Proposal (948) to South American Classification Committee

Treat Flame-rumped Tanager (Ramphocelus flammigerus) as two species

Background:

The Flame-rumped Tanager (Ramphocelus flammigerus) occurs from Panama south through Colombia to southwestern Ecuador and adjacent northwestern Peru. It consists of two subspecies: R. f. flammigerus in the northern part of the range and R. f. icteronotus in the southern part of the range. They meet and hybridize above 800 meters in the western Andes of Colombia. Both males and females of each subspecies have very different plumage. Males of both subspecies are mostly black, but R. f. flammigerus has a bright red back and rump, whereas R. f. icteronotus has a yellow back and rump. Females of R. f. flammigerus also have a reddish orange band across their chest and reddish-orange rump, undertail coverts, and uppertail coverts, whereas females of R. f. icteronotus are solid yellow below and on their rumps

Taxonomic lists and regional field guides differ in how they treat these two taxa:

The two forms are treated as subspecies of the same species by Storer (1970), Ridgley and Tudor (1989), Sibley and Monroe (1990), Dickinson (2003), Hilty (2021), and Clements et al. (2021). They are treated as separate species by Meyer de Schauensee (1970), Hilty and Brown (1986), Ridgely & Greenfield (2001), Restall et al. (2007), del Hoyo and Collar (2016), and Gill et al. (2022).

The HBW-Birdlife list (del Hoyo and Collar 2016) provides the following rationale for treating them separate: Usually treated as conspecific with R. flammigerus, and genetic differences apparently minimal; moreover, owing to recent deforestation the two taxa now reportedly meet and interbreed in W Andes of Colombia (along a narrow but stable band at middle elevations on upper Pacific slope); even so, visual divergence striking.

New information:

There is no new information per se; this proposal is to provide feedback to IOC’s WGAC, which is working to reconcile all world lists. However, there is relatively new research in the hybrid zone that doesn't appear to be considered previously in decisions to classify these taxa as one species or two.

A hybrid zone between the two taxa in western Colombia has been known for many years (Chapman 1917). The zone was studied in some detail by Sibley in the 1950’s (Sibley 1958). More recently, Morales-Rozo et al. (2017) studied the genetics, plumage color, and morphology of birds across the hybrid zone. The hybrid zone occurs in the Cauca River Valley where the lower elevation R. f. icteronotus meets the higher elevation R. f. flammigerus around 800 meters. The hybrid zone occurs along a 140 km transect and the birds in this area show a gradient from bright yellow to bright red (see Fig. 1, copied from the paper and inserted below). Morales-Rozo et al. (2017) also mentioned that the two forms meet and hybridize in additional contact zones further north in the Cordillera Occidental; however, these areas where not studied at the time of publication. Morales-Rozo et al. (2017) looked at historical specimens and collected fresh specimens in 2007-2010. They sequenced cyt b from recent samples as well as from toe pads of the specimens collected by Sibley. Their genetic analyses included samples within the hybrid zone as well as samples far from the hybrid zone in Ecuador and Panama. They also included 6 morphological characters and measured rump plumage coloration using a spectrophotometer. These phenotypic characters were defined into three time periods to study the temporal dynamics of the cline (prior to 1911, 1956-1986, and 2007-2010).

Morales-Rozo et al. (2017) found overall low levels of sequence divergence. Within Colombia, samples differed on average by only 0.3%, and samples between Colombia and Panama differed by only 0.4%, but between Colombia and Ecuador, samples differed by 1.6%. Samples from Ecuador and Colombia could be separated in their tree, but otherwise no clades were associated with specific geographic regions or plumage colors. In many cases, individuals with different rump colors and from different geographic regions had the exact same sequence. In addition, no genetic structure was detected across the transect. As the authors stated: “In contrast to multiple studies on hybridization in birds finding significant mtDNA divergence between populations located away from the center of hybrid zones and clinal variation in haplotype frequencies across them, mtDNA variation was not geographically structured in our study system, a likely consequence of recent divergence of the hybridizing populations or of high levels of introgression.” In addition, the authors have niche modeling and demographic data indicating that the two taxa have expanded their range and come into contact after prior isolation. Although the hybrid zone is often thought to be the result of recent anthropogenic activity (deforestation and conversion to crops creating scrub and second growth), the authors’ analyses show it to be much older than expected – around 6,000 years before present. However, anthropogenic activity could still have increased the degree of hybridization.

In contrast to the lack of pattern with the genetic data, the authors did find clines for the morphological data and for the plumage color data. For each period of time, the clines for these two character sets where coincident, and the clines appear to have moved slightly to the east and upwards in elevation. In addition, the cline is much narrower than expected under a model of neutral diffusion. Thus, the authors propose that the hybrid zone is a tension zone, where dispersal of parental forms and selection against hybrids balance each other out.

There is actually a talk at next week’s AOS meeting on the same hybrid zone (authors = Castaño, Cadena, and Uy), this time using genome-wide SNP data. According to the abstract, the findings look similar to the mtDNA study: “We found low genetic divergence and genetic structure across the hybrid zone, and a discordance in the width and cline center between the genome-wide loci and the plumage clines previously reported. Our results suggest that there are few intermediate individuals (F1 hybrids) and pure and backcrossed individuals of the icteronotus subspecies appear to be distributed across allopatric and sympatric populations.” I will go to the talk and hopefully chat with the authors, but it looks pretty consistent with the degree of gene flow found with the mtDNA data. The abstract also mentions plans to study the other areas of contact mentioned in the Morales-Rozo et al. (2017) study.

Recommendation:

These two taxa would clearly be recognized as separate species under many other species concepts, like the phylogenetic species concept, that recognizes past evolution of characters as defining evolutionary units. In this case, the plumage differences in both males and females are pretty dramatic and indicate evolution of characters in allopatry. However, this committee follows the biological species concept, which downplays the importance of these events in the face of gene flow, or in the case of allopatric taxa, potentially significant gene flow. The published genetic study based on mtDNA does indicate movement of genes across the hybrid zone and the unpublished nuclear data seems to find a similar pattern. The authors characterize the hybrid zone as a tension zone; thus, there is some selection against hybrids. However, there appears to be enough gene flow for this committee to consider them one biological species. Thus, following this concept, I recommend a NO vote on this proposal to split these two taxa. The situation is analogous to the relatively recent lumping by this committee of R. passerinii and R. costaricensis. These two taxa also differ in the plumage in a similar way as the taxa under consideration in this proposal.

English names:

If we were to vote to split, the English names that are in common use are Lemon-rumped for R. icteronotus and Flame-rumped for R. flammigerus sensu stricto.

Literature cited:

Clements, J. F., T. S. Schulenberg, M. J. Iliff, S. M. Billerman, T. A. Fredericks, J. A. Gerbracht, D. Lepage, B. L. Sullivan, and C. L. Wood. 2021. The eBird/Clements checklist of Birds of the World: v2021.

Chapman, F. M. 1917. The distribution of bird-life in Colombia; a contribution to a biological survey of South America. Bull. Am. Mus. Nat. Hist. 36:1–729.

Del Hoyo, J., and N. J. Collar. 2016. HBW and BirdLife International Illustrated Checklist of the Birds of the World. Volume 2. Lynx Edicions, Barcelona, Spain.

Dickinson, E. C. 2003. The Howard and Moore Complete Checklist of the Birds of the World. Princeton University Press, Princeton.

Gill F, D. Donsker, and P. Rasmussen (eds). 2022. IOC World Bird List (v12.1). doi: 10.14344/IOC.ML.12.1.

Hilty, S. L., and W. L. Brown. 1986. A Guide to the Birds of Colombia. Princeton University Press, Princeton, New Jersey.

Hilty, S. L. 2021. Birds of Colombia. Lynx Edicions, Barcelona.

Meyer de Schauensee, R. 1970. A Guide to the Birds of South America. Livingston Publishing Company, Wynnewood, Pennsylvania.

Morales-Rozo, A., E. A. Tenorio, M. D. Carling, and C. D. Cadena (2017). Origin and cross-century dynamics of an avian hybrid zone. BMC Evolutionary Biology 17:257.

Restall, R. L., C. Rodner, and M. R. Lentino. 2007. Birds of Northern South America: An Identification Guide. Yale University Press, New Haven, Connecticut.

Ridgely, R. S. and Greenfield, P. J. 2001. The birds of Ecuador: Status, Distribution, and Taxonomy. Cornell University Press, Ithaca, New York.

Sibley, C.G., 1958. Hybridization in some Colombian tanagers, avian genus "Ramphocelus". Proceedings of the American Philosophical Society 102:448-453.

Sibley, C. G., and B. L. Monroe, Jr. 1990. Distribution and Taxonomy of the Birds of the World.  Yale University Press, New Haven, Connecticut. 

Kevin J. Burns, June 2022

_______________________________________________________________________________

Comments from Stiles: NO. Here, I note that the study by Andrea Morales and Daniel Cadena dealt exclusively with the same transect as used by Sibley to first document the hybridization between these two forms. However, this is not the only such case of hybridization over the cordillera Occidental from the Cauca valley into the Pacific lowlands. I obtained a similar picture of hybridization in a brief survey transect to the north in Risaralda in 1991 from La Virginia at ca. 800 m on the río Cauca (all flammigerus) up via Apía (at ca. 1500 m, where a mixture of all forms, mostly intermediates, was seen) and on to Pueblo Rico and down to Santa Cecilia at ca. 300 m (here, nearly all pure icteronotus): in all, ca. 70 birds counted over this route. Moreover, there is apparently a third and more extensive area of hybridization exists further north, from Puerto Triunfo at ca. 300 m on the río Magdalena (all birds seen were icteronotus) up to Medellín (1800m, where observers there inform me that most birds are intermediates). From Medellín down southwest to the río Cauca, the country is densely settled, and various roads reach the river over at least 50 km along the river; at least at Bolombolo, where I stayed for several days in 1994, all birds seen were flammigerus. Thus, the hybridization of these two forms is far more widespread than might have been noted previously, and strongly favors continuing to maintain them as conspecific.”

Comments from Remsen: “NO. That the contact zone consists of a hybrid swarms with mostly intermediate birds indicates absence of assortative mating – these two populations treat each other as “same” when it comes to mat choice, so, in my opinion, why shouldn’t we? The cline in variation between the two is best seen in the rump color data in Morales-Rozo et al.’s Fig. 6D (based on Sibley’s 1956 samples). The hybrid swarm that comprised the contact zone is sufficient reason to treat them as conspecific.”

Comments from Claramunt: “NO. The situation could be explained by a color polymorphism, geographically structured, but within a single species. The lack of genetic differentiation and the multiple instances of hybridization and introgression suggest a single species.”

Comments from Robbins: “NO, for reasons presented in the proposal, plus those comments by Gary Stiles stating that the zone of intergradation is broader (“far more extensive”) than what is presented in the literature.”

Comments from Areta: “NO. Extensive hybridization from several localities across a broad front and across a considerable distance, coupled to little genetic differentiation (in mtDNA) away of the area of the hybridization, plus minor plumage differences (yellow vs. red carotenoid-based plumages), no association of geographic origin to mtDNA clades, and long-term persistence of this situation indicates that these two are behaving as a single unit in places in which they meet. It might also be the case that the "orange" birds are equally palatable to "yellow" and "red" birds, and thus even in the face of selective disadvantage, there is gene flow between the extreme populations. If reinforcement will happen or not we do not know, but so far, I don´t see compelling evidence to recognize them as separate species. The widespread hybridization across geography and the wide hybrid cline that was studied in detail indicates that flammigerus and icteronotus are not behaving as separate species. I would contend that given the rampant hybridization, no species concept would have an easy time supporting their recognition as different species.”

Comments from Lane: “NO. The presence of multiple areas where there appears to be extensive interbreeding between yellow and scarlet rumped populations, evidence supporting their separation is overwhelmed by that supporting continuing to consider them a single biological species.”

Comments from Bonaccorso: “NO. Even if the hybrid zone is a “tension zone” there is enough geneflow for genes to intergrade from one form to the other and maintain a perfect cline. I agree with Van´s comment: “these two populations treat each other as “same” when it comes to mate choice, so, in my opinion, why shouldn’t we?”

Comments from Zimmer: “NO, for all of the reasons elucidated in the proposal. As different as the two taxa are in plumage at their extremes, the existence of a rather broad hybrid swarm in the contact zone would seem to be sufficient evidence that the two taxa are not discriminating between one another to the extent one would expect if, indeed, they represented distinct biological species.”

Comments from Pacheco: “NO. Because there is no genetic differentiation and the hybridization area is more extensive (Gary) than previously admitted.”