Proposal (790) to South American Classification Committee

Change species limits within Ramphocaenus melanurus

Effect on SACC classification: Change number of species of Ramphocaenus from one to as many as three.

Background: From SACC notes:

The rufiventris subspecies group of Middle America and western Colombia was formerly (e.g., Cory & Hellmayr 1924) considered a separate species from Ramphocaenus melanurus, but see Zimmer (1931) for rationale for treating them as conspecific. Harvey et al. (2014) proposed that the subspecies sticturus (with obscurus) should be treated as a separate species from other Amazonian populations because of local sympatry without any sign of interbreeding along with strong vocal differences. SACC proposal needed.

For much of the past century, Ramphocaenus has consisted of a single, polytypic species: R. melanurus. As stated above, Zimmer provided the rationale for combining the trans-Andean R. rufiventris with the cis-Andean R. melanurus. Recent fieldwork and lab work have shown that this may not have been the correct move, and that in fact multiple species are involved within the genus. The two cis-Andean groups (melanurus and sticturus) have been shown to be locally sympatric, differing strongly in voice (Harvey et al. 2014). The recently published phylogeny of the Polioptilidae further shows that a clade containing sticturus/obscurus is sister to the remainder of melanurus and that the latter has a deep node within it to suggest that two species may be involved with the biogeographic break across the Andes (Smith et al. 2018).

Analysis and Recommendation:

Smith et al. (2018) provided a fairly exhaustive phylogeny of the Polioptilidae. Within this study, they sampled Ramphocaenus adequately and provided the first phylogenetic tree of the genus, showing that it comprised three clades: the widespread cis-Andean R. melanurus group (including taxa austerus, albiventris, badius, duidae, melanurus, amazonum, and trinitatus; though some of these were only represented by ND2 sequences, and not nuclear datasets), a cis-Andean sticturus group (including sticturus and obscurus) restricted to southern and western Amazonia, and a trans-Andean rufiventris group (including ardeleo, rufiventris, and sanctaemarthae; again the last only represented by ND2). The position of the Colombian form griseodorsalis (another ND2 only sample) remained unresolved. The branching pattern placed sticturus as sister to the other two groups. These results complement field observations reported by Harvey et al. (2014), in which vocal distinctions and local sympatry supported considering obscurus and badius as biological species (representing the sticturus and melanurus groups respectively).

As noted in the SACC note above, older authorities have considered cis- and trans-Andean forms of R. melanurus to be separate species, and although the vocal differences between them are not as strong as between the sticturus group and the remaining forms, there appear to be vocal and plumage distinctions that would support returning to separate species across the Andes. The lack of resolution of Colombian form griseodorsalis (and the confounding variation in songs in the Magdalena and Cauca valleys as per Xeno-canto) may weaken the evidence for this split until more Colombian material is available to resolve the relationships of taxa there.

I will break this proposal into two votes:

790A: separate Ramphocaenus sticturus (including obscurus) from the remaining R. melanurus. I strongly recommend a YES on this vote based on phylogenetics, voice, plumage, and sympatry.

790B: separate trans-Andean R. rufiventris (including ardelo and sanctaemarthae) from cis-Andean R. melanurus. Frankly, I’m somewhat ambivalent on this vote. The Colombian taxon griseodorsalis muddies the issue enough that I am not sure where to place it with respect to the rest of the complex. It may be prudent to await further sampling to clarify relationships.

If these two votes pass, I recommend the following English names (the compound names from Hellmayr do not seem appropriate for these daughter species):

R. sticturus—Chattering Gnatwren

R. melanurus—Trilling Gnatwren

R. rufiventris—Northern Gnatwren

Literature cited:

Harvey, M. G., D. F. Lane, J. Hite, R. S. Terrill, S. Figueroa R., B. T. Smith, J. Klicka, and W. Vargas C. 2014. Notes on bird species in bamboo in northern Madre de Dios, Peru including the first Peruvian record of Acre Tody-Tyrant (Hemitriccus cohnhafti). Occasional Papers of the Louisiana State University Museum of Natural Science 81:1-38

https://sites01.lsu.edu/wp/mnspapers/files/2016/05/81.pdf

Smith, B. T., R. W. Bryson, Jr., W. M. Mauck III, J. Chaves, M. B. Robbins, A. Aleixo, and J. Klicka. 2018. Species delimitation and biogeography of the gnatcatchers and gnatwrens (Aves: Polioptilidae). Molecular Phylogenetics and Evolution 126:45-57.

https://doi.org/10.1016/j.ympev.2018.03.012

Daniel Lane, April 2018

Comments from Robbins:

“A. I vote YES for elevating Ramphocaenus sticturus to species level based on multiple data sets.

“B. As Dan points out, additional sampling is needed across northern Colombia to help elucidate whether trans-Andean rufiventris should be treated as a species separate from cis-Andean melanurus. So, for now, I'll go with the status quo and vote NO for recognizing the rufiventris group as a species.”

Comments from Stiles: “A: YES, this split is justified by multiple sources of data. B: NO for now; hopefully, someone here will provide data to resolve the inconsistencies in the Colombian situation.

Comments from Pacheco: “A. YES. The available data give a good safety in the decision on splitting Ramphocaenus sticturus.

“B. NO. In this case, it is cautious to wait for more data.”

Comments from Stotz: “A. YES

“B. NO. I suspect we will eventually make this split, but currently it seems like the data in support are not sufficient.”

Comments from Jaramillo: “A – YES, well-backed by multiple data sets.

“B – Yes, I will go against the grain on this one. If one is basing a good deal of this decision on the genetic data that is available, it seems odd to split two of the lines but not the third. Although there may be some unresolved issues, the major pattern of three clades (at least) stands. So, I think it is best to separate all three as species. The cis-trans Andean separation in the molecular data is pretty striking to me.”

Comments from Areta: “A. YES. Local sympatry, marked vocal differences, and genetic data support recognition of Ramphocaenus sticturus as a distinct species from R. melanurus.

“B. NO. This may well be a valid split, but until more thorough genetic, morphological and vocal analyses convincingly sort out geographic variation into distinct clusters (or not) I feel more comfortable without recognizing R. rufiventris as a species-level entity.”

Comments from Claramunt: “A. YES. The lack of signs of interbreeding between obscurus and amazonus along a wide area of parapatry in Peru indicates intrinsic reproductive isolation.

“B. YES. It was considered a separate species until the lumping madness infected avian systematics in the mid XX century. griseodorsalis was considered a subspecies of rufiventris. The node that unites griseodorsalis to the sticturus clade does not have statistical support and I suspect, given previous taxonomies and its conspicuous gray mantle, that griseodorsalis is indeed related to the trans-Andean clade, as expected.