Proposal (961) to South American Classification Committee

Treat Tolmomyias viridiceps as a separate species from T. flaviventris

Our SACC note reads:

"77. Tolmomyias flaviventris almost certainly involves more than one species; see Bates et al. (1992) and Ridgely & Tudor (1994). The subspecies viridiceps is almost certainly a distinct species, and was so considered by Ridgely and Greenfield (2001) and Hilty (2003). However, Zimmer (1939a) considered them conspecific because he considered the subspecies subsimilis and dissors to represent taxa that were intermediate between the two, and this treatment was followed by Fitzpatrick (2004) in the absence of published data supporting a split. SACC proposal needed."

Ridgely & Tudor (1994) makes for an interesting reading as an introduction to the situation:

The unpublished BSc thesis of Almeida (2017) provides a wealth of data on the phylogeny of T. flaviventris (based largely on mtDNA data) and some evidence on the aspect and vocalizations. All but one figure in the proposal were extracted from this thesis.

This study obtained genetic sequences of subspecies flaviventris, aurulentus and dissors in the flaviventris group, and only nominate viridiceps from the viridiceps group (missing zimmeri and subsimilis).

They found that viridiceps (Clade A, blue) was sister to the other samples in the flaviventris group (where four clades were identified). There is at least on area of known of known overlap between the two groups of taxa in the Madeira-Tapajos interfluvium (Almeida 2017) in which three genetic clusters are found (Clades A, B and E), but they could potentially meet across a broad area in the western Amazonia of SE Colombia, CW Brazil and N Bolivia.

The genetic p-distance between viridiceps and the flaviventris group were large (between 3.1 and 4.5%):

One sample of what should be viridiceps, from Cobija (Pando, NW Bolivia), and one from flaviventris from Piaui (NE Brazil) were sister in the work of Harvey et al. (2020). Dashed lines indicate 2 MY:

The separation into two species seems to capture most of the variation in song (see for example descriptions in Herzog et al. (2016) for Bolivia, and Ridgely & Tudor (1994 posted above). Although there are sampling gaps in the critical areas in which both taxa presumably approach closely, it is worth highlighting that recordings of both forms in N Bolivia, CW Brazil. and possibly also SE Colombia can be easily identified to the corresponding taxa without signs of intermediacy.

https://xeno-canto.org/species/Tolmomyias-flaviventris
https://xeno-canto.org/species/Tolmomyias-viridiceps

The "intermediate" plumages of subsimilis and dissors discussed by Zimmer (1939) might be a cause for concern to some. However, note that vocalizations of subsimilis indicate its belonging to the viridiceps group (no genetic samples available), while both vocalizations and genetics (mostly mtDNA) indicate the belonging of dissors to the flaviventris group.

Below is a sample of specimens from each group (perhaps showing the extremes and not the "intermediate" populations):

Recommendation: I recommend a YES vote to the split of T. viridiceps. Given the degree of vocal and morphological differentiation it seems difficult to envision that there will be genetic flow in the Madeira-Tapajos interfluvium, and other possible zones of parapatry/sympatry. Also, if there is some flow, it seems to be restricted based on the close proximity of vocal and genetic types." 

References:

HARVEY, M. G., G. A. BRAVO, S. CLARAMUNT, A. M CUERVO, G. E. DERRYBERRY, J. BATTILANA, G. F. SEEHOLZER, J. S. MCKAY, B. C. O’MEARA, B. G. FAIRCLOTH, S. V. EDWARDS, J. PÉREZ-EMÁN, R. G. MOYLE, F. H. SHEDLON, A. ALEIXO, B. T. SMITH, R. T. CHESSER, L. F. SILVEIRA, J. CRACRAFT, R. T. BRUMFIELD, AND E. P. DERRYBERRY. The evolution of a tropical biodiversity hotspot. Science 370: 1343-1348.

Marques Almeida, C., 2017. Filogeografia de Tolmomyias flaviventris (Wied, 1831). Aves: Rhynchocyclidae. BSc Thesis. Museu Paraense Emílio Goeldi, Belem, Brazil).

J. I. Areta, February 2023

Note from Remsen on English names: If the proposal passes, we will need a separate proposal on the English name, so be thinking about this. Ridgely & Tudor (2001) used the name “Olive-faced Flatbill” for T. viridiceps.

Comments from Lane: YES to splitting T. viridiceps from T. flaviventris. The distinctiveness of the voices of these two groups, and the apparent sympatry pointed out by Almeida, make this split quite straightforward for me!”

Comments from Stiles: “YES to splitting viridiceps from flaviventris, based primarily on the Almeida data, which look to be solid evidence.”

Comments from Zimmer: “YES. This one is pretty straightforward given the genetic data and apparent sympatry of the two groups in the Madeira-Tapajós interfluve as presented in Almeida (2017), and, given the solid vocal differences between the two groups, which has been known to fieldworkers for some time.”

Comments from Remsen: “NO. Although I’m certain from what is assembled here that two species are involved, I’m going to be very picky about our standards of evidence. What we have is anecdotal, qualitative information combined with an unpublished BSc. thesis that did not sample two of the three taxa assumed to be associated with viridiceps, and sample sonograms from xeno-canto. I have no reason to doubt that any of the evidence presented is incorrect but plenty of reason to doubt that this is sufficient for changing the status quo. I have the feeling that we are rushing this one through because most of us ‘know’ two species are involved.”

Comments from Claramunt: “YES. The nice study by Almeida is convincing.”

Comments from Bonaccorso: “NO. Judging by the mitochondrial evidence, there is good enough genetic differentiation between viridiceps (clade A) and the other four populations in the tree (B, C, D, E), but this could be just genetic (population-level) structure. I don´t see clear evidence of reproductive isolation among these populations. Plumage differences are so subtle that I bet it would be challenging to identify potential hybrids or intergrades; also, we are basing a decision on “apparent sympatry.” Finally, I don´t know much about calls, but it seems that more data (from several individuals) should be used to support the case.”

Comments from Mario Cohn-Haft (voting for Pacheco): “NO. The flaviventris group is less obvious to me based on the info presented. First off, I'm not sure exactly what the proposal is. Is it to split only viridiceps out of a still polytypic flaviventris? That was my initial understanding based on the wording. But, if as is implied in the discussion of voices, subsimilis is to be part of viridiceps and the rest (?) presumably to stay in flaviventris, then i think that needs to be made explicit. In other words, I'm not sure how to vote without knowing exactly what taxa are supposed to go where. I guess I'd be inclined to vote NO as currently worded (or at least as i currently understood the proposal) for lack of clear evidence of what information supports what relationships.”

Additional comment from Areta: “Mario: The proposal is to include viridiceps, subsimilis, and zimmeri under T. viridiceps.”

Comments from Robbins: “YES. Although we haven’t been given a copy/access to the unpublished thesis, it has long been appreciated, and can readily be heard via on-line audio resources, that more than one species is involved in Tolmomyias flaviventris. So, even though I appreciate Van’s sentiments concerning published evidence, this seems clear enough to at least recognize the population of the viridiceps clade that was sampled as a species. Moreover, to be consistent with my evaluation of the Myiophobus fasciatus proposal, which has less documentation (e.g., no genetic data), I vote “Yes” for recognizing viridiceps as a species.”

Comments from Glenn Seeholzer (voting for Jaramillo): “YES. Correspondence between vocalizations and genetics places the burden of proof on those that would keep these vocal+genetic groups as a single species.

“- Mario is correct that the proposal doesn't state explicitly what subspecies go where. For me, it is implied that subspecies viridiceps, subsimilis, and zimmeri go in viridiceps and subspecies flaviventris, aurulentus, and dissors go in flaviventris. Perhaps Nacho can amend the initial proposal to make this clear

“- vocalizations have been shown to more closely track evolutionary history than plumage in Zimmerius (Rheindt et al. 2008, 2014). Given the well-known vocal differences amongst subspecies (ahem, species) within Tolmomyias assimilis and sulphurescens with relatively minor plumage variation, I'm inclined to also not worry too much about the 'intermediate' plumages of subsimilis and dissors. The eye-ball-a-series approach of Zimmer and others can uncover remarkably subtle variation, but I'd prefer something more quantitative before fully buying into the idea that there are intermediate plumages and the implication of gene flow.

“- peer-reviewed and published would be ideal, but this thesis is a far more complete analysis than what was available when these species were split by other taxonomies in 2016. There is also a grey-literature analysis by Peter Boesman (2016) (https://birdsoftheworld.org/bow/ornith-notes/JN100123) of these species vocalizations confirming what many have long observed, that there are two vocalization groups each with easily distinguished primary vocalizations with non-overlapping variation.

“- There are only two genetic samples that come from a contact zone between Clades A (viridiceps) and Clade B+C+D+E (flaviventris) along the Amazon close to the Tapajos. More would be desirable, but this does mean that we can be reasonably confident that the genetic clades match the vocal groups.

“- The three regions below are where these forms are documented to come close to being in contact based on vocalizations (see Figure 1). I've included linear distance between the nearest localities. Especially for Bolivia and Colombia, there are no obvious intervening biogeographical barriers, so it seems likely that they come into close contact with possibly syntopy somewhere in these regions. For Central Brazil along the Amazon, these taxa are riverine / disturbed habitat species and probably not greatly affected by the Amazon.

- Central Brazil along the Amazon (358 km)
- N Bolivia (134 km)

- Central Colombian Amazon - Central and Southern Serrania de Chiribiquete (125 km, wouldn't that be an adventure to find that contact zone!).

flaviventris from N Chiribiquete
https://search.macaulaylibrary.org/catalog?view=list&unconfirmed=incl&captive=incl&taxonCode=yebfly3&mediaType=audio&regionCode=CO-CAQ (ML252714 is an outlier, but the other two recordings from Alvarez are clearly flaviventris)

viridiceps from S Chiribiquete
https://search.macaulaylibrary.org/catalog?view=list&unconfirmed=incl&captive=incl&taxonCode=yebfly4&mediaType=audio&regionCode=CO-CAQ

“Figure 1 - Map of records of the viridiceps group (purple) and the flaviventris group (green). Triangles are Macaulay Library and circles are Xeno-Canto.

Boesman, P. (2016). Notes on the vocalizations of Yellow-breasted Flycatcher (Tolmomyias flaviventris). HBW Alive Ornithological Note 123. In: Handbook of the Birds of the World Alive. Lynx Edicions, Barcelona. https://doi.org/10.2173/bow-on.100123

Rheindt, Frank E., Matthew K. Fujita, Peter R. Wilton, and Scott V. Edwards. “Introgression and Phenotypic Assimilation in Zimmerius Flycatchers (Tyrannidae): Population Genetic and Phylogenetic Inferences from Genome-Wide SNPs.” Systematic Biology 63, no. 2 (March 2014): 134–52. https://doi.org/10.1093/sysbio/syt070.

Rheindt, Frank E., Janette A. Norman, and Les Christidis. “DNA Evidence Shows Vocalizations to Be a Better Indicator of Taxonomic Limits than Plumage Patterns in Zimmerius Tyrant-Flycatchers.” Molecular Phylogenetics and Evolution 48, no. 1 (July 2008): 150–56. https://doi.org/10.1016/j.ympev.2008.04.016.