Proposal (996) to South American Classification Committee

 

 

Treat Xenops minutus as three species

 

 

That Xenops minutus may include multiple biological species has long been suspected (Remsen 2003). It is the objective of the present proposal to suggest that currently available evidence is sufficient to split X. minutus into three biological species.

 

Relevant information:

 

1. Xenops minutus includes three populations that differ markedly in vocalizations. An analysis of geographic acoustic trait variation documented three distinct vocal groups in Xenops minutus: one comprising the nominate taxon in the Atlantic rainforest south of the São Francisco River, one comprising all taxa in lowland Amazonia plus X. m. alagoanus of the Atlantic rainforest north of the São Francisco River, and one comprising trans-Andean birds from Central America to northwestern South America (Boesman 2016). Differences between the first two groups are particularly remarkable. Vocal differences between the nominate taxon and the Amazonian taxa sound as great or greater than those among the three species currently recognized in the genus Xenops.

 

2. Vocal differences between Xenops minutus populations generate behavioral discrimination. Field playback experiments between X. m. obsoletus (part of the Amazonian vocal group) and X. m. littoralis (part of the trans-Andean vocal group) revealed strong behavioral discrimination between these two vocal groups (Freeman & Montgomery 2017). Given that the vocal differences between these two vocal groups translate into behavioral discrimination (and presumably into premating isolation), the relatively greater vocal differences between the nominate subspecies and these two vocal groups should also expected to act as effective premating barriers if they were to come into contact.

 

3. The three vocally distinct Xenops minutus populations show high levels of genome-wide differentiation. A phylogenomic analysis of Xenops minutus found three deeply divergent clades experiencing little to no gene flow, entirely congruent with the three groups delineated by vocal variation (Harvey & Brumfield 2015).

 

Currently available evidence suggests that Xenops minutus is best treated as three biological species:

 

Xenops minutus — monotypic, in the Atlantic rainforest south of the São Francisco River.

Xenops genibarbis — polytypic, including obsoletus, ruficaudus, remoratus, and alagoanus.

Xenops mexicanus — polytypic, including ridgwayi, littoralis, olivaceus, and neglectus.

 

 

Note: Two of the three resulting species taxa are paraphyletic in mtDNA gene trees (Burney 2009; Smith et al. 2014, Harvey & Brumfield 2015). However, they are all reciprocally monophyletic in species trees inferred from genome-wide nuclear markers (Harvey & Brumfield 2015). Importantly, nuDNA variation is much more consistent with phenotypic variation than is mtDNA. For example, the nominate taxon, which is the most distinct in plumage and song, was not recovered as the most genetically distinct lineage in mtDNA gene trees, but was so in nuDNA trees. The latter are more likely to reflect the true species tree.

 

References:

Boesman, P. (2016). Notes on the vocalizations of Plain Xenops (Xenops minutus). HBW Alive Ornithological Note 85. In Handbook of the Birds of the World Alive. Lynx Edicions. https://doi.org/10.2173/bow-on.100085

Burney, C. W. (2009). Comparative phylogeography of Neotropical birds. PhD thesis. Louisiana State University. https://repository.lsu.edu/gradschool_dissertations/2682/

Freeman, B. G., & Montgomery, G. A. (2017). Using song playback experiments to measure species recognition between geographically isolated populations: A comparison with acoustic trait analyses. The Auk, 134(4), 857–870. https://doi.org/10.1642/AUK-17-63.1

Harvey, M. G., & Brumfield, R. T. (2015). Genomic variation in a widespread Neotropical bird (Xenops minutus) reveals divergence, population expansion, and gene flow. Molecular Phylogenetics and Evolution, 83, 305–316. https://doi.org/10.1016/j.ympev.2014.10.023

Remsen, J. V., 2003. Family Furnariidae (Ovenbirds). In: del Hoyo, J., Elliott, A., Sargatal, J., Christie, D.A. (Eds.), Handbook of the Birds of the World. Lynx Edicions, Barcelona, Spain, pp. 162–357.

Smith, B. T., et al. (2014). The drivers of tropical speciation. Nature, 515(7527), 406–409. https://doi.org/10.1038/nature13687

 

 

Rafael D. Lima, April 2024

 

 

Note from Remsen: Here is a photograph of specimens that includes the three major groups, from top to bottom: X. m. minutus, two from the mostly Amazonian genibarbis group (X. m. remoratus and X. m. ruficaudus), and X. m. mexicanus.

 

 

 

Note from Remsen: Here is the BirdLife International/HBW rationale for their 2-way split of X. genibarbis from X. minutus provided by Marshall Iliff:

 

“Hitherto treated as conspecific with X. genibarbis, but differs in its bold plain white chin and throat connecting to white submoustachial streak with relatively little brown on malar below it (prominent and completely separating chin from-submoustachial streak in X. genibarbis, all races of which have brown-streaked whitish throat) (2); less pale streaking-extending onto breast (1); loudsong a series of 4–5 upslurred notes, the first slightly lower-pitched and subdued, vs much-faster-delivered (2), more numerous and overslurred (3) notes (1). Claimed smaller size (in HBW) not apparent vs (at least)-nominate X. genibarbis, but further study needed; molecular evidence supports the split (2). Monotypic.-Distribution-E & SE Brazil (Bahia S to Santa Catarina), E Paraguay (E from Canindeyú, Caaguazú) and NE Argentina (Misiones).”

 

HBW/BLI did not mention splitting mexicanus group from genibarbis group.  The vocal information also comes from the Boesman analysis cited in the proposal.

 

Note on English names: If the proposal passes, we will do a separate one on English names.  HBW/BLI used “White-throated Xenops” for minutus and left everything else as Plain Xenops.

 

___________________________________________________________________

 

Comments from Robbins: “YES. Based on vocal and genetic data this seems to be a straightforward split.  Additionally, the clear white throat of nominate appears to be distinct from the other two taxa (based on the photo included in this proposal and Marshall's assessment in BirdLife/HBW. Thus, I vote to treat the current Xenops minutus as three species.”

 

Comments from Mike Harvey (voting for Bonaccorso): “YES. I believe this proposal is well justified and the recommendation is eminently reasonable. Although our 2015 genetic study was one of the first to use RAD-Seq for phylogeography/phylogeny, recent years have established the utility of the approach for this purpose. I think this fact, combined with the subsequent vocal and playback analyses, provides more-than-sufficient support for the existence of these three species.”

 

Comments from Del-Rio: “YES. Based on nuclear genome structure and phenotypic differentiation. Vocalizations are also pretty distinctive.”

 

Comments from Lane: “YES to the split into three species. The voices of these three groups are quite distinctive (more so than between several of the other species of Xenops, to my ear), and combined with plumage features, I think it makes sense to separate these three groups into species-level taxa.”

 

Comments from Stiles: “YES to the 3-way split of Xenops minutus, which is justified by multiple lines of evidence. Again, a proposal on E-names should be pending.”

 

Comments from Claramunt: “YES. I think the combination of plumage, voice, and genetic differences make a compelling case for treating this complex as three species. The mitochondrial information is somewhat muddling but it can be explained as vagaries of the coalescent process in large populations. The assignment of subspecies to each species looks correct. Note that Arbeláez-Cortés (2020) confirmed with mtDNA data that the sis-Andean neglectus, described as very similar to littoralis but with cinnamon rather than rufous wings and tail (Cory & Hellmayr 1923), is closer to the trans-Andean and Central American forms. It would be super interesting to see what happens in a potential contact zone in Colombia, but with the information at hand, the three-way split is the most reasonable classification.”

 

Arbeláez-Cortés, E. (2020). Defining the phylogeographic relationship between cis-and trans-Andean populations of Dendrocincla fuliginosa and Xenops minutus in Colombia. Revista Mexicana de Biodiversidad, 91: e912984

 

Comments from Areta: “YES. The Boesman note indicates the existence of 3 vocal types, each of which corresponds to the main deep breaks in phylogenetic structure shown by Harvey & Brumfield. However, no proper analysis has been published and we do not know where the samples that Boesman analysed came from. In a rapid survey of sounds, I found that this recording (https://xeno-canto.org/245253) in the eastern slope of the eastern Andes sounds like mexicanus on the other side of the Andes, and it therefore seems that the simple cis/trans Andean pattern is an illusion. Incidentally, this would seem to provide more support for the split of mexicanus from genibarbis. Overall, a look at available sounds recordings is consistent with Boesman´s analysis. The Atlantic Forest minutus is clearly distinct in plumage and vocalizations and it is difficult to argue in favour of its merger with the other two groups. In this case, it seems that vocalizations trump mtDNA in terms of taxonomic weight. I would much prefer to see a convincing integrative work discussing the details of vocalizations, genetics and plumage, but I think that there is enough data to move forward and accept the 3-way split.”

 

Comments from Remsen: “YES.  Vocal differences carry the most weight with me.  We even have the Freeman-Montgomery playback trial for one of the combos that is consistent with these vocalizations being important to species recognition, and we can use that to extrapolate (“if, then”) for the other combination.  I think the proposal makes it clear that burden-of-proof clearly falls on treating these taxa as a single species.

 

Comments from Curtis Marantz: “Looking at the spectrograms, I am not sure I would find the vocal differences between the Amazonian and Trans-Andean taxa to be overly compelling, at least if Xenops vary vocally in the way that most woodcreepers do.  The spectrograms presented in Boesman's additional notes demonstrate what I would consider to be compelling differences between the Atlantic Forest and other taxa, but not the other two, which are more reflective of variations on a theme as opposed to the different themes that I like to see for splitting taxa on the basis of vocal differences.”

 

Comments from Zimmer: “YES.  Make that a strong YES for recognizing nominate minuta as a distinct species from the others, based upon diagnostic vocal differences, genetic data, and consistent plumage differences.  The vocal differences between Atlantic Forest birds and all other populations currently treated in minuta, are much greater, at least to my ears, than say, the differences between the songs of X. tenuirostris and most subspecies of X. rutilans.  I’ve advocated for this split for a long time, and I already had a separate species account planned for the Atlantic Forest nominate subspecies in my forthcoming Brazil field guide.  As for the separation of Central American + trans-Andean populations from Amazonian populations, I’m a bit more hesitant here.  I think that both the plumage distinctions and vocal distinctions, although real, are less clear-cut – at least, I can’t personally vouch for the vocal distinctions of some Central American populations including all Trans-Andean taxa, so the break, as Nacho suggests, may not be the Andes.  However, the genetic work suggests otherwise, so I’m inclined to treat these two groups as distinct from one another as well.”