Proposal (966) to South American Classification Committee

Note from Remsen: This is a proposal submitted to and rejected (4 to 8) by NACC. Although the comments are not yet public, the NO voters agreed with the synopsis in the proposal, i.e., insufficient data to make formal changes in a complex situation, although multiple species likely involved.

Split Mionectes olivaceus into two species, M. olivaceus and M. galbinus

Background: Mionectes olivaceus is a small fruit-eating flycatcher with a range from Costa Rica to eastern Venezuela and Bolivia along lower montane slopes and nearby lowlands. The species was described in 1868 by Lawrence from Costa Rica. There are 5 recognized subspecies all of which were originally described as subspecies of olivaceus.

New Information: Boesman (2016) examined the voices of the various subspecies of olivaceus and concluded that there were 4 vocal groups. One was the nominate subspecies, olivaceus, found in eastern Costa Rica and western Panama. A second consisted of the taxa hederaceus and galbinus. The subspecies hederaceus occurs from Veraguas, Panama, south largely on the western side of the Andes to southern Ecuador whereas galbinus is known from the Santa Marta Mountains of Colombia. The third vocal group consists of venezuelanus from northern Colombia and northern Venezuela and fasciaticollis on east slope of Andes from southern Colombia to Bolivia. The fourth group is from Santander, Colombia and is not clearly associated with any of the recognized taxa. Boesman evaluated the groups using the Tobias (2010) criteria and scored olivaceus versus the other taxa as 8. He scored hederaceus/galbinus versus venezuelanus/fasciaticollis as 6.

The sonograms here are from the Birds of the World account (Fitzpatrick et al 2020). A is nominate canfasciaticollis/ venezuelanus, D is the Santander population of uncertain subspecies, and E is a single recording from the far eastern end of the northern Venezuela mountain, which currently would be assigned to venezuelanus. Recordings of songs from populations in the Perijá Mountains and Trinidad do not seem to be available.

These songs are squeaky and extremely high pitched, basically between 8 and 10 thousand MHz I personally cannot hear the sounds on recordings of nominate olivaceus (at least through my computer) at all, and am not hearing the other populations’ vocalizations well. The literature describes these songs as insect- or hummingbird-like.

Morphological variation among the five subspecies is minor. The main variation is in the brightness of the green upperparts, the brightness and tone of yellow on abdomen, the extent of streaking and paleness of streaking. None of this variation would be apparent in the field. In del Hoyo and Collar (2016), they assign some points toward the Tobias criteria score based on plumage. However, I would say, from examining specimens and the discussion in Fitzpatrick et al (2020), that olivaceus is not at the extreme in any of the characters varying across the 5 subspecies. It is generally pale and bright, but to my eye, galbinus is the palest and brightest of these taxa, whereas hederaceus is overall the dullest.

At the time of del Hoyo and Collar (2016) there was no genetic evidence available for this complex. However, Harvey et al (2020) examined multiple samples of olivaceus as well as the other species of Mionectes. In their analysis, Mionectes olivaceus was paraphyletic, with nominate olivaceus, sister to a clade with Mionectes striaticollis and samples corresponding to galbinus and venezuelanus. M. striaticollis is a similar species broadly sympatric with olivaceus in South America, overlapping with the subspecies hederaceus, fasciaticollis, and venezuelanus.

Recommendation: This is a complicated issue. I think it is very likely there are multiple species within the current Mionectes olivaceus. Splitting nominate olivaceus is supported by a distinctive voice and the genetic evidence that it is not sister to the rest of the species. If NACC did this, we would add a species to the overall list, because olivaceus occurs in Costa Rica and western Panama, and hederaceus representing Mionectes galbinus occurs in central and eastern Panama. However, it seems likely that there are multiple species to be recognized in galbinus. Unfortunately, I think we currently lack sufficient information to define the various species that would be left in galbinus. There has not been genetic analysis done at a relevant scale for that question, and many populations (including galbinus) do not have recordings in Xeno-canto or Macauley collections. One vocal group (D from Santander above) has not been clearly assigned to a named taxon. For NACC, the issue is that the name galbinus may not be applicable in the end to the populations in Panama. This case reminds me of Schiffornis turdinus for SACC. A proposal to split in 2007 failed to pass, but with additional data, in 2011 a new proposal to split Schiffornis turdinus into 5 species did pass.

My weak recommendation is a NO vote, awaiting further evidence that will allow us to define more clearly the multiple species that likely make up Mionectes olivaceus.

English names: Del Hoyo and Collar (2016) use Olive-streaked Flycatcher for M. olivaceus and retain Olive-striped Flycatcher M. galbinus. Because the ranges are very uneven in size with M. galbinus much more widespread than M. olivaceus, I think the use of Olive-striped Flycatcher for M. galbinus is justified under our English name criteria. Further, given that the splitting up of galbinus into two or more species seems likely eventually, we (or SACC) would have to coin new names for the daughter species at a later time. I recommend using Olive-streaked Flycatcher for Mionectes olivaceus and Olive-striped Flycatcher for Mionectes galbinus if we decide to split these two groups. These are the names used by del Hoyo and Collar 2016, and used for the groups corresponding to the species recognized by del Hoyo and Collar in the Birds of the World account (Fitzpatrick et al. 2020).

References:

Boesman, P. F. D. 2016. Notes on the vocalizations of Olive-striped Flycatcher (Mionectes olivaceus). Birds of the World Ornithological Note 117. In: Birds of the World. Cornell Lab of Ornithology, Ithaca, NY. (retrieved from birdsoftheworld.org/bow/ornith-notes/JN100117 on 25 March 2022.).

del Hoyo, J., and N. J. Collar. 2016. HBW and BirdLife International Illustrated Checklist of the Birds of the World. Volume 2. Passerines. Lynx Edicions, Barcelona, Spain.

Fitzpatrick, J. W., J. del Hoyo, N. Collar, E. de Juana, G. M. Kirwan, and A. J. Spencer. 2020. Olive-striped Flycatcher (Mionectes olivaceus), version 2.0. In Birds of the World (T. S. Schulenberg and B. K. Keeney, Editors). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.olsfly1.02

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. C. Faircloth, S. V. Edwards, J. Pérez-Emán, R. G. Moyle, F. H. Sheldon, A. Aleixo, B. T. Smith, R. T. Chesser, L. F. Silveira, J. Cracraft, R. T. Brumfield, and E. P. Derryberry. 2020. The evolution of a tropical biodiversity hotspot. Science 370:1343-1348.

Tobias, J.A., N. Seddon, C. N. Spottiswoode, J. D. Pilgrim, L. D. C. Fishpool, & N. J. Collar 2010. Quantitative criteria for species delimitation. Ibis 142:724-746.

Doug Stotz, February 2023

Note from Remsen on English names: if this passes, a separate proposal would be needed; see discussion of possibilities above.

Comments from Remsen (as in my NACC comments): “NO. Tough call for me. As noted in the proposal, M. olivaceus is a paraphyletic taxon certainly consisting of multiple species, and Boesman’s preliminary analysis indicates vocal differences consistent with species rank for several populations. But this is a case in which a piecemeal approach seems unwise, especially in the case of galbinus-hederaceus, because we don’t have genetic data from what would be the taxon in the NACC area (hederaceus) (Harvey et al. included only 3 subspecies of M. olivaceus) and because according to the proposal, galbinus and hederaceus are the most divergent taxa in terms of plumage. Boesman and Harvey et al. have set the stage for a future, thorough analysis that will allow us to sort all this out in terms of taxonomy.

“One could make the argument that it is more important to split olivaceus into two species to remove the problem of maintaining a paraphyletic species than it is to wait to sort out all the details, which affect mainly extralimital taxa (South America) anyway. I understand that rationale, and I am convinced from the published sonograms and measurements of the vocal parameters in Boesman (2016) that the NACC area has two species-level taxa (nominate olivaceus vs. hederaceus). I am not opposed to piecemeal taxonomic change, would support it if we were reverting to a prior treatment rather than a novel one, and would support that split if the data behind it were solid. But are they solid? The genetic data for paraphyly are based on Harvey et al. 2020, excerpted below:

“Note that only 5 individuals were sampled: 2 M. striaticollis and 3 M. olivaceus. The proposal notes that the 3 olivaceus samples represent 1 sample each of nominate olivaceus, galbinus, and venezuelensis (the latter two the sisters in the tree). That only leaves out Boesman’s Santander vocal group* (which evidently cannot be assigned to subspecies yet). But it also leaves out two critical subspecies, including the one from the NACC area: hederaceus. Splitting everything from nominate olivaceus as one species assumes that hederaceus and fasciaticollis group with galbinus and venezuelensis rather than olivaceus. That is the most likely outcome based on biogeography, but it is not a certainty. Hederaceus is the closest taxon (parapatric evidently) to nominate olivaceus. We are assuming hederaceus groups with galbinus because of Boesman’s vocal analysis (more on that later), but the proposal noted that they are at opposite extremes of plumage variation in the complex, so that is a concern. Even fasciaticollis, the most distant taxon from nominate olivaceus, occurs as close as southern Colombia, and cannot be dismissed completely as a potential disjunct, sister taxon to nominate olivaceus.

“And then there is another potential problem: there are two additional unsampled named taxa that were subsumed into venezuelensis on the say-so of Fitzpatrick in HBW without any real analysis: pallidus Chapman, 1914 (TL = Buena Vista, 4500 ft., above Villaviscensio, Meta, Colombia), and meridae Zimmer, 1941 (type loc = near Mérida, Venezuela). Mel Traylor, in his Tyrannidae chapter in “Peters” 1979 recognized both subspecies. Traylor gave the range of pallidus as “Upper tropical and lower subtropical zones of Eastern Andes of Magdalena and northern Meta, Colombia” and of meridae as “Subtropical zone of northwestern Venezuela from western Zulia and Falcon to Táchira, and Norte de Santander and Boyacá in adjacent northeastern Colombia.” Note that according to the proposal Norte de Santander is region from which Boesman described a separate vocal group, and so I strongly suspect Fitzpatrick unjustly sunk meridae; here’s what Fitzpatrick said: “pallidus …. and meridae … are indistinct, intermediate forms within a cline, both treated as synonyms of venezuelensis.” That’s it. It should be noted that Fitzpatrick synonymized a number of tyrannid subspecies based on statements like this, and that some of these synonymized taxa are actually vocally distinct (e.g. Phyllomyias griseiceps, which he treated as monotypic).

“As far as Boesman’s vocal data go, his N for nominate olivaceus was 6 and for hederaceus, 19. So that looks solid, and just eyeballing the sonograms suggests to me that these are the most different of the vocal groups. Then, for galbinus from the Santa Martas, N=1. On the basis of that one recording, Boesman groups galbinus with our hederaceus. Again eyeballing the sonograms, that single recording does bounce around like hederaceus, but to my eye, the note shapes (lopsided “U”) of galbinus look more like Andean venezuelensis and fasciaticollis than they do those of hederaceus ….Which actually makes more sense biogeographically in terms of relationships. Boesman does not have a specific metric for note shape, by the way.

“So, in summary, we would be making a novel taxonomic change based on 5 individual genetic samples that do not include the critical taxon hederaceus, which happens to be the one of the largely South American group represented in the NACC area. Yet we would be assigning the species name galbinus to that species, and hederaceus would be a subspecies of galbinus. And that taxonomy would be based on Boesman’s analysis of a single recording, the sonogram of which to my eye looks more like the Andean taxa, not hederaceus. Further, our case for paraphyly is actually weak with respect to the two taxa in our area because we don’t have a genetic analysis that included hederaceus. Am I the only one queasy about all this? Yes, we’ve got two species in the NACC area, and we can punt the other problems to SACC, but it just seems awfully sloppy and laden with compound assumptions. Imagine the egg-on-face fallout if there is something wrong with our current understanding once we have genetic and vocal data from all 5 (or 7) taxa? I don’t see why we should take that risk.

“* my downloaded version of Boesman (2016) states that this vocal group is from Caripe, Venezuela, and thus pertains to venezuelensis, so I’m not sure what’s going on here except that Doug must have access to a more recent revision.”

Comments from Areta: “Harvey et al. (2020) show that olivaceus is deeply paraphyletic (Panama sample would be olivaceus, Colombian samples would be galbinus from Santa Marta and hederaceus from Tamá):

“The problem I see is that by splitting the nominate olivaceus from the rest, we would have a new use for the name galbinus to be applied to what seems to be at least 2-3 species, thereby leading to a new faulty taxonomy.

“The variation in vocalizations of these Mionectes requires more thorough quantification. In a rapid examination in XC and in the NACC proposal, the differences in sounds do not match the species limits outlined by BirdLife, and because so much variation remains unexplained, I don´t think a good case has been made to adopt this split, even when it seems quite clear that there are multiple species within M. olivaceus as currently circumscribed.

“I am seriously concerned by these “express” splits. I have no doubt that olivaceus is a different species, but as a taxonomic committee, I think that we should be worried about deeper issues, such as the new application of the name galbinus (a Santa Marta endemic?) at the species level for another array of species. Then this rather unsatisfactory usage will spread, only to be changed sooner than later. As such, I prefer to stick to known historical errors of wide application instead of incurring in a new kind of error affecting more taxa (e.g., venezuelanus would in all probability have now been erroneously considered to pertain to olivaceus AND galbinus). To me, we do a disservice to stability by accepting splits such as this one, and this is why to me not all the "clear splits" are equally acceptable. We have to deal with species-level entities and their names, and this is where things get muddy. In this case, I don´t think that the piecemeal approach to taxonomy is adequate. This complex needs to be sorted in one movement, lest we create a large number of ephemeral taxonomies and ensuing confusion.”

Comments from Lane: “NO, largely for the reasoning Van puts forth. Whereas I suspect splits will be necessary once we gain a better understanding of the various taxa involved (both molecularly and vocally), the taxonomy makes me leery of splitting using the wrong daughter names (much as was almost the case with Herpsilochmus rufimarginatus and frater). As it happens, I literally made a recording yesterday in the Santa Martas-- as “bycatch” in the background of another recording (and thus ID not visually confirmed)--that has the same vocalization that Niels identified as M. o. galbinus (XC235896). My recording uploaded here: https://macaulaylibrary.org/asset/534205891.

“I note that XC235896 has in its notes that the singer was *not seen* and so I don’t know what grounds Niels used to identify that voice as M. o. galbinus! However, I used Niels' recording to play back to a bird I was watching later yesterday, and it seemed briefly to respond positively, approaching me once before returning to foraging (and not showing interest again despite more playback). Given the lack of representation of key taxa in a phylogeny, the lack of confirmation of ID of several key recordings (such as the XC recording above), and the reluctance to have to backtrack on species names if the results of those future studies contradict what has been proposed, I would rather stay at status quo until things are clearer.”

Comments from Robbins: “NO. Clearly, multiple species are involved, but as Van and Dan point out we need additional information on a number of facets of this complex before any splits are made.”

Comments from Stiles: “NO for splits at this time; the plumages of all are very difficult to tell apart, and unfortunately, my post-malarial ear is deaf to the songs, so I’d like to see a more thorough genetic analysis with birds also recorded and if possible, a colorimetric analysis to get this one sorted out.”

Comments from Del-Rio (who has Pacheco vote): “NO. I would like to see a more complete study of this complex, including sound recordings and molecular data.”

Comments from Zimmer: “NO. The molecular data are pretty clear that M. olivaceus, as currently constituted, is paraphyletic. But Boesman’s vocal presentation, while a good start, leaves the question of how many splits should be recognized as the taxonomic equivalent of a “jump ball”. I would want to see a more broadly sampled, detailed, quantitative vocal analysis, not just a handful of spectrograms, no matter how much those tracings would appear to differ.”

Comments from Claramunt: “NO. Vocal and genetic data really suggest that the taxon in Costa Rica and W Panama is a different species, but I agree with others in that more evidence is needed, in particular, data about birds in central and eastern Panama.”