AOU Committee on Classification and Nomenclature (North & Middle America)
Votes on Proposals 2010-B
2010-B-1: Transfer eight species from the Turdidae to the Muscicapidae
2010-B-2: Transfer three species of muscicapids to different genera
2010-B-3: Revise two type localities
2010-B-4: Change Amazilia chionopectus to Amazilia brevirostris
2010-B-5: Create a new order, Pteroclidiformes, for the sandgrouse
2010-B-6: Recognize Anas diazi as a species—again
2010-B-7: Change the linear sequence of the furnarioid families
2010-B-8: Recognize the Bahama Warbler Dendroica flavescens as a distinct species
2010-B-9: Recognize Geothlypis (aequinoctialis) chiriquensis as a distinct species
2010-B-10: Adopt a new generic classification for the Parulidae
2010-B-10-Amendment #1: Retain Leucopeza semperi and Oporornis agilis in their current genera (versus lumping into Geothlypis)
2010-B-10-Amendment #2: Retain Catharopeza bishopi in its current genus (versus lumping into Setophaga)
2010-B-11: Restore authority for the genus name Peucedramus to Coues
2010-B-12: Split Mexican Jay Aphelocoma ultramarina into two species
2010-B-13: Lump Ramphastos swainsonii with R. ambiguous
Transfer eight species from the Turdidae to the Muscicapidae
YES - 8 without comment.
YES. The data seem quite clear on this point.
YES. Very clearly needed.
YES. Solid data make this the only option.
Transfer three species of muscicapids to different genera
YES - 2 without comment.
YES. No need to wait.
YES. Contra to the motion, Bluethroat is regular in small numbers in the UK. The only two that share the same behavior (tail quivering) are Rufous-tailed Robin and Siberian Blue Robin, and I am pleased to see that they work out to be in the same genus.
NO. I buy that L. cyane and sibilans need a new name given the published tree. However, only Bayesian analyses of combined data definitely separate L. svecica and calliope from one another. I'd consider a separate proposal for recognition of Larvivora.
NO. Although surely there will be changes, I agree that further corroboration that these are the right changes to make is needed, especially given the authors’ statement of uncertainty regarding this.
NO. I agree with other comments that the support in the trees for the subclades look weak. There are potential concerns for the stability of the names as well. Both Calliope and Larvivora are old names (both 1837), but Brachypteryx in Clade D3a with Larvivora is older (1821). It would seem like we’d need to decide that we wanted to split up D3a before we could use Larvivora for sibilans and cyane. However, I don’t know the type species for Brachypteryx, and it could be that a species of Brachypteryx not in the tree and not in Clade D3a is the type, in which case the nomenclature would require yet more tweaking. The comment about not having any basis other than genetics for recognizing these clades as genera is also a concern, although the comment about tail quivering gives one hope that they could be characterized by morphology or behavior. I am always a little concerned about making changes based on the twiggery of a tree that was really looking at the deeper branches.
NO. Wait for better data.
NO. For two reasons: 1) The division of Clade D in Sangster et al. into subclades to which they would like to apply new generic names appear to be based entirely on molecular characters rather than morphological ones. Yet there are many clades that are not highlighted as such for which generic status might be equally strongly argued. Very little attempt has been made to use phenotype to help guide where generic limits should be placed within Clade D. I agree with Sangster et al. 2010 and the proposal authors that these three taxa belong in different genera, but I am unconvinced that adequate evidence has been presented to determine exactly which genera (what generic limits) will apply when this is completely resolved. 2) As the proposal states, “details of the composition of these clades may be subject to change” This also means that nomenclature may also change. Why adopt names that might be short-lived before this situation is better clarified? Finally, Clade D in Sangster et al. appears to contain a number of clades that are not supported by most or all individual gene trees, yet which have high Bayesian support and low or no ML support, including two of those units that this proposal focuses on (D3d and D3a). This, coupled with missing taxa, makes me leery to adopt any change at this time.
NO. I agree that something needs to be done with Luscinia but this seems like a piecemeal approach. There are other species within the three clades containing 'Luscinia' that need to be dealt with as well, and I think it's better to wait and treat them all at once. I also agree with others regarding the statement of uncertainty and with the problem of low overall support (except Bayesian) at the base of those clades.
NO. I agree with other no votes on poor support of nodes.
Revise two type localities
YES - 10 without comment.
YES, but for Spizella, why not include the entire recommended type locality given in Banks and Gibson (= “Western North America, California, New Mexico = Black Hills, Dak[ota Territory] = Laramie….”)?
Change Amazilia chionopectus to Amazilia brevirostris
YES - 9 without comment.
YES. SACC has already adopted this. Be sure to also cite Chebez et al. (2008).
YES. This is a convoluted mess, but I think that Weller and Schuchmann got it right.
Create a new order, Pteroclidiformes, for the sandgrouse
YES - 8 without comment.
YES. Demanded by multiple independent datasets. This order has a long tradition in bird classification.
YES, but—Was this ordinal name used in any of the references cited? I am concerned that the check-list may not be the proper place to formulate a new name. I would rather that it be created elsewhere and adopted by us as, I believe the ordinal names we inserted last year were.
YES. I am not sure why it would be bad for us to coin the new order, especially given that it is mono-familial, as long as we have the spelling correct.
Recognize Anas diazi as a species—again
YES - 4 without comment.
YES. I was really on the fence about this, because the published genetic
evidence isn’t the greatest when it comes to species limits. However, it does show us that
a new dichromatic lineage on the continent is hybridizing to some degree into an endemic
monochromatic lineage that we would ordinarily call a species. Further, Rhymer (2006,
Acta Zoologica Sinica) suggests that there are still genetically pure populations in central
Mexico. So the biology looks good, though it should become a priority for new research to make it better, because this decision may cause it to become re-listed as federally
Threatened or Endangered.
NO. While the data of McCracken et al. are suggestive, they are consistent with a number of possible scenarios. For instance, the pattern of intergradation described by Hubbard's work reminds me quite strongly of the pattern seen in a number of other North American taxa (e.g., Dendroica coronata complex, Juncos, chipping sparrows), where resident southern taxa seem to have given rise relatively recently to migratory forms. I see no reason why this couldn't have been linked to the evolution of sexual dichromatism. It seems to me that the pattern of haplotype sharing between mallards and Old World versus New World taxa could just as easily be explained by the reverse of the scenario proposed by McCracken et al.--invasion of the Old World by a North American native mallard that then acquires Eurasian haplotypes by hybridization. I think we need to see sampling and careful analysis of multiple loci from both Old World and New World taxa in order to have this worked out properly.
NO, for now. I do not see how Hubbard’s work can be discounted. He showed widespread introgression with Mallards in the northern half of the Mexican Duck range. Even if the Mallard is not its closest relative, there seems to be less reproductive isolation between them then between Mallard/American Black Duck and Mallard/Mottled Duck (at least historically). I would like to see the supposed introgression explained as something other than rampant hybridization and lack of reproductive barriers. Whether we should lump Mottled and American Black Duck with Mallard is another question.
NO. Although diazi may indeed be closer to rubripes and fulvigula than any is to platyrhynchos, not only is that result based on an mtDNA gene tree but also the N for diazi was only 4 individuals. Further, those 4 came from San Luis Potosí, 500+ km from the contact zone and deep within the area with phenotypically pure diazi. That such a tiny, geographically restricted sample would fairly represent haplotype diversity in diazi seems highly unlikely, and that it would show any sign of gene flow from not only seems unlikely.
Further, reliance on just that data set dismisses completely the large data set constructed by Hubbard (1977) on the geography of phenotype variation. Using classical hybrid index scores, Hubbard grouped his samples from the range of diazi (N >100 museum specimens) into 8 geographic regions and found that his NM and AZ samples all had average scores that he interpreted as diazi X platyrhynchos, and his samples south of the border south to Durango averaged as either “nearer diazi” or “very near diazi”, i.e. showing signs of introgression. Only his two southernmost samples, from Zacatecas to Puebla (which includes the region from which the 4 genetic samples were taken) were phenotypically pure diazi. Therefore, at least in terms of the signal from phenotype scores, there is evidence of extensive gene flow between the groups. Hubbard also found evidence of increasing hybridization through time, with his 1938-1960 sample from the contact zone showing a 15% increase in % hybrids over his pre-1920 sample; 88% of that later sample consisted of birds showing phenotypic signs of hybridization.
Hubbard also reported results of an unpublished field study (Nymeyer 1975) that suggested that mating was assortative: mixed pairs in the Rio Grande Valley only constituted 9% of the pairs studied. However, the assessment of whether females were diazi, platyrhynchos, or hybrids was based on field identification, which Hubbard considered unreliable. Nonetheless, Hubbard noted that there may be behavioral and ecological differences between the two in the contact zone and was impressed by the abrupt shift from diazi-like birds to essentially pure Mallards in n. New Mexico.
Even if diazi is more closely related to fulvigula and rubripes than any are to platyrhynchos, there is no reason to expect a priori that the nature of their contact zones would be the same. For starters, as noted by Hubbard, female platyrhynchos are more similar to diazi than they are to females of the two other, much darker-plumaged groups, so on that basis alone, one might make a bold prediction of sloppier mate choice. More broadly, paraphyletic species are a predicted outcome of evolution at the population level and are fairly frequent in birds (Kevin Omland papers).
What is needed here is a modern investigation of the nature of the contact zone between the two taxa.
NO - 2 originally voted YES but changed vote to NO based on other comments.
NO. Originally voted YES but changed vote to NO based on other comments. I should have looked at Hubbard. I recalled the long-held conventional wisdom that all birds in the U.S. were hybrids...i.e. no pure Mexican Ducks. I don't think that's the case recently from the many I've seen recently in the Southwest, but have hardly tried to look at them critically, something that needs to be done. Perhaps as distinct from Mallard as Mottled and American Black, but for now I'm happy to keep the status quo.
Change the linear sequence of the furnarioid families
YES - 7 without comment.
YES. I agree with the change in sequence, but think we should retain Furnariidae subfamilies for now (a different proposal should be submitted to elevate them to family).
YES. The proposed sequence more accurately reflects the phylogeny of this group of families.
YES. Let’s leave the Furnariidae subfamilies as subfamilies.
YES, retaining Furnariid subfamilies. One question though. Why Thamnophilidae first rather than last? In our current sequence Furnariidae is first. I guess unless there is a compelling reason to do so, I would continue with that, so the new sequence would be:
Recognize the Bahama Warbler Dendroica flavescens as a distinct species
YES - 3 without comment.
YES. If you measure one, or look at one, or hear one sing, or sequence its mtDNA, you know what it is with probability of 100%. What else do you want?
YES. The original merger of flavescens into dominica was not based on strong evidence for conspecificity, and the congruence of genetic (albeit only a short fragment of control region sequence from a small number of individuals), ecological, phenotypic, and vocal data (especially the lack of response to songs of the other taxon) all argue that flavescens should be resurrected as a species.
YES, although just barely. Sample sizes are small, but the preponderance of
evidence, including degree of plumage difference, overwintering dominica and song
playback, and habitat specialization argue in favor of species status.
YES. But we have Bahama Yellowthroat already, so there are two endemic warblers now in the Bahamas. Interestingly, despite an extended spring trip to Abaco, no flavescens were singing, yet wintering nominate dominica were singing. This is in early April. Their call notes were alike. Resident flavescens certainly was more prone to forage on tree trunks (Caribbean Pines). I still can't figure out why I didn't hear a single flavescens sing.
YES. Although plumage and morphological differentiation may be somewhat less between flavescens and dominica than in some other pairs of Dendroica warblers, the whole genus has undergone hyper plumage differentiation in a short amount of time. Given the foraging specialization, vocal differentiation, and plumage differences, I think that these two taxa would behave reproductively isolated if they came in contact. Audubon’s and Myrtle Warblers have less plumage and especially morphological differentiation, more similar songs (but not calls), and nearly no foraging/ecological differences, and yet they are right at the cusp of being species level taxa.
NO. The behavioral and plumage differences between the Bahama and mainland subspecies are small, similar to those among other passerine subspecies and smaller than is often the case between mainland and island conspecifics. The magnitude of mtDNA divergence is also very small. Retaining the Bahama form as a subspecies of D. dominica seems to me to be most consistent with long-standing AOU practice. We should expect island forms to have some level of differentiation and only split them at the species level when there is good evidence of a sustained history of isolation and/or substantial character divergence
NO. At least for now. That flavescens has unique haplotypes only reflects its geographic isolation and per se is irrelevant to species limits. The plumage differences indicate that it is a valid taxon, but not necessarily a species under BSC; parulid taxa ranked as subspecies differ in plumage characters of the same degree of difference as does flavescens and mainland dominica. As for habitat, parulids are notoriously flexible in breeding habitat, even within monotypic species (e.g., Swainson’s). Even within mainland Yellow-throated, breeding habitat varies from pine-dominated (much as in flavescens, just a different Pinus) to 100% broadleaf. So, that leaves voice. With many parulids notoriously variable in voice, with many species having multiple song types, a more through analysis is required here, including some playback trials. Bahama flavescens may indeed have diverged to the level consistent with species rank, but I think we need a more rigorous analysis.
NO. This is close to the line, but I remain more comfortable with it as a subspecies of dominica than as a separate species.
Recognize Geothlypis (aequinoctialis) chiriquensis as a distinct species
YES - 5 without comment.
YES. In contrast to 2010-B-8, the genetic evidence here is compelling, and suggests that the earlier classifications within this complex were confused by its low degree of morphological differentiation in plumage traits.
YES. I am not completely happy with this. I think that we need to consider the semiflava piece of this as well. My sense is that we would not place chiriquensis into an expanded semiflava, so I think we can split chiriquensis now, but semiflava clearly needs more work.
NO. Based on the mtDNA data, it looks like semiflava needs revision. I would recommend waiting on this one until more extensive sampling of semiflava is done (preferably with multilocus data).
NO. Although I agree that chiriquensis may be a valid species, this study included only a single sample of this form and relied entirely on mtDNA. This is suggestive but I don’t consider the evidence sufficient to change the status of chiriquensis. I vote NO pending additional sampling and additional types of data that bear on the question of species status.
NO for now. Clearly the Costa Rican sample of G. aequinoctialis is separate from the South American samples of that species. However, as the proposal points out - and assuming that the semiflava (bairdi) from Costa Rica is identified correctly - the relationships are (semiflava (semiflava [bairdi], aequinoctialis [chiriquensis])) with strong support. Thus, I think we need to treat semiflava at the same time. Further, this study is based only on mtDNA from a relatively small number of individuals, and I would like to see corroboration from other data (nuclear DNA, voice).
NO. At least for now. I strongly suspect that additional, more rigorous analyses will provide the data for ranking chiriquensis as a species, but as noted in the proposal, there are sufficient doubts at present that additional data are needed: bigger N for individuals, localities, and loci examined.
NO. This proposal rests entirely on the MtDNA sequence of one individual. Too much opportunity for ID or mislabeling problems. Some sort of vocal/ecological/morphological analyses to back up the semiflava/chiriquensis link would be nice.
Adopt a new generic classification for the Parulidae
YES - 2 without comment.
YES. This is the state of the art, and unlikely to be revised for some time.
YES. Originally voted NO, but changed vote to YES based on other comments and addition of amendments #1 and #2. I agree that this is a well-resolved, thorough phylogeny that is the current state of knowledge for this group, so I think we should go ahead with the changes.
YES. Get it over with and too bad about Dendroica. I still have a hard time thinking that Setophaga is just another Dendroica, or put correctly that all Dendroica are now Steophaga. I would have guessed that Mniotilta was allied with Dendroica, so by priority they would all be Mniotilta. Either way Dendroica loses out. It will be interesting to see how Oporornis agilis finally sorts out. It is very different from the other Oporornis.
YES. Originally voted NO, but changed vote to YES. The lumping of Phaeothlypis in Myiothlypis is fine, as is Baileuterus. Euthlypis could still be retained, but it's not necessary. I think the expanded Geothlypis (with Leucopeza and Op. agilis) works fine.
YES. This phylogeny is exceptionally complete in terms of taxon sampling and gene sampling, so I doubt that we’re going to get additional data that will change the branching pattern any time soon. The proposed generic limits are consistent with the nodes with strong support, and actually make lots of “sense” in terms of warbler biology. However, I was tempted to vote NO so that we wouldn’t have to adopt the entire classification as is because (1) there are indeed a few alternatives in some cases, and (b) the phenotypic integrity of some genera (e.g., inclusion of Leucopeza and “O.” agilis in Geothlypis) could be improved with some tweaking. Nonetheless, I think those tweaks could be addressed in subsequent proposals – the broad classification is a good starting point.
YES. I hate to lose Dendroica and we could argue about whether to continue to recognize some monotypic genera that are basal to some of the big clades, e.g. Catharopiza, Euthlypis, but this looks pretty solid, and I can't see we gain from splitting off the basal taxa of some of the big clades.
NO for now. Wait for clarification and proposals on individual cases.
NO for now. This is a tough one. This situation is much cleaner than 2010-B-02, but there are a number of options available. As much as I liked the Lovette et al. 2010 paper, and I know the taxa involved well enough to do a lot of reading between the lines, I do not think there has been enough effort to confidently place generic limits in the context of morphology. “What is a genus?” is a really tough question, but in my mind it is not defined solely by a well supported molecular clade. If we go the route of just choosing molecular clades that we like as genera, we’re doomed to a lot of future nomenclatural instability. I think we should change generic limits when we have robust phenotypic characteristics upon which to rest them. As I look at the tree I can see that doing that is possible (though in places like Leucopeza it’s really tough), but that several possibilities exist and we probably need to be roughly evenhanded across the family.
ABSTAIN - 1 without comment.
Retain Leucopeza semperi and Oporornis agilis in their current genera (versus lumping into Geothlypis)
YES - 8 without comment.
Retain Catharopeza bishopi in its current genus (versus lumping into Setophaga)
YES - 7 without comment.
YES. I vote to retain Catharopeza in its genus, or rather to take no action on merging it. It seems to me that there should be clear and compelling evidence to change the status quo, which is perhaps lacking in this case, at least for now.
YES. Both Kepler and Parkes 1972 (based on perceived similarities between Catharopeza bishopi and Dendroica angelae) and Bond 1972 (in his annual supplement to his 1956 Checklist of Birds of the West Indies, following Kepler and Parkes’ publication) suggested that Catharopeza be merged into Dendroica. The AOU did not follow this for either the 1983 or 1998 editions. Bond, presciently, had remarked in 1956 that Catharopeza is probably an aberrant, primitive Dendroica, and we now have molecular data that indicate that it is indeed an aberrant, primitive “Dendroica” (= Setophaga), so perhaps we should simply merge it into Setophaga. However, Bond 1972 notes that there is actually “little similarity” between bishopi and angelae other than an eye-ring in bishopi and a broken eye-ring in angelae. The immatures are completely different (“quite distinct” in Bond), bishopi completely dark brown and very unlike any Dendroica, and angelae olive-green and warblerlike (as are immatures of plumbea and pharetra). According to Bond the song of Catharopeza is “unlike that of any species currently associated with the genus Dendroica.” Robbins and Parker 1997 obviously considered Catharopeza to be phenotypically distinct, publishing an entire paper devoted to the question of its relationships (“What is the closest living relative of Catharopeza”). They concluded – on vocal, plumage, structural, and behavioral grounds – that Catharopeza is not closely related to Dendroica.
The molecular split between bishopi and Setophaga is better supported than that between agilis-Leucopeza and Geothlypis. In Fig 2 (combined data) both splits are supported at posterior probability of 1.0. In Fig 4 (mitochondrial), the probability of bishopi separate from Setophaga is 1.0, whereas that for agilis-Leucopeza separate from Geothlypis is 0.88. Neither split is supported in the tree using only the nuclear introns (Fig 3): agilis is in the middle of the Geothlypis clade and bishopi forms a clade with Wilsonia citrina, Dendroica magnolia, and the 3 West Indian Dendroica. The probability that agilis and Leucopeza are sisters ranges from 0.51 to 0.57. There is no support in any of these trees for the superspecies mentioned by Kepler and Parkes (plumbea, pharetra, angelae, andbishopi).
NO. I vote "NO" on keeping Catharopeza, given the conflict in the molecular data, and its uncertain affinities based on phenotypic characters.
Restore authority for the genus name Peucedramus to Coues
YES - 10 without comment.
YES. If this were a novel treatment, then I think we should require that the proposal be published separately; however, all we are doing is switching to an alternative treatment.
Split Mexican Jay Aphelocoma ultramarina into two species
YES - 4 without comment.
YES, seems well supported.
YES. According to McCormack et al. 2008, these taxa are distinguishable over >95% of the time by phenotype alone (a combination of plumage traits and standard external measurements), they are geographically disjunct, reciprocally monophyletic in mtDNA, and share no haplotypes in two autosomal loci (McCormack et al. 2011). These latter loci don't demonstrate reciprocal monophyly, but this is a very high bar for autosomal genes (e.g., red-winged blackbird and tricolored blackbirds typically don't achieve this, and in fact fall out almost identically, with tricolored nested within red-wing variation). Although there is evidence of gene flow within A. wollweberi (at least between central and eastern populations), there is no hint of it between A. wollweberi and A. ultramarina. All of this points to species recognition.
YES. Really? Diverged in the Miocene? I find that hard to accept.
YES. Note that if the type reference given for A. wollweberi is correct, then Kaup on the preceding line should not be in parentheses.
YES, for reasons given in the proposal.
NO. I would like to see some vocal data to support a split. For what's is worth the Texas birds and Arizona birds sound identical to my ear. Bottom line: how about some additional studies of the Transvolcanic birds in the field?
NO. The rationale of this proposal is concentrated on the 9% divergence in mtDNA, the reciprocal monophyly, and calibrations that suggest a Miocene split … in other words, that the genetic differences between Ultramarine and Mexican are consistent with species rank. However, these degrees of genetic divergence are really not that different from population-level differences within species of sedentary tropical birds. Of course compared to the North American species that this Committee typically deals with, these genetic differences are large. However, they are really not impressive when compared to those among populations ranked as members of single species of tropical birds (e.g., phenotypically indistinguishable populations of West African Bleda syndactylus that differ by 8-11% in sequence divergence).
Further, having spent some time roaming through our good series of these jays, all I see (with all appropriate caveats) are minor differences in size and color that are expected in geographic variation within taxa we rank as species. In step-clinal fashion, the intensity and extent of blue on the head increases from northernmost arizonae southward to the ultramarina group. The differences between these two groups are less than those between populations of Steller’s Jays, Gray Jays, and Green Jays, for example, or even among the Scrub Jay species group. I encourage those of you with access to collections to check the skins; if you don’t, check Plate 2 in Madge & Burns “Crows and Jays” book. Looking at Howell & Webb (1995), who were vigilant in mentioning species-level phenotypic divergence in voice and other features within species of Mexican birds, nothing is mentioned concerning any differences; also, consistent with an absence of major plumage differences, they only illustrated wollweberi (in contrast to 3 other jays with separate subspecies illustrated). In fact, Howell & Webb’s illustration of Western Scrub-Jays within Mexico (subspecies hypoleuca vs. cyanotis) shows as much or more variation within Mexican subspecies of the latter than exists between the proposed separate species of Mexican Jays. Obviously, I do not intend these observations to represent hard evidence one way or another, but only to describe qualitative levels of the plumage differences involved. So, without some other data (besides genetic divergence) on the degree of differentiation between these two groups, e.g., vocal or behavioral, I see no real evidence for species rank. If we base species rank solely on genetic distance and reciprocal monophyly, then we might as well turn over these decisions to the bar-coders.
Lump Ramphastos swainsonii with R. ambiguous
YES - 3 without comment.
YES. The evidence suggests that subspecies status is warranted.
YES. Although the data are weak, all existing data point towards conspecificity. There are no known vocal differences between the two, and the only differences between them are in characters not associated with breaks in gene flow in other toucans (e.g., bill color). There is absolutely no evidence so far that these two merit species rank.
YES. Although I think it is desirable to have the NACC and SACC lists conform as much as possible, that is not driving my vote in this case. The evidence is fairly ambiguous, but I think leans more toward subspecies rather than species status.
YES, given that they are not more distinct in any way than other confamilial taxa treated as subspecies. If it could be established that facial and bill coloration causes assortative mate choice, I would vote to split.
NO. It seems to me that there is not enough information to act either way. I am content to have occasional differences between the two lists, and would not vote to change merely for conformity.
NO. Evidence is mixed, but not enough to change anything. I think that is what "No" means.
NO. There isn’t good evidence to argue for either treatment in my view.