Proposal (993) to South American Classification Committee

Treat Thamnophilus ruficapillus as conspecific with T. torquatus

A recent proposal in this committee (SACC Proposal 968) suggested maintaining the traditional treatment of the Thamnophilus ruficapillus complex into two species taxa (T. ruficapillus and T. torquatus), as opposed to adopting the split of T. ruficapillus into two species proposed by Del Hoyo & Collar (2016), or any other possible additional splits. A possibility that has not been considered is to treat the T. ruficapillus complex as a single biological species (i.e., treating ruficapillus as conspecific with torquatus), which is proposed below.

The taxonomic history of the complex and other relevant pieces of information were summarized in Proposal 968. The argument here for treating the complex as a single biological species is twofold:

1. Divergence in vocalizations are currently assumed to be the key phenotypic proxy for biological species-level divergence in antbirds, and no vocal differences are known between ruficapillus and torquatus. The current two-species treatment of this species complex is based on early analyses of plumage differences (Cory and Hellmayr 1924, Peters 1951), and this treatment has remained until now just because no one has reassessed it in light of the realization that vocalizations, not plumage differences, are the main phenotypic proxy for species-level divergence in antbirds. In other words, the ornithological literature currently treats ruficapillus and torquatus as separate biological species only due to inertia.

2. Recent genetic work indicates that the current taxonomic treatment of the T. ruficapillus complex is incorrect. Phylogenetic work based on genome-wide genetic markers showed that T. ruficapillus and T. torquatus as currently defined are not reciprocally monophyletic (Harvey et al. 2020, Bolívar‐Leguizamón et al. 2024). Although paraphyly is not a problem in species delimited under the BSC, treating ruficapillus and torquatus as a single species eliminates paraphyly. Most importantly, a recently published analysis of population genetic structure of the complex found range-wide signatures of genomic introgression between ruficapillus and torquatus (Bolívar-Leguizamón et al. 2024). This finding is particularly remarkable when considering that they only had a few samples from the contact zone between the two taxa, thus suggesting that introgression is likely to be pervasive. In Proposal 968, Dan Lane provided several photographs of apparently phenotypically intermediate birds from the contact zone between ruficapillus and torquatus. Therefore, there is both genetic and phenotypic evidence indicating that ruficapillus and torquatus are weakly reproductively isolated, and so the current treatment of these two taxa as two biological species is incorrect.

Similar proposals (e.g., 173, 174, 977) have not passed because several people prefer to wait for more comprehensive analyses to avoid changing the taxonomy multiple times. I understand that point of view, but, personally, I prefer to see species limits as hypotheses formulated within the context of currently available evidence, and current evidence indicates that the classification of ruficapillus and torquatus as two separate biological species is incorrect. Even though this complex may potentially include multiple biological species, I think that recognizing a broadly defined species taxon for now, until a thorough analysis is published, is better than maintaining species taxa that are plainly improperly defined. In addition, I think it’s more likely that future taxonomic work will address the taxonomy of the complex as a whole if it’s treated as a single species for now. In fact, a single-species treatment will not only eliminate the possibility that future taxonomic revisions address only part of the complex (e.g., as in the BirdLife split), but also it may stimulate further taxonomic study of the complex by showing more clearly that this system is an interesting one whose taxonomy is poorly resolved.

References:

Bolívar‐Leguizamón, S. D., F. Bocalini, L. F. Silveira, and G. A. Bravo (2024). The role of biogeographical barriers on the historical dynamics of passerine birds with a circum‐Amazonian distribution. Ecology and Evolution 14:e10860.

Cory, C. B., and C. E. Hellmayr (1924). Catalogue of birds of the Americas and the adjacent islands in Field Museum of Natural History. Volume XIII, Part III. Field Museum Press, Chicago.

Harvey, M. G., et al. (2020). The evolution of a tropical biodiversity hotspot. Science 370:1343–1348.

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.

Peters, J. L. (1951). Check-list of the birds of the world. Museum of Comparative Zoology, Cambridge.

Rafael D. Lima, March 2024

Comments from Claramunt: “NO. I am concerned about the non-monophyly of the traditional T. ruficapillus, but I think the solution would be to split the Andean forms from the Atlantic forms, not to keep making the “complex” more complex by adding more differentiated lineages such as torquatus to the mix. The new paper by Bolívar-Leguizamón et al. (2023) is very insightful and we should revisit Proposal 968 instead. The new genetic data plus the photos compiled by Dan do suggest that there is some hybridization going on between ruficapillus and torquatus. But, as far as I can see, it could be a narrow hybrid zone with not much impact on the integrity of the two lineages. Without further genetic sampling of specimens from Minas Gerais and a formal analysis of plumage variation there, little we can say about the significance of those hybridization events.”

Comments from Lane: “NO. For the time being, I would say sticking with status quo until a deeper study of this complex is conducted is my preferred stance, so NO. I think we humans see the cap color as more of a character that would prevent hybridization than the birds probably do, but I'd like to see information on voice analysis and better-sampled genetic analyses before we change the taxonomy of the group. Until then, I don't see what harm maintaining the present taxonomy will do.”

Comments from Stiles: “NO: I agree with Dan that more evidence is needed to approve this lump, especially more thorough vocal and genetic data from the putative contact zone of these taxa to evaluate the degree and extent of hybridization between ruficapilla and torquatus.”

Comments from Bonaccorso: “NO. The proposal mentions that there are no differences in song, but in proposal 968, Areta & Pearman stated that there are some differences. Still, none of these statements are valid until a formal analysis (with sufficient sampling) is conducted. The hybrids found in Bolívar-Leguizamón et al. (2023) are from the contact zone, and the relatively small (but not insignificant) geographic sample does not indicate massive introgression from T. r. ruficapillus to T. torquatus or vice versa. These preliminary data seem to indicate that both entities are maintaining their evolutionary independence despite hybridization in the contact zone. It even seems that, with enough data, we should be looking into recognizing T. cochabambae, T. ruficapillus, and T. torquatus as three different species, instead of lumping them together.”

Comments from Areta: “NO. In our proposal we alluded to torquatus as an "undisputed" species with very similar vocalizations to those of ruficapillus, and indicated that a proper vocal study was needed in the complex (including ruficapillus sensu lato and torquatus). It was not our intention to lend support for species status of torquatus, but rather to highlight that current species in the group had very similar vocalizations, which complicated any rapid assessment of vocal differences in the group. The integration of phenotypic (plumage, vocal) and genetic data for the whole complex is still pending, and I think that going for a massive lump at this stage is inadequate given the many unknowns. The bottom line is the same that underlies our proposal: we need a fully comparative dataset to assess species limits in ruficapillus, in which different units seem recognizable, while it is not clear at what level these should be recognized.”

Comments from Robbins: “NO. I agree with some of the comments made in the proposal, but as Dan pointed out earlier and others have commented on the current proposal, more information is needed before making any species-level changes. Thus, I vote NO for now.”

Comments from Remsen: “YES. I am swayed by Rafael Lima’s point in the proposal:

“I prefer to see species limits as hypotheses formulated within the context of currently available evidence, and current evidence indicates that the classification of ruficapillus and torquatus as two separate biological species is incorrect. Even though this complex may potentially include multiple biological species, I think that recognizing a broadly defined species taxon for now, until a thorough analysis is published, is better than maintaining species taxa that are plainly improperly defined.”

“Of course, a comprehensive analysis would be preferable, but I think a lump here would help dramatize the need for one.”

Comments from Gustavo Bravo (voting for Glaucia Del-Rio): “NO. Based on current evidence, merging T. ruficapillus and T. torquatus would be an unwarranted move. I think that, if anything, the Andean forms of ruficapillus (subfasciatus, marcapatae, jaczewskii and cochabambae – hereafter subfasciatus group) should be split from nominate ruficapillus, and torquatus should be maintained as is. Still, we currently lack the required phenotypic data to support the split.

“Our recently published analyses (Bolívar-Leguizamón et al. 2024) show that divergence time based on demographic modeling between the subfasciatus group and nominate ruficapillus (2.34 Ma) is considerable and identical to that between torquatus and nominate ruficapillus [not that these values can be used as a yardstick, but they are greater than inferred divergence times between various others of antbird sister pairs]. Also, when considering inferred areas of low migration, our genomic data support a three-way split (subfasciatus group, nominate ruficapillus, and torquatus).

“I think that as more genomic and phenotypic data are included, northern and southern Andean populations may need to be split, too. In fact, Bukowski et al (2024) published a preprint integrating mitochondrial, genomic, and vocal data of some T. ruficapillus populations (unfortunately torquatus was not included as well as marcapatae and jaczewskii) suggesting that birds identified as cochabambae (southern Andes) can be distinguished genomically and vocally from birds identified as subfasciatus.”

“At this point, the evidence is spotty, but it points to a situation in which we are dealing with more than one species, we just need to know how many. We’ll only be able to figure it out based on a solid, comprehensive study integrating genomic, vocal, and plumage data of all populations involved. As I have mentioned in other cases (S. canadensis and T. shumbae), phenotypic variation in this complex is likely linked to environmental gradients in precipitation and seasonality, and the needed study must consider that. Until then, I am perfectly happy living with the status quo.

New reference:

B Bukowski, L Campagna, G Cabanne, P Tubaro, D Lijtmaer. 2024 Genetic and phenotypic differentiation in a Neotropical passerine with a disjunct distribution in the Andean and Atlantic forests (Thamnophilus ruficapillus) Authorea Preprints”

Comments from Zimmer: “NO. I’m very much on the fence with this one. I agree with many of the points that Rafael makes in the Proposal, but a couple of things give me pause. He makes the statement that “no vocal differences are known between ruficapillus and torquatus”. Indeed, to my ears, songs of the two are pretty indistinguishable. But, as far as I’m aware, no one has yet to do a quantified vocal analysis actually demonstrating that there are no differences. Also, I would suggest that any such vocal analysis would need to include calls as well as loudsongs in the mix. As has been demonstrated by the vocal analyses of the Hypocnemis cantator-complex (Isler et al. 2007), there are some thamnophilid species-complexes in which diagnostic differences in calls are concordant with plumage distinctions in supporting the treatment of populations as specifically distinct, even in the absence of any diagnosable differences in loudsongs. I’ve heard, and tape-recorded, multiple, structurally simple calls from both torquatus and ruficapillus, and I believe that I have noted some qualitative differences between them. However, without an actual analysis, that includes visual inspection of spectrograms for note shape differences, and, which included adequate samples of clearly homologous vocalizations, clearly identified to sex of the calling bird, I would be loath to make any definitive claim one way or the other. Elisa’s points regarding the hybrids found in Bolívar-Leguizamón et al (2023) being from the contact zone, and that “the preliminary data seem to indicate that both entities are maintaining their independence despite hybridization in the contact zone” are well taken. It seems to me that ecological/habitat distinctions between the two taxa are probably limiting their potential contact to areas of anthropogenic habitat degradation. Whatever the mechanism, the fact that I have yet to encounter an obvious hybrid or intergrade anywhere, suggests to me that the hybrid zone is narrow in scope. Given that, and the lack of any real vocal analysis, I’m reluctant to upset the status quo at this time, but readily acknowledge that more comprehensive genetic sampling and a thorough vocal analysis that includes calls as well as loudsongs, may well provide strong support for lumping these two as a single species.”