day, the choice of such a cut-off is still either arbitrary or circular (“let’s choose a cut-off that will allow us to keep calling the current families families”). Personally, I find arguing about what should or should not be a family tedious and pointless: again, there is no way to really “win” that argument, because it’s ultimately decided by convention, and you’re not learning anything about your group in the process. The time would be better spent on using your phylogeny to investigate questions of actual biological interest: things like trait evolution, diversification rates, historical biogeography, and their potential connections. Most biologists today feel the same way.
That’s a very interesting point; thank you for raising it. You’re right that the position of Lymnocryptes within Limosinae is driven by RAG1; none of the other four loci we have for the taxon (12S, Cox1, CytB, ND2) backs it up. That explains why the result doesn’t occur in the ASTRAL analysis. As for the alternatives:
Any evidence that KY419888 comes from the Kentish rather than the long-billed plover? As I’m sure you know (the GenBank annotation you link to says as much), the mitogenome was published in a paper specially written for that purpose (Lee et al. 2017; doi:10.1080/23802359.2017.1292473), and the publisher’s website doesn’t link to any later corrections. It would be a pretty astonishing failure if a paper solely dedicated to sequencing the mitogenome of the long-billed plover actually sequenced a different species by accident.
The fact that a short Cox1 fragment favors a different position for the taxon than the mitogenome as a whole doesn’t mean much. That happens all the time; stochastic error is pervasive in short-sequence phylogenies.
That’s because there are currently no data available for it, as we explicitly say in section 4.1.
Agreed; this was one of the toughest nuts to crack during the taxonomic reconciliation process. As it happens, we used no sequences for Esacus/Burhinus magnirostris at all. It’s one of the 31 species whose position in the tree is based solely on the morphological dataset assembled by Strauch (1978) and corrected by Chu (1995; see the preprint for the full citations). Our Burhinus magnirostris is Strauch’s Esacus magnirostris; Strauch’s Burhinus magnirostris is our Burhinus grallarius. I actually remember the exact moment I became really paranoid about whether we got this right; I checked our matrix cell by cell against Strauch’s Table 1, and to my great relief found out that everything was as it was supposed to be.
Incidentally, if you’re skeptical about the paraphyly of Burhinus with respect to Esacus, this pretty much explains where that finding comes from. The phylogenetic signal in the 69 morphological characters is very weak, and so their ability to discriminate between competing hypotheses about taxon placement is very low: in technical phylogenetic terms, we would speak of a flat likelihood surface. The 69 characters contribute so little to the overall tree likelihood that if you have nothing else from a taxon, its placement will be close to random. Not completely random – there is some signal for burhinid interrelationships in the Strauch matrix – but close.
Maybe it is. Bayesian node-dating is hard. Calibration choice exerts huge influence on the results; in our tree, a lot probably hinges on whether a single fossil specimen from the late Eocene is a pan-alcid or not. Calibration priors interact both with one another and with the branching-process prior. The choice of the relaxed clock model – in more understandable terms, the assumptions you make about how the rate of DNA evolution itself evolves along the tree – matters a lot. The DNA sequence data can rule out certain combinations of branch rates and branching times, but for any given rate, you can always find a branching time that will make it work, and vice versa.
In the preprint, we go over these things in some detail, and we note that studies employing greater amounts of sequence data will likely be able to do better. For now, though, I have no reason to suspect that our node ages are systematically overestimated. Is it weird, as another user mentioned, that the divergences among the dowitchers – members of the same genus – are older than some passerine families? Well, not really, because taxonomic ranks like “genus” and “family” have no biological meaning. Traditionally, rank assignment has been based on phenotypic distinctiveness and the number of included species, both of which do correlate with the time elapsed since a group’s origin, but only imperfectly. Unequal rates of diversification and phenotypic evolution can cause that correlation to break down in all sorts of ways.
Interesting; do you happen to have a citation for that?
Please note that nowhere in the preprint do we proclaim “this is The One True Phylogeny; bow down before it” – there is a reason for that! We used multiple analytical methods and even multiple datasets (molecules only vs. molecules + morphology); all of these yielded estimates that differed from one another to various degrees. Even within any given tree, not all nodes are equally trustworthy; some are robust while others are barely any better than the next best alternative. I think that in the preprint, we were rather upfront about all this: we dedicated a lot of space to pointing out poorly supported relationships, discussed conflicts between different trees and their possible causes, and highlighted areas in need of improvement. So I find it surprising that the discussion here tends to ignore all of that uncertainty. Are those taxa’s positions unexpected? Well, which ones? We found multiple!
I’m one of the authors of the preprint under discussion. Thank you all for your interest in our work; it’s very exciting to see it provoking thoughtful discussions before it has even been published. At the same time, let me take this opportunity to remind everyone that for the time being, our study is really just that: a preprint, not a published paper. It hasn’t passed peer review yet. That’s a good thing – it means that any potential errors can still be fixed!
I thought I could perhaps clear up potential confusion and respond to some of the comments here:
We didn’t tackle this question because it can’t really be tackled. Unlike clades, families are not objectively real biological entities whose existence can be inferred from data. The number of families is not decided by analysis but by convention. I’m well aware that there have been efforts to make ranking schemes objective by basing them on divergence time cut-offs, but at the end of the day, the choice of such a cut-off is still either arbitrary or circular (“let’s choose a cut-off that will allow us to keep calling the current families families”). Personally, I find arguing about what should or should not be a family tedious and pointless: again, there is no way to really “win” that argument, because it’s ultimately decided by convention, and you’re not learning anything about your group in the process. The time would be better spent on using your phylogeny to investigate questions of actual biological interest: things like trait evolution, diversification rates, historical biogeography, and their potential connections. Most biologists today feel the same way.
That’s a very interesting point; thank you for raising it. You’re right that the position of Lymnocryptes within Limosinae is driven by RAG1; none of the other four loci we have for the taxon (12S, Cox1, CytB, ND2) backs it up. That explains why the result doesn’t occur in the ASTRAL analysis. As for the alternatives:
- Cox1 has Lymnocryptes nested within a paraphyletic Limnodromus;
- ND2 shows Limnodromus to be paraphyletic with respect to both Lymnocryptes and the rest of Scolopacinae;
- CytB has Lymnocryptes inside a clade that includes all the usual representatives of Scolopacinae and Tringinae but doesn’t find those two subfamilies to be reciprocally monophyletic;
- With 12S, the sandpipers as a whole are a massively polyphyletic mess in which nothing makes sense and none of the five established subfamilies come out monophyletic.
Any evidence that KY419888 comes from the Kentish rather than the long-billed plover? As I’m sure you know (the GenBank annotation you link to says as much), the mitogenome was published in a paper specially written for that purpose (Lee et al. 2017; doi:10.1080/23802359.2017.1292473), and the publisher’s website doesn’t link to any later corrections. It would be a pretty astonishing failure if a paper solely dedicated to sequencing the mitogenome of the long-billed plover actually sequenced a different species by accident.
The fact that a short Cox1 fragment favors a different position for the taxon than the mitogenome as a whole doesn’t mean much. That happens all the time; stochastic error is pervasive in short-sequence phylogenies.
That’s because there are currently no data available for it, as we explicitly say in section 4.1.
Agreed; this was one of the toughest nuts to crack during the taxonomic reconciliation process. As it happens, we used no sequences for Esacus/Burhinus magnirostris at all. It’s one of the 31 species whose position in the tree is based solely on the morphological dataset assembled by Strauch (1978) and corrected by Chu (1995; see the preprint for the full citations). Our Burhinus magnirostris is Strauch’s Esacus magnirostris; Strauch’s Burhinus magnirostris is our Burhinus grallarius. I actually remember the exact moment I became really paranoid about whether we got this right; I checked our matrix cell by cell against Strauch’s Table 1, and to my great relief found out that everything was as it was supposed to be.
Incidentally, if you’re skeptical about the paraphyly of Burhinus with respect to Esacus, this pretty much explains where that finding comes from. The phylogenetic signal in the 69 morphological characters is very weak, and so their ability to discriminate between competing hypotheses about taxon placement is very low: in technical phylogenetic terms, we would speak of a flat likelihood surface. The 69 characters contribute so little to the overall tree likelihood that if you have nothing else from a taxon, its placement will be close to random. Not completely random – there is some signal for burhinid interrelationships in the Strauch matrix – but close.
Maybe it is. Bayesian node-dating is hard. Calibration choice exerts huge influence on the results; in our tree, a lot probably hinges on whether a single fossil specimen from the late Eocene is a pan-alcid or not. Calibration priors interact both with one another and with the branching-process prior. The choice of the relaxed clock model – in more understandable terms, the assumptions you make about how the rate of DNA evolution itself evolves along the tree – matters a lot. The DNA sequence data can rule out certain combinations of branch rates and branching times, but for any given rate, you can always find a branching time that will make it work, and vice versa.
In the preprint, we go over these things in some detail, and we note that studies employing greater amounts of sequence data will likely be able to do better. For now, though, I have no reason to suspect that our node ages are systematically overestimated. Is it weird, as another user mentioned, that the divergences among the dowitchers – members of the same genus – are older than some passerine families? Well, not really, because taxonomic ranks like “genus” and “family” have no biological meaning. Traditionally, rank assignment has been based on phenotypic distinctiveness and the number of included species, both of which do correlate with the time elapsed since a group’s origin, but only imperfectly. Unequal rates of diversification and phenotypic evolution can cause that correlation to break down in all sorts of ways.
Interesting; do you happen to have a citation for that?
Please note that nowhere in the preprint do we proclaim “this is The One True Phylogeny; bow down before it” – there is a reason for that! We used multiple analytical methods and even multiple datasets (molecules only vs. molecules + morphology); all of these yielded estimates that differed from one another to various degrees. Even within any given tree, not all nodes are equally trustworthy; some are robust while others are barely any better than the next best alternative. I think that in the preprint, we were rather upfront about all this: we dedicated a lot of space to pointing out poorly supported relationships, discussed conflicts between different trees and their possible causes, and highlighted areas in need of improvement. So I find it surprising that the discussion here tends to ignore all of that uncertainty. Are those taxa’s positions unexpected? Well, which ones? We found multiple!
Last edited:
andrew147
Well-known member
Thank you for detailed reply to people's comments. I certainly wasn't holding you accountable for a "One True phylogeny" - just thought there were a few interesting talking points.
My oystercatcher comment based on:
Senfeld et al. (2019) Taxonomic status of the extinct Canary Islands Oystercatcher Haematopus meadewaldoi; Ibis 162 (3): 1068-1074
+ presented but as yet unpublished paper discussed here: (cladogram below)
+ unpublished analysis of various genes with reference to osculans (file below)
Attachments
Mysticete
Well-known member
Oh I am familiar with the knowledge that taxonomic ranks (even species) are in the big picture arbitrary and there are more interesting things to do with phylogenies (I am a assistant professor in a biology department by the way, with a background in phylogenetics, I just don't work on birds).
I just brought that up because their there have been a fair number of papers dealing with songbirds where folks have been interested in trying to make the different ranks "equivalent", and SACC has clearly made some decisions on ranking based on relative age. So clearly there are people who find these exercises to be useful, even if its not something you are interested in.
I just brought that up because their there have been a fair number of papers dealing with songbirds where folks have been interested in trying to make the different ranks "equivalent", and SACC has clearly made some decisions on ranking based on relative age. So clearly there are people who find these exercises to be useful, even if its not something you are interested in.
Acanthis
Well-known member
Or it's true affinities may lie elsewhere Tom
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
Unless surprised, e.g. potential paraphyly of Ortyxelos within Turnix
Acanthis
Well-known member
Hi David.
I think many or most of us here on this forum are not biologists or even particularly interested in the methods used to investigate bird phylogeny. I suspect many just cut straight to the trees to find out if there are any surprises, unexpected placements, or species not studied before.
The joy is in the reveal!
So i just wish say thanks to yourself and your team for all the hard work and for allowing access to the preprint.
And on a personal note thanks for revealing more of the relationships within the Charadrii, a group I'm kind of fond of 🙂
Acanthis
Well-known member
Or perhaps basal to the coursers?
thyoloalethe
Well-known member
I've also thought it seemed rather courser-like!
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
Not stupid
Acanthis
Well-known member
Only time will tell
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
It doesn't look like a turnix indeed
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
If new samples are available (for Ortyxelos, could you easily get some from voucher?), It would be interesting to study these families or genera in separate works.
Mysticete
Well-known member
Taxonomy in flux has a substantial update focused on Charadriiformes:
Magellanic Plover: The Magellanic Plover family, Pluvianellidae, has been demoted to a subfamily (Pluvianellinae) of the sheathbill family Chionidae.
[Chionidae, Charadriiformes, 3.06]
Thick-knees: Based on Paton et al. (2003) and Černý and Natale, (2021), the genus Burhinus is divided into three genera:
[Burhinidae, Charadriiformes, 3.06]
Stilts and Avocets: The Banded Stilt, Cladorhynchus leucocephalus, is basal in family Recurvirostridae, which has been rearranged accordingly.
[Recurvirostridae, Charadriiformes, 3.06]
Far Eastern Oystercatcher, Haematopus osculans: Based on Senfeld et al. (2020b), the Far Eastern Oystercatcher, Haematopus osculans, has been split from the Eurasian Oystercatcher, Haematopus ostralegus.
[Haematopodidae, Charadriiformes, 3.06]
Oystercatchers: Based on Senfeld et al. (2020b), the American Oystercatchers have been separated in genus Prohaematopus (Matthews 1913), type ater.
[Haematopodidae, Charadriiformes, 3.06]
Charadriidae Genus Changes:
[Charadriidae, Charadriiformes, 3.06]
Buttonquails: The buttonquail tree was constructed by combining the species groups described by Debus (1996) with the 7-species phylogeny of Černý and Natale (2021). I was surprised there were no conflicts.
[Turnicidae, Charadriiformes, 3.06]
Coursers and Pratincoles:
[Glareolidae, Charadriiformes, 3.06]
Jaegers and Skuas: The Stercorariidae have been rearranged slightly to conform with Černý and Natale (2021).
[Stercorariidae, Charadriiformes, 3.06]
Tropical Murrelets: The three tropical murrelets, Craveri's Murrelet, Synthliboramphus craveri, Scripps's Murrelet, Synthliboramphus scrippsi, and Guadelupe Murrelet, Synthliboramphus hypoleucus, have been transferred to genus Endomychura (Oberholser 1899, type hypoleuca).
[Alcidae, Charadriiformes, 3.06]
Terns and Skimmers: The Terns have been promoted to a family, Sternidae, with two subfamilies, Rynchopinae (skimmers) and Gyginae (white terns). The gull and tern families form the superfamily Laroidea.
[Sternidae, Charadriiformes, 3.06]
Gulls and Noddies: The Noddies (Anous = Anouinae) are treated as a subfamily of the gulls (Laridae) based on the genetic trees from Černý and Natale (2021). The relevant material can be found in Figures A-3 and A-5 of the supplementary material.
[Laridae, Charadriiformes, 3.06]
New Larid Genera: Based on Černý and Natale (2021), I've recognized three additional genera in the gulls:
The TiF list has adopted Figs. A-5 and A-6 from Černý and Natale for the Larus gulls.
[Laridae, Charadriiformes, 3.06]
Mew Gull: As in the AOS Supplement #62, the Mew Gull, Larus canus, is split into:
[Laridae, Charadriiformes, 3.06]
Magellanic Plover: The Magellanic Plover family, Pluvianellidae, has been demoted to a subfamily (Pluvianellinae) of the sheathbill family Chionidae.
[Chionidae, Charadriiformes, 3.06]
Thick-knees: Based on Paton et al. (2003) and Černý and Natale, (2021), the genus Burhinus is divided into three genera:
- An unnamed genus designated "Burhinus", which includes the Double-striped Thick-knee, "Burhinus" bistriatus, and the Peruvian Thick-knee, "Burhinus" superciliaris
- Burhinus itself is now reduced to a single species, the Bush Stone-Curlew, Burhinus grallarius
- Oedicnemus (Temminck 1815), type oedicnemus, the Eurasian Stone-Curlew. This genus includes all four other species formerly included in Burhinus.
[Burhinidae, Charadriiformes, 3.06]
Stilts and Avocets: The Banded Stilt, Cladorhynchus leucocephalus, is basal in family Recurvirostridae, which has been rearranged accordingly.
[Recurvirostridae, Charadriiformes, 3.06]
Far Eastern Oystercatcher, Haematopus osculans: Based on Senfeld et al. (2020b), the Far Eastern Oystercatcher, Haematopus osculans, has been split from the Eurasian Oystercatcher, Haematopus ostralegus.
[Haematopodidae, Charadriiformes, 3.06]
Oystercatchers: Based on Senfeld et al. (2020b), the American Oystercatchers have been separated in genus Prohaematopus (Matthews 1913), type ater.
[Haematopodidae, Charadriiformes, 3.06]
Charadriidae Genus Changes:
- The Hooded Dotterel is referred to as "Thinornis" cucullatus
- Forbes's Plover is referred to as "Afroxyechus" forbesi
- The Pied Lapwing / Pied Plover is transferred to the monotypic genus Hoploxypterus (Bonaparte, 1856)
- Vanellus is restricted to the Northern Lapwing
- Southern Lapwing and Andean Lapwing are transferred to Belonopterus (Reichenbach 1852, type chilensis)
- Until we know more, the other 20 lapwings have been placed in Hoplopterus (Bonaparte 1831, type spinosus). I did not find the total evidence tree useful here as it contradicts the available genetic data. I retained the previous arrangement of the Vanellinae, with the above genera pulled out.
- The Double-banded Plover, Anarhynchus bicinctus, is transferred to the monotypic genus Nesoceryx (Mathews, 1920).
- The New Zealand Plover, Anarhynchus obscurus, is transferred to the monotypic genus Pluviorhynchus (Bonaparte 1856).
- Finally, Ochthodromus is divided into Leucopolius (Bonaparte 1856), type marginatus, Helenaegialus (Mathews 1913), type sanctaehelenae, and Ochthodromus (Reichenbach 1852), type wilsonia.
[Charadriidae, Charadriiformes, 3.06]
Buttonquails: The buttonquail tree was constructed by combining the species groups described by Debus (1996) with the 7-species phylogeny of Černý and Natale (2021). I was surprised there were no conflicts.
[Turnicidae, Charadriiformes, 3.06]
Coursers and Pratincoles:
- The Double-banded Courser, previously Rhinoptilus africanus, has been transferred to Smutsornis (Roberts 1922, monotypic).
- The Gray Pratincole, Glareola cinerea, and Small Pratincole, Glareola lactea have been transferred to Galachrysia (Bonaparte 1856, type lactea). The movement that prompted this was based on DNA analyzed by Černý and Natale (2021).
- The Rock Pratincole, Glareola nuchalis, and Madagascan Pratincole, Glareola ocularis, have been transferred to Subglareola (Mathews 1913, type ocularis).
- I have restored the monotypic genus Stiltia (G.R. Gray, 1855, type isabella) which I had previously submerged in Glareola.
[Glareolidae, Charadriiformes, 3.06]
Jaegers and Skuas: The Stercorariidae have been rearranged slightly to conform with Černý and Natale (2021).
[Stercorariidae, Charadriiformes, 3.06]
Tropical Murrelets: The three tropical murrelets, Craveri's Murrelet, Synthliboramphus craveri, Scripps's Murrelet, Synthliboramphus scrippsi, and Guadelupe Murrelet, Synthliboramphus hypoleucus, have been transferred to genus Endomychura (Oberholser 1899, type hypoleuca).
[Alcidae, Charadriiformes, 3.06]
Terns and Skimmers: The Terns have been promoted to a family, Sternidae, with two subfamilies, Rynchopinae (skimmers) and Gyginae (white terns). The gull and tern families form the superfamily Laroidea.
[Sternidae, Charadriiformes, 3.06]
Gulls and Noddies: The Noddies (Anous = Anouinae) are treated as a subfamily of the gulls (Laridae) based on the genetic trees from Černý and Natale (2021). The relevant material can be found in Figures A-3 and A-5 of the supplementary material.
[Laridae, Charadriiformes, 3.06]
New Larid Genera: Based on Černý and Natale (2021), I've recognized three additional genera in the gulls:
- The Slender-billed Gull, Chroicocephalus genei, is separated from the other Chroicocephalus by about 10 million years. It is also distinctive, and I've moved it to genus Gelastes (Bonaparte 1856, monotypic).
- The division between Dolphin and Gray Gulls on one hand, and Laughing, Franklin's, and Lava Gulls on the other is about 6 million years, enough to support different genera. Moreover, they obviously form two, or even three groups. Accordingly, the Laughing, Franklin's and Lava Gulls have been moved to Atricilla (Bonaparte 1854, type atricilla). It would not be unreasonable to also split the Gray Gull, Leucophaeus modestus. In that case it would be genus Blasipus (Bruch 1853).
- The band-tailed gulls (4 species) are transferred to genus Gabianus (Bruch 1853, type pacificus).
The TiF list has adopted Figs. A-5 and A-6 from Černý and Natale for the Larus gulls.
[Laridae, Charadriiformes, 3.06]
Mew Gull: As in the AOS Supplement #62, the Mew Gull, Larus canus, is split into:
- Common Gull, Larus canus
- Short-billed Gull, Larus brachyrhynchus
[Laridae, Charadriiformes, 3.06]
Jim LeNomenclatoriste
Je suis un mignon petit Traquet rubicole
Forgotten Scolopacidae :
Lymnocryptes Issues: The sample of the Jack Snipe (Lymnocryptes) used by Černý and Natale (2021) seems to be a chimera, so I have ignored their results for it. Baker et al. (2007) found Lymnocryptes sister to a dowitcher (they used Limnodromus scolopaceus), so I have put it there.
New Zealand Snipe: The New Zealand Snipes (Coenocorypha) are now considered to include 5 exant and recently extinct species based on Baker et al. (2010) and Worthy et al. (2002). The Snares and South Island Snipes are quite closely related, with an estimated divergence time of about 50,000 years. Their status as separate species rests on the lack of an aerial display for the Snares Snipe, as well as genetic and plumage differences. The other Coenocorypha are somewhat more distant relatives.
Gibson (2010) and Gibson and Baker (2012) found that Imperial Snipe is more closely related to the New Zealand snipes than to the other snipes. The Imperial Snipe and two other snipes have sometimes been separated as Chubbia (Mathews 1913), and that is how I treat them here. It is clear that this arrangement of the snipes is not fully satisfactory, and they deserve further study.
Gallinago Split: Černý and Natale (2021) made clear there is a deep division in Gallinago, which I recognize by putting them in separate genera.
- The Great Snipe, Gallinago media, Solitary Snipe, Gallinago solitaria, Wood Snipe, Gallinago nemoricola, Swinhoe's Snipe, Gallinago megala, Pin-tailed Snipe, Gallinago stenura, and Latham's Snipe, Gallinago hardwickii are all transferred to genus Telmatias (Boie 1826, type stenura).
- The other 8 Gallinago snipes remain in Gallinago.
Willets: The Willet, Tringa semipalmata, has been split into Western Willet, Tringa inornata, and Eastern Willet, Tringa semipalmata, based on Oswald et al., (2016).
Tringa Split: Černý and Natale (2021) provide a better-calibrated tree than past efforts. There's not a good calibration point near the Tringinae, so I assume ages are substantially over-estimated. Nonetheless, it provides some guidence concerning where to draw genus boundaries. The previous list had used three subgenera for Tringa and I'm promoting them to regular genera. As a result,
- Tringa (Linnaeus 1758, type ochropus) is restricted to the Green Sandpiper, Tringa ochropus, and the Solitary Sandpiper, Tringa solitaria
- The two tattlers are transferred to genus Heteroscelus (Baird 1858, type brevipes)
- The remaining Tringa sandpipers are placed in genus Totanus (Bechstein 1803, type totanus).
Aphriza merged into Calidris: It has long been suspected the Surfbird is close to the knots (e.g., Jehl, 1968). This is exactly what Bororwik and McLennan (1999) found in their DNA tree. Indeed, their results suggest the Surfbird and knots are congeneric. The recent analysis by Gibson (2010) and Gibson and Baker (2012), using additional data, concured, as did Černý and Natale (2021). Based on this, I've merged Aphriza into Calidris.
Tribes in Arenariinae: Černý and Natale (2021) have a calibration point that somewhat constrains dates for the broad Calidris sandpipers. To me, it suggests dividing the subfamily Arenariinae into three tribes: Prosoboniini (Bonaparte 1850), Arenariini, and Calidrini (Reichenbach 1849 (1838)). This creates the handy term Calidrine, which can be used to refer to the species formerly in the very broad Calidris.
Černý and Natale's results alao suggest that the broad genus Calidris should be split into a number of genera. I had previously done such a thing, and then undone it after the AOU, BOU, and H&M 4th ed. have all merged all of these species into Calidris, as in Banks (2012).
Well, I'm undoing it and then some!
- Calidris (Merrem 1804, type canutus) is restricted to three species: Great Knot, Calidris tenuirostris, Red Knot, Calidris canutus, and Surfbird, Calidris virgata.
- The Ruff, Calidris pugnax, is transferred to Philomachus (Merrem 1804, monotypic).
- Broad-billed Sandpiper, Calidris falcinellus, and Sharp-tailed Sandpiper, Calidris acuminatus are transferred to Limicola (Kaup 1816, type falcinellus).
- Curlew Sandpiper, Calidris ferruginea is transferred to the monotypic genus Erolia (Vieillot 1816).
- Stilt Sandpiper, Calidris himantopus is transferred to the monotypic genus Micropalama (Baird 1858).
- Temminck’s Stint, Calidris temminckii, Long-toed Stint, Calidris subminutus, Red-necked Stint, Calidris ruficollis, and Spoon-billed Sandpiper, Calidris pygmeus are placed in Eurynorhynchus (Nilsson 1821, type pygmeus).
- Buff-breasted Sandpiper, Calidris subruficollis is placed in the monotypic genus Tryngites (Cabanis 1856).
- Sanderling, Calidris alba, Dunlin, Calidris alpina, Purple Sandpiper, Calidris maritima, and Rock Sandpiper, Calidris ptilocnemis are transferred to Pelidna (Cuvier 1816, type alpina).
- Baird's Sandpiper, Calidris bairdii, Little Stint, Calidris minuta, Least Sandpiper, Calidris minutilla, White-rumped Sandpiper, Calidris fuscicollis, Pectoral Sandpiper, Calidris melanotos, Semipalmated Sandpiper, Calidris pusilla, and Western Sandpiper, Calidris mauri are transferred to Ereunetes (Illiger 1811, type pusilla).