Proposal (873) to South American Classification Committee

 

Modify species limits in Forpus: (A) Treat Forpus crassirostris as a separate species from F. xanthopterygius, and (B) Treat Forpus spengeli as a separate species from F. passerinus

 

Background: The distinctive, morphologically homogeneous Parrotlet genus Forpus is usually treated as being comprised of seven species (e.g. Forshaw 1973, Sibley and Monroe 1993, Dickinson 2003, Dickinson and Remsen 2013, Clements et al. 2019), although Peters (1937) only recognized five. Most are allopatric, with only one species (F. modestus) overlapping broadly geographically with other species. All are sexually dichromatic, and most are polytypic. Not surprisingly, species limits have long been contentious, and nomenclatural issues have caused further confusion (e.g., Collar 1997, Juniper and Parr 1998, Whitney and Pacheco 1999, SACC proposal #4).

 

The most widespread species as currently recognized by most authorities is Blue-winged Parrotlet Forpus xanthopterygius. Its member taxa were often treated as three different species: F. xanthopterygius (= vivida); F. crassirostris; and F. spengeli (e.g. Ridgway 1916, Cory 1918), while others (e.g., Hellmayr 1907, Peters 1937) considered them all races of Green-rumped Parrotlet Forpus passerinus. Gyldenstolpe (1945, not seen), however, showed that crassirostris and passerinus are narrowly parapatric in western Brazil, without evidence of intergradation (Juniper and Parr 1998, Whitney and Pacheco 1999), and on this basis and their obviously different rump colors, he and subsequent authors have mostly treated them as separate species (although with crassirostris as a subspecies of xanthopterygius). Collar (1997) and Juniper and Parr (1998) have suggested that spengeli may be more closely related to or conspecific with the broadly allopatric Mexican Parrotlet Forpus cyanopygius. For a more in-depth summary of the taxonomic history of F. xanthopterygius, see Bocalini and Silveira (2015).

 

New information:

 

Smith et al. (2013), in a phylogenetic analysis of mtDNA and nuclear loci of all species and most subspecies of Forpus, found that spengeli of northern coastal Colombia is embedded (on the basis of mtDNA only, no nuclear data being available) within F. passerinus rather than F. xanthopterygius (see their Fig. 1 below). Thus, although Dickinson (2003) had treated spengeli as a subspecies of xanthopterygius, Dickinson and Remsen (2013) treated it as a race of passerinus, and Remsen et al. (2020) provide the rationale. However, this treatment does not address the seemingly considerable morphological disparity between spengeli and other subspecies of passerinus, especially F. p. cyanophanes of arid north-eastern Colombia (between the Santa Marta and Perijá mountains). These two appear to be essentially parapatric, but cyanophanes has conspicuous, extensive violet-blue on upper- and underwing coverts, quite unlike those of spengeli (see photo below), which also has a brilliant turquoise rump (vs. green in cyanophanes). In addition, this change to species attribution of spengeli appears to have been made solely on the basis of mtDNA.

 

Smith et al. (2013) also found evidence that crassirostris is sister to the clade comprised of most Forpus taxa, except modestus and cyanopygius. This result was strongly supported on the mtDNA tree but not well supported in the nuclear DNA and species tree.

 

 

Fig. 1 of Smith et al., mtDNA

 

 

Fig. 3 of Smith et al., species tree

 

 

Bocalini and Silveira (2015) analyzed geographic variation in morphology of 518 specimens of the F. xanthopterygius complex, and concluded that spengeli should be considered a distinct species (see their Fig. 1 below). However, their study did not evaluate the possibility that spengeli may be conspecific with F. passerinus. They also considered that crassirostris (along with the other subspecies of xanthopterygius traditionally recognized) is not diagnosable phenotypically and should thus be considered a synonym of F. xanthopterygius, which they treat as monotypic (Bocalini and Silveira 2015). However, they did confirm that crassirostris is smaller overall than the nominate (with overlap, see their Fig. 3 below), which they attribute to Bergmann’s Rule, given its more northerly (Southern Hemisphere) range. However, they did not place this finding in context of other ecogeographic studies, and it does not seem clear from the literature that Bergmann’s Rule applies in any consistent way to fauna of tropical and subtropical lowlands.  Also, Bocalini and Silveira (2015) did not address the other morphological differences summarized in Cooper (1973): “like xanthopterygius, but all blue markings paler; primary-coverts pale greyish violet-blue contrasting with darker violet-blue secondary-coverts; upper mandible compressed laterally at the centre”, or its relatively large bill (Hellmayr 1907) so their study does not negate the putative existence of these differences.

 

 

 

Fig. 1 (part) from Bocalini and Silveira (2015); spengeli above, xanthopterygius below

 

 

Fig. 3 from Bocalini and Silveira (2015); Factor 1 is a general size axis and Factor 2 is mainly influenced by culmen length.

 

Donegan et al. (2016) reexamined the question of whether spengeli should be split from xanthopterygius under the view that the best yardstick is whether differences exceed those between sympatric species of the same genus. From examination of AMNH specimens (see their Figs. 3-4, below) they determined that differences between spengeli and xanthopterygius were substantial, especially compared to those between F. modestus and F. xanthopterygius, and in addition noted that spengeli is found in drier habitat. They also compared spengeli with F. passerinus viridissimus at AMNH (see their Fig. 5 below) and noted further plumage distinctions, and they discussed the potential for a contact zone between viridissimus and spengeli and the lack of clear evidence for intergradation (Donegan et al. 2016).

 

 

 

 

Fig. 5 from Donegan et al. (2016). In each, the two specimens on the left are F. passerinus viridissimus and the two on the right are spengeli.

 

In summary, although mtDNA places spengeli within the F. passerinus clade, and it clearly does not belong with F. xanthopterygius, it is as distinctive morphologically as most other Forpus treated as species and it appears to be parapatric, without reported intergradation to my knowledge, with the quite different-looking F. passerinus cyanophanes. Although SACC (Remsen et al. 2020) treats spengeli as a subspecies of passerinus, Clements et al. (2019) maintain it within xanthopterygius, and del Hoyo and Collar (2014) and Gill and Donsker (2015) consider spengeli a full species, the aptly descriptive Turquoise-winged Parrotlet.

 

And, although crassirostris (including the sometimes recognized ollalai of east-central Amazonas) is only subtly distinct in plumage, the mtDNA tree places it as sister to most other Forpus (except modestus and cyanopygius, and it differs from other taxa of F. xanthopterygius in its smaller size but relatively larger bill, but with a reportedly laterally compressed culmen. Although other authors maintain crassirostris within xanthopterygius, Gill and Donsker (2015) consider it a full species (as was done by Ridgway 1916 and Cory 1918), adopting the common name Large-billed Parrotlet from Cory (1918) for crassirostris.

 

Another option would be to reunite all these taxa (xanthopterygius s.l. + passerinus s.l.) under Forpus passerinus, as in Hellmayr (1907) and Peters (1937), but that is argued against by the greater morphological disparity of such a grouping relative to other Forpus species, the greater branch length on the mtDNA tree than between the undisputed species in the coelestis + xanthops and conspicillatus clade, and the two zones of apparent parapatry (between spengeli and cyanophanes in northeastern Colombia and between crassirostris and Forpus passerinus deliciosus in Amazonas).

 

Effect on AOS-SACC area:

 

This proposal would elevate up to two subspecies endemic to South America to species status.

 

Voting:

 

A YES vote on (A) would be to split crassirostris from F. passerinus.

     IF (A) passes, a YES vote on (B1) would be to adopt the English name Large-billed Parrotlet for F. crassirostris.

IF (A) passes, a YES vote on (B2) would be to retain the English name Blue-winged Parrotlet for the more widely distributed F. xanthopterygius s.s.

 

A YES vote on (C) would be to split spengeli from F. xanthopterygius.

If (A) passes, a YES vote on (D1) would be to adopt the English name Turquoise-winged Parrotlet for F. spengeli.

If (A) passes, a YES vote on (D2) would be to retain the English name Green-rumped Parrotlet for the much more widely distributed F. passerinus s.s.

 

Literature Cited:

 

Bocalini, F., and L. F. Silveira (2015). Morphological variability and taxonomy of the Blue-winged Parrotlet Forpus xanthopterygius (Psittacidae). Revista Brasileira de Ornitologia 23(1):64–75.

Clements, J. F., T. S. Schulenberg, M. J. Iliff, S. M. Billerman, T. A. Fredericks, B. L. Sullivan, and C. L. Wood (2019). The eBird/Clements Checklist of Birds of the World: v2019. Downloaded from https://www.birds.cornell.edu/clementschecklist/download/ 

Collar, N. (1997). Family Psittacidae (Parrots). In del Hoyo, J., A. Elliot, and J. Sargatal (Editors). Handbook of the Birds of the World. Volume 4. Sandgrouse to Cuckoos. Lynx Edicions, Barcelona.

Cory, C. B. (1918). Catalogue of Birds of the Americas. Part 2 No. 1. Field Museum of Natural History Zoological Series 13(197).

del Hoyo, J., and N. J. Collar (2014). HBW and BirdLife International Illustrated Checklist of the Birds of the World. Volume 1: Non-passerines. Lynx Edicions, Barcelona.

Dickinson, E. C. (Editor) (2003). The Howard and Moore Complete Checklist of the Birds of the World. Revised and enlarged 3rd edition. Christopher Helm, London.

Dickinson, E. C., and J. V. Remsen, Jr. (Editors) (2013). The Howard and Moore Complete Checklist of the Birds of the World. 4th edition. Volume One. Non-passerines. Aves Press Ltd., Eastbourne, UK.

Donegan, T., J. C. Verhelst, T. Ellery, O. Cortés-Herrera, and P. Salaman (2016). Revision of the status of bird species occurring or reported in Colombia 2016 and assessment of BirdLife International’s new parrot taxonomy. Conservación Colombiana No. 24:12-36.

Forshaw, W. T. (1973). Parrots of the World. Doubleday, Garden City, New York.

Gill, F., and D. Donsker (Editors) (2015). IOC World Bird List (v 5.3). http://www.worldbirdnames.org/

Gyldenstolpe, N. (1945). The bird fauna of Rio Juruá in Western Brazil. Kongliga Svenska Vetenskaps-Akademeins Handlingar 22(3):1-338.

Hellmayr, C. E. (1907). Another contribution to the ornithology of the lower Amazons. Novitates Zoologicae 14:1-39.

Juniper, T., and M. Parr (1998). Parrots. A Guide to the Parrots of the World. Pica Press, Sussex, UK.

Peters, J. L. (1937). Check-list of the Birds of the World. Volume 3. Harvard University Press, Cambridge, Massachusetts.

Remsen, J. V., Jr., J. I. Areta, C. D. Cadena, S. Claramunt, A. Jaramillo, J. F. Pacheco, J. Perez Emán, M. B. Robbins, F. G. Stiles D. F. Stotz, and K. J. Zimmer (Version 11 February 2020). A Classification of the Bird Species of South America. American Ornithological Society. http://www.museum.lsu.edu/~Remsen/SACCBaseline.htm

Ridgway, R. (1916). The birds of North and Middle America. Bulletin of the United States National Museum No. 50 Part 8.

Sibley, C. G., and B. L. Monroe, Jr. (1993). A World Checklist of Birds. Yale University Press, New Haven, Connecticut.

Smith, B. T., C. C. Ribas, B. M. Whitney, B. E. Hernández-Baños, and J. Klicka (2013). Identifying biases at different spatial and temporal scales of diversification: a case study in the Neotropical parrotlet genus Forpus. Molecular Ecology 22:483–494.

Whitney, B. M., and J. E. Pacheco (1999). The valid name for Blue-winged Parrotlet and designation of the lectotype of Psittaculus xanthopterygius Spix, 1824. Bulletin of the British Ornithologists’ Club 119:211-214.

 

 

Pamela C. Rasmussen, August 2020

 

 

 

 

 

Comments from Areta: “A) YES. To be consistent with recognition of other species in Forpus.  The relatively deep divergence of crassirostris and the lack of monophyly this implies for xanthopterygius makes this the only reasonable alternative at hand without a major overhaul of taxonomy in Forpus.

 

“B) NO. Smith et al (2012) found that spengeli is a taxon more closely related to (and embedded within) passerinus as currently delineated. Given the minor genetic differentiation, the not really impressive plumage differences, and the difficulty in understanding what the plumage differences might mean in a genus characterized by complicated plumage variation that does not clearly correspond with phylogenetic relationships, I prefer to leave spengeli within passerinus.”

 

Comments from Stiles: “YES to A: recognizing crassirostris as a species. YES to B1 and B2 (E-names). C: YES to this split as well, and YES to B1 and B2 (E-names). The genetics, morphology and distributions fit well.”

 

Comments from Robbins: “A) YES, to recognizing crassirostris as a species based on the Smith et al. genetic data. C) NO, to spengeli as a species, as I agree with comments by Nacho.”

 

Comments from Zimmer:

“(A) YES, using the “yardstick” of differentiation seen in other recognized species of Forpus.

“(B1) YES for adopting the English name of Large-billed Parrotlet for F. crassirostris.

“(B2) YES for retaining the familiar English name of Blue-winged Parrotlet for the more widely distributed F. xanthopterygius sensu stricto.

“(C)  I’m a little confused here, because I thought the molecular data showed spengeli to be embedded within passerinus, which is where I thought we currently treat it.  Yet, in the Voting instructions at the end of the Proposal, it is stated that a “YES vote on (C) would be to split spengeli from F. xanthopterygius” (bold-face mine).  I can only assume this is an error, since the title of the Proposal reads “(B) Treat Forpus spengeli as a separate species from F. passerinus.”  Going with that assumption, then we would appear to have a conflict between the molecular data and what are, to me, fairly clear plumage distinctions that are at least on a par with the distinctions between other recognized species of Forpus.  Given that conflict, I am tentatively persuaded by the described parapatry between spengeli and cyanophanes without evidence of intergradation.  So, a tentative “YES” for splitting spengeli from passerinus, and treating it as a distinct species.

“(D1) YES to establishing Turquoise-winged Parrotlet as the English name for a split spengeli.

“(D2) YES” for retaining Green-rumped Parrotlet for the widespread F. passerinus.”

 

Comments from Pacheco: “As Kevin has already warned, the proposal's title and its recommendations are not perfectly aligned. Having said that, my votes are:

 

“A – NO, tentatively. Although there is evidence by mtDNA that crassirostris is clearly distinct, I am concerned about the alleged lack of diagnosability pointed out by Bocalini and Silveira (2015).

 

“C – YES, tentatively. Clarifying that this “yes” is for splitting spengeli from passerinus, and treating it as a separate species.”

 

Comments from Claramunt:

 

“A. YES. Given the conservative plumage of this genus, I think that the mtDNA evidence is compelling. 

 

“C. YES. I think there is a chance that the lack of reciprocally monophyly between spengeli and passerinus is artifactual. Further analyses are needed, but I give spengeli the benefit of the doubt.”

 

Comments from Lane:

“A) YES. I should note that Vitor Piacentini sent me a private email that made me hesitate by suggesting that two forms of Forpus might be involved within the Amazon of northern Peru, based on orbital skin color and iris color in the photos available in Macaulay Library. I am not sure if these character states are necessarily taxon-driven but rather may be age-driven (or perhaps even just effects of lighting, etc.). If the characters are taxon driven--meaning there could be two "xanthopterygius-types" in northern Peru, then we'd have to be very careful we apply the name "crassirostris" correctly! But after reviewing photos on Macaulay again, I am thinking that this may be entirely either age-driven or lighting effects, and not taxonomic characters at all (in addition, I would expect there to be vocal characters that would offset with two potentially sympatric Forpus, but I am unaware of any beyond those between F. xanthopterygius/crassirostris and F. modestus). So I will say YES, but wonder if there may be a more complicated issue in western Amazonia?

“B1) YES to accepting Large-billed Parrotlet as English name for F. crassirostris.

“B2) NO to retaining "Blue-winged Parrotlet" for restricted F. xanthopterygius. Even though Smith and all showed that crassirostris is not sister to xanthopterygius, it occupies a huge portion of the range (I think? This is not really clear to me. I guess it is the western Amazonian form, if not more widespread?) of that species sensu lato. So I think it would be confusing to retain that name for the sensu stricto version. Thus, a new name would be necessary by my estimation. Hellmayr isn't very helpful in providing a reasonable name for the form sensu stricto, so a novel name may be the best move. I would propose something like "Cerrado Parrotlet", but that's just a first attempt, and without fully understanding where the break between the two species occurs, and their preferred habitats.

“C) YES

“D1) YES to accepting Turquoise-winged Parrotlet as English name for F. spengeli.”

“D2) YES to retaining Green-rumped Parrotlet for restricted F. passerinus.”

 

Comments from Remsen:

“A) YES, reluctantly.  I suppose the genetic data are solid (N=4 crassirostris samples; nDNA shows same basic pattern as mtDNA), but those same data found spengeli embedded in passerinus – so something is fishy.  But I share Fernando’s concern.  Contrary to statements in the proposal, Bocalini and Silveira (2015) did measure culmen length and bill width, and mean differences are about 0.4 and 0.1 mm respectively; even so, culmen width was the main influence on Factor 2, yet crassirostris does not occupy discrete morphospace.

“B1) YES, reluctantly.  Large-billed has some history and syncs with the scientific name, but the Bocalini-Silveira analysis indicates that this is a trend, not a real character.  Does this bird stand out in the field as having a larger bill than other Forpus in the group?  Could applying this name be misleading?”

“B2) NO!  For the same reasons as outlined by Dan above.  Crassirostris is not some peripheral isolate but rather occurs in 4 countries in western Amazonia.  For those of us who have worked there, this is the taxon we called Blue-winged Parrotlet.

“C) YES. Parapatry with no sign of gene flow is sufficient evidence for species rank for any taxa, and as quantified by the Bocalini-Silveira analysis, this taxon is fairly distinctive by Forpus standards.

“D1) YES - Turquoise-winged Parrotlet already in use and a good name.

“D2) YES.  In contrast to the crassirostris situation, spengeli is a peripheral taxon with a vastly smaller range.

 

Comments from Schulenberg on B1 and B2: “As far as Forpus crassirostris and Forpus xanthopterygius are concerned, I vote NO on both 'Large-billed' (crassirostris) and 'Blue-winged' (retaining this name for xanthopterygius). 'Large-billed' just isn't a great name to begin with, given that the difference in bill size between nominate crassirostris and the other taxa is not large and or consistent a difference (as noted by Van: "crassirostris does not occupy discrete morphospace"). I wonder if anyone would accept something like 'Riparian Parrotlet' (similar to the case of Riparian Antbird Cercomacroides fuscicauda), in a nod to the fact that it occupies open, river edge habitats (and now, of course, a lot of second growth etc.). this habitat preference isn't unique in the genus, but ... we're going to be limited by color-based names, and 'Amazonian Parrotlet' already is taken (Nannopsittaca dachilleae). or can anyone come up with anything better? otherwise, crassirostris occupies a large enough geographic range that retaining 'Blue-winged' doesn't seem wise to me. I don't think 'Cerrado Parrotlet' would work well, since I take it that the range of Forpus xanthopterygius extends to west to Beni, and for that matter well outside of the cerrado in eastern Brazil. would 'Blue-rumped Parrotlet' work? I think that's been used before for Mexican Parrotlet Forpus cyanopygius, but perhaps sufficiently long ago that it wouldn't be a problem. or any other ideas?

 

“On the other hand, I'm perfectly fine with 'Turquoise-winged' for spengeli, and with retaining 'Green-rumped' for passerinus.”

 

Comments from Jaramillo on B1 and B2: I read Tom's comments, and like that Riparian name. I will change my votes on the two.

F. crassirostris - NO to Large-billed. Yes to Riparian if that is put forward. 
F. xanthopterygius – NO, as it seems like we are going down the need to get a new name put forward, I will help it get there more quickly then. 

 

Comments from Donsker on on B1 and B2:  “I recommend using the English names "Large-billed" Parrotlet for F. crassirostris and “Cobalt-rumped Parrotlet” for F. xanthopterygius.

 

“Although the name Large-billed Parrotlet may not be at all helpful for diagnosis, the name is at least a reasonable translation of the Latin species epithet, which may be as useful a reason as any other to use it. (Thick-billed Parrotlet would be a better translation, but I feat that’s too close to Thick-billed Parrot and is probably best avoided). Besides the English name Large-billed Parrotlet has an established history of usage at least traceable back to Brabourne & Chubb and to Cory.

 

“I’d favor changing the English name of F. xanthopterygius to avoid the potential for confusion that Dan and Van have both expressed. “Blue-winged” Parrotlet is the well-established English name for both the western Amazonian crassirostris and the eastern Brazilian/Bolivian xanthopterygius.  This becomes even more confusing given the unsettled history of the scientific name applied to Blue-winged Parrotlet (sensu lato), which has bounced around between F. crassirostris and F. xanthopterygius over the mid to late 20th century.  But there is no other suitable historical English name for this form that I am aware of.  It’s a dilemma.  I like Tom’s suggestion of Blue-rumped Parrotlet, but I don’t think it would be wise to use it given the association of that name with Mexican Parrotlet.  But perhaps a similarly constructed name? So, I’d propose Cobalt-rumped Parrotlet.  "Cobalt" is a reasonable modifier for the shade of the dark blue rump of this species. This has been variously described as "rich blue" (Ridgely et al. 2016); ”violet-blue" (Forshaw & Knight 2010); "cobalt-blue" (Juniper & Parr 1998); and "azul-colbalto" (F. x. flavissimus) or "azul-violeta" (nominate xanthoptergius) (Grantsau 2010).”

 

Additional comments from Remsen: “I’m switching from Y to N on crassirostris just to force a reconsideration of these names.”

 

Comments from Bonaccorso:

“A. YES. The branch that conducts to F. x. crassirostris is among the deepest in Forpus and is supported by both mitochondrial and nuclear data.

“B. NO. Forpus passerinus spengeli is well within F. passerinus to be considered another species. Until data on other characters suggest that they are reproductively isolated from other F. passerinus, they should remain as part of F. passerinus.”

 

Comments from Stiles: “This proposal (Forpus parrotlets) suffered from a confusion between the proposal (A: split crassirostris from xanthopterygius, and C: split spengeli from passerinus), and the way the voting chart was given (A: split crassirostris from passerinus, and C: split spengeli from xanthopterygius).  Fortunately, the E-name proposals of the voting chart under A and C got it right, and I am assuming that people followed the proposal regarding the splits, although several simply said YES or NO without specifying. Here are the splits: A passed 9:1. C passed 7:3. Because some people did not vote on E-names, nothing reached quorum but both Green-rumped for passerinus and Turquoise-winged for spengeli were 5:0 for YES. Large-billed was 3YES: 2NO for crassirostris, with two YES for Riparian; for xanthopterygius, YES votes were 2 for Blue-winged, 1 for Cerrado(?), 1 for Cobalt-winged (a late-comer, but I’m willing to switch to it as well (so give it 2). This stalemate prompted taking up the issue again in Proposal 900 (see below).”

 

Comments from David Donsker: B and D: “NO. to all I’d support "Riparian" Parrotlet” for F. crassirostris and “Cobalt-rumped Parrotlet” for F. xanthopterygius.”