Proposal (1015) to South American Classification Committee

Treat Rauenia darwinii as a separate species from Rauenia bonariensis

Note: This is a high-priority issue for WGAC.

Background: Our SACC note on this is as follows:

19bb. The Andean subspecies darwinii was formerly (e.g., Chapman 1926, Zimmer 1930) considered a separate species from Pipraeidea (ex-Thraupis) bonariensis, but Hellmayr (1936) treated them as conspecific as in all subsequent classifications, including Zimmer (1944). Del Hoyo & Collar (2016) treated the northern subspecies darwinii as a separate species (“Green-mantled Tanager”) based in part on differences in song described by Boesman (2016k).

Dickinson & Christidis (2014) treated these as single polytypic species, with 4 subspecies placed in two groups:

(1) darwinii from Andes of Ecuador through Peru to n. Chile and n. Bolivia (La Paz) + composita from c. Bolivia in Cochabamba and Santa Cruz

(2) schulzei from se. Bolivia, w. Paraguay, and nw. Argentina + nominate bonariensis of ne. Argentina, Uruguay, and se. Brazil

Their construction of the group composition is a mistake. Darwinii stands alone in having an olive-green back and being completely yellow below, whereas the other three have black backs and orange underparts.

Zimmer (1944) reversed his earlier species limits in treating them all as conspecific. He did not give explicit reasons, but in describing composita and discussing schulzei he noted individual variation in the direction of darwinii:

“Several specimens of bonariensis males have the black mantle feathers margined, in varying degree, with olive. Many specimens of bonariensis, schulzei, and composita have a band of olive separating the black of the mantle from the orange of the rump, but many lack it. Some females (if the specimens are correctly sexed) have as much blue on the head as certain females of darwinii, although other examples of the Peruvian form reach an extreme not covered by the other three subspecies. All these characters, however, are indicative of the close relationship of all forms.”

Isler & Isler (1987) did not address species limits but said: “In Bolivia, black-backed males sometimes have back feathers edged olive or have an olive band between back and rump (Eisenstraut 1935)”

Here is the plate by Hilary Burns in Hilty (2011):

Here are photos of LSUMNS specimens; unfortunately, we don’t have any schulzei, for which I have left a gap in each of the photos as a reminder that we’re not seeing that taxon. (We actually have surprisingly few specimens, period – I think this because we usually see them in people’s gardens and chacras.) When Zimmer described composita (which my auto-correct continues to insist must be “composted”), he distinguished it from schulzei as follows: “Compared to T. b. schulzei of [Paraguay and] northwestern Argentina, composita is larger, the blue of the head averages darker, and the orange colors of the lower under parts and rump are a little less intense.”

Here is a photo of darwinii from Paul Molina A from Azuay from Macaulay Library:

Here is a photo of composita from Cochabamba by Phillip Edwards from Macaulay Library. By the way there is a photo from Cuzco, Peru, of a bird indistinguishable from this one as far as I can tell; it had been identified as darwinii , and I flagged it for review, so Dan or Tom should have this in their review queue already. I mention this because it indicates that these birds have good dispersal abilities, which is easy to predict from their adaptability and association with human-altered habitat.

New information:

Del Hoyo & Collar (2014) treated darwinii and bonariensis as separate species based on the Tobias et al. point scheme as follows (provided by Pam Rasmussen):

"Usually considered conspecific with P. bonariensis, but differs in its (in male) olive-green vs black mantle, back, scapulars and chest side (3); yellow vs flame-orange breast and rump (3); (in female) bluer crown, wings and tail (2); more olive-yellow underparts and rump (ns[2]); and song with lower-pitched notes in narrow frequency range (2) (1). Monotypic."

The information on song comes from Boesman (2016k), whose examination of recordings yielded the following conclusions:

“Song is quite variable, typically a sub-phrase of 2-3 notes repeated several times, sometimes with slight changes. Often alternating downslurred and (lower-pitched) upslurred notes. Pace also quite variable. Some song (?) phrases lack the typical regular pattern. Overall, song is quite similar over entire range, but: darwinii can be recognized from all other races by the presence of lower-pitched notes, which reach a max. frequency of c 3.5-6kHz (vs c 6-8kHz in southern races, even more in bonariensis) and cover a much smaller frequency range (c 2-3kHz vs 3-6kHz) (score 1+1=2).

“bonariensis apparently lacks a clear pattern of alternating downslurred and upslurred notes, notes are higher-pitched and less melodious.

“We can thus conclude that race darwinii can be identified by alternating high-pitched and low-pitched notes, races composita/schulzei has alternating upslurred and downslurred notes, while race bonariensis has high-pitched repeated notes but lacks upslurred notes.”

Boesman (2016k) presented sonograms of 11 individuals, which I encourage you to look at to see the variability in song noted by Boesman. Better yet, go right to xeno-canto and do some sampling or Macaulay (where you will find recordings by Ted Parker, Mark Robbins, and regular contributors Niels Krabbe and Gary Rosenberg). Boesman had a lot of patience to deal with all these predominantly weak recordings, often with lots of background noise. This is clearly a difficult species to record the song well (as are a lot of thraupids). Nonetheless, I’m pretty sure I can hear the differences he described between the three groups.

Discussion and Recommendation: This is another tough case. It lacks a definitive, quantitative study, so at this point we have to base a decision on fragmentary evidence. I can understand voting no on this one until the missing pieces are in place, especially study of the contact zone, which may never be thoroughly characterized if we endorse a split. However, I think I will vote YES on this for two reasons.

First, darwinii and composita are basically parapatric without little sign of hybridization. A number of taxa are separated by the dry canyon of the La Paz river system that isolates taxa of humid forest on either side. On the surface, this could be one of those, but this is not a bird restricted to humid cloud forest but instead is an edge species that does very well in towns and farmland and in fairly dry areas. Note that darwinii is found on both sides of the North Peruvian Low/Marañon, the biggest barrier to dispersal for humid forest birds in the Andes, with no signs of phenotypic differentiation on either side. Ditto the Urubamba and other dry valleys. Composita occurs in coastal Peru and n. Chile. As noted by Isler & Isler (1987), southern populations are migratory. These birds clearly have strong dispersal abilities. At a small scale, the area where they likely come in contact in Bolivia is difficult to access and is poorly studied bird-wise. So, you never know but … all evidence so far suggest there is some sort of hard barrier to gene flow that is not strictly geographic, given these birds’ habitat tolerance. Zimmer’s interpretation of the occasional olive feathers in the backs of bonariensis is not necessarily due to any ongoing gene flow, but could reflect past connectivity or shared ancestry. The Eisenstraut observation of olive dorsal feathers in black-backed Bolivia birds (Isler & Isler 1987) is more suggestive of gene flow and bears further investigation.

Second, we have some evidence of differences in song, and I can hear the subtle differences discovered by Boesman. Would we rather have a larger, formal, peer-reviewed study with dense geographic sampling? Of course. But I think we have enough information to suggest that the initial findings would hold up, although it looks to me that composita-schulzei might be as different from bonariensis s.s. as darwinii is in terms of song. Strictly on the vocal information, one might argue for a 3-way split.

Finally, a minor, subjective point on color differences. I am not particularly impressed with the ventral color differences, which also appear to vary in intensity within the bonariensis group, although that could be due to age. The Macaulay photo galleries show lots of individual variation, although some of that might be digital artifacts. However, I am impressed with the difference in back color. It’s an abrupt shift from green to black. No sign of a cline. All in all, I think the available evidence shifts burden-of-proof on treating them all as conspecific, but I could be talked out of that view. I would really like to see an analysis of dorsal plumage variation in males in central Bolivia – that’s not too much to ask for.

English names: BirdLife International retained Blue-and-yellow Tanager for the much more widely distributed R. bonariensis (including composita and schulzei) and used Green-mantled Tanager for R. darwinii. Although retaining the parental name for the more widely distributed species is perhaps in accord with our SACC guidelines on English names, in this case, darwinii is found from Ecuador to n. Bolivia and is a familiar bird on bird tours to those countries. Also, part of the reason for the split was that the two differ in belly color, with darwinii actually being the species that is “the” Blue-and-YELLOW Tanager.” To avoid that confusion, if the proposal passes, I will do a proposal to make bonariensis s.s. the “Blue-and-ORANGE Tanager” to highlight its most conspicuous plumage feature and to avoid causing confusion for observers and to call attention to the change in species limits.

References: (see SACC Bibliography for standard references)

BOESMAN, P. 2016k. Notes on the vocalizations of Blue-and-yellow Tanager (Thraupis bonariensis). HBW Alive Ornithological Note 402. In: Handbook of the Birds of the World Alive. Lynx Edicions, Barcelona.

Van Remsen, June 2024

Comments from Robbins: “NO. Plumage morphology: I looked at our (Univ. of Kansas) material and one question that I have that might explain some of the variation (ventral color?) within each form is that this tanager may have delayed maturation; specimens with completely ossified skulls and no bursa are clearly not in definitive plumage. Perhaps this is in already in the literature (?). Clearly the dorsal coloration does separate darwinii from the other forms. With regard to ventral concolor, at least all of our male specimens in definitive plumage from dept. Cochabamba have bright orange underparts, whereas a limited sample of darwinii, which are clearly in definitive plumage, have yellow underparts.

“Vocalizations: I listened to what I consider primary song (non-call notes) on both Macaulay and Xeno-canto.  It is clear that there is considerable variation and there may be two different primary vocalizations in both groups.  If one listens and compares relatively decent audio recordings on Macaulay from just Ecuador you realize there is a fair amount of variation within darwinii. Ditto for black-backed forms, e.g., compare recordings from Cochabamba and northern Argentina on Macaulay. So, in sum, at a minimum, there needs to be a rigorous comparison of analogous primary vocalizations.

“Although I've on the fence on this one, I would like to see a more rigorous vocal analysis and a genetic data set to help guide what the best course of action is. Thus, for now, a tentative NO.”

Comments from Areta: “An unhappy NO. I am of the view that these two are different species, but that no one has tackled the situation properly and splitting just because of the different plumage aspect seems insufficient. There is geographic variation in male and female plumage within the bonariensis group that needs to be analysed and a genetic perspective could provide really strong evidence in favour of the split. I have sound-recorded Rauenia bonariensis extensively, and there is a great deal of variation across geography, but I think that the vocalizations are quite different from those of darwinii, even allowing for variation within each. However, the lack of formal bioacoustic analyses and the lack of thorough studies on plumage variation in Bolivia where the two groups could interact (I see photographs of pure black-backed birds very close to green-backed ones in the Coroico area; is this seasonal or is this true breeding parapatry/overlap?), leads me to vote NO to the split for the time being.”

Comments from Stiles: “NO, for now: the difference in plumage certainly suggests that darwinii may be a different species, but a more thorough vocal analysis as well as genetic data could tip the balance either way.”

Comments from Lane: “NO. It sounds like this is a ripe study for someone in Bolivia to tackle, but until we understand what happens at the contact zone, it is probably best to leave this complex as is.”

Comments from Claramunt: “YES. I agree with the arguments in the proposal. If these two forms were reproductively compatible, they would be hybridizing in N Bolivia. Unless this is an unusually sedentary tanager. And given that they differ in at least two traits (intensity of ventral yellow, and black versus green back), hybrids would be readily distinguishable by having unmatching combinations of these two traits at least, and maybe intermediate amounts of black on back. Therefore, the evidence points to reproductive isolation and independent evolution despite geographical proximity.”

Comments from Niels Krabbe (voting for Jaramillo): “NO. I have also noted that they take two years to attain full adult plumage (at least males of darwinii). The lack of clear intermediates in La Paz may suggest that two species are involved, but how many specimens were examined? Vocally, these oscines do differ notably, but still show some similarity, possibly enough for them to accept each other (compare with the much more striking geographical variation in Geospizopsis unicolor song). A genetic and morphological study would settle it, but needs to be done before as split is warranted.”