Proposal (957) to South American Classification Committee

Generic limits in Ochthoeca, Silvicultrix, Tumbezia and Colorhamphus

Our SACC notes read as follows:

"118. Lanyon (1986a) merged Tumbezia into Ochthoeca based on morphological data, followed by Sibley & Monroe (1990), and genetic data confirm that it is embedded within Ochthoeca (Tello et al. 2009; cf. García-Moreno et al. 1998). SACC proposal badly needed.

“119. Lanyon (1986a) separated Ochthoeca frontalis, O. jelskii, O. diadema, and O. pulchella into the genus Silvicultrix, and genetic data (Tello et al. 2009; cf. García-Moreno et al. 1998) confirmed that the Silvicultrix group is more closely related to Colorhamphus. SACC proposal badly needed."

Phylogenetic data in Tello et al. (2009) and Harvey et al. (2020) provides a solid framework to work on generic limits in chat-tyrants. I would say that there are 3 issues here, so please indicate your choice on these when voting:

A: Use Silvicultrix for frontalis, pulchella, diadema and jelskii?
B: Merge Tumbezia salvini into or Ochthoeca?
C: Merge Colorhamphus parvirostris into Ochthoeca?

Also please see Miller & Greeney (2008) for nests of Silvicultrix as different from Ochthoeca. The high-pitched vocalizations of Silvicultrix are also quite different from those of Ochthoeca.

Tree from Tello et al. (2009):

Tree from Harvey et al. (2020) (lines spaced by 2MY):

Discussion and recommendations: My recommendations are as follows:

A: YES to Silvicultrix. Lanyon (1986) makes an excellent case for the recognition of this genus.

B: YES to include Tumbezia in Ochthoeca, following Lanyon (1986) and the position as sister to the remainder of Ochthoeca.

C: NO to including Colorhamphus in Ochthoeca. Colorhamphus is phenotypically and genetically quite different from Ochthoeca sensu stricto (and Tumbezia).

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.

LANYON, W. E. 1986a. A phylogeny of the thirty-three genera in the Empidonax assemblage of tyrant flycatchers. American Museum Novitates 2846: 1–64.

Miller, E.T. & Greeney, H.F. 2008. Clarifying the nest architecture of the Silvicultrix clade of Ochthoeca chat-tyrants (Tyrannidae). Ornitologia Neotropical 19: 361–370.

TELLO, J. G., MOYLE, R. G., D. J. MARCHESE, AND J. CRACRAFT. 2009. Phylogeny and phylogenetic classification of the tyrant flycatchers, cotingas, manakins, and their allies (Aves: Tyrannides). Cladistics 25: 429-467.

J. I. Areta, February 2023

Comments from Stiles: “YES to A,B,C; the combination of morphological and genetic data is unbeatable (and in my limited experience with the Colombian species, vocalizations also fall into line with the proposal).”

Comments from Lane: “YES to A; NO to B and C. I have made some effort to get to know the members of Ochthoeca (sensu lato) fairly well, at least in the northern and central Andes, and I support Silvicultrix being removed and returned to its own genus. Tumbezia, however, is decidedly distinct from the remaining Ochthoeca in elevation, behavior, coloration, and vocalizations. It does not seem to perform regular duets which the true Ochthoeca do often… in fact, individuals of Tumbezia are quite solitary by comparison (I've not seen them moving in pairs at all). They are also not bold and extroverted, as most of the true Ochthoeca are (perhaps with the exception of the O. cinnamomea group). If I understand the wording of the proposal correctly, the reasoning to absorb Tumbezia into Ochthoeca is because Ochthoeca (sensu lato) was polyphyletic in the Tello et al. and Harvey et al. trees... but if one removes Silvicultrix from Ochthoeca, that is *no longer the case*! So rather than adding this odd-one-out species into an otherwise fairly uniform Ochthoeca, I think it makes far more sense to continue recognizing Tumbezia as a monotypic genus. I don’t know Colorhamphus, but recordings on XC don’t give me the impression that they regularly duet. So, if I vote for Tumbezia to remain monotypic, then by the phylogeny built by Harvey et al, then Colorhamphus must also remain separate from Ochthoeca, as is recommended in the proposal.”

Comments from Areta on B: I am on the fence as what to do with Tumbezia (I have always been): I have never seen it, but both Mark Pearman privately and now Dan point out that in the field it does not have the behavior or look of an Ochthoeca sensu stricto (to which it is sister). The depth of the divergence of Tumbezia in relation to Ochthoeca does not provide a firm anchoring point to discuss whether it might be better recognized as a separate genus or not. Based on ages of genera across the Tyrannidae, Tumbezia would be among the most recent monotypic genera, and seems to be somewhat younger than most genera in the family. However, one could go either way on this argument. Lanyon (1986) provided his rationale as follows:

Comments from Claramunt: “NO. Although the proposed solution is sound and valid, there is an alternative solution: expand Ochthoeca to include salvini and parvirostris. These two species have been included in Ochthoeca in the past. The resultant expanded Ochthoeca would not be unusually large or heterogeneous. We need large genera to accommodate the astonishing diversity of the Tyrannidae.

“Silvicultrix is diagnosable but mostly based on traits that are not immediately observable: some are internal (syrinx, skull), others are behavioral (nests, songs). Not ideal. In general, differences among general should be more pronounced than differences among species. Silvicultrix, Colorhamphus, and Tumbezia can be recognized at the level of subgenus, instead.”

Additional comments from Stiles: “A-YES to recognize Silvicultrix: genetics confirm Lanyon’s data on morphology B-NO, i.e. Yes to maintaining Tumbezia as a monotypic genus: its separation antedates such universally accepted divisions as that for Empidonax vs. Contopus, among others, and Dan´s information on the biology of this species tips the balance: C-NO to merging Colorhamphus into Ochthoeca, i.e. retain Colorhamphus as a separate genus.”

Comments from Remsen: “As for Santiago’s suggestion of merging all in a single large genus, that is a possibility, but the estimated age of that lineage would be 10+ MY, older than many or most tyrannid genera, and would cause unnecessary instability in tyrannid classification. I don’t think that notions of “too many genera” can be defended – what is more defensible, in my opinion, is using rough divergence times to recognize how many separate lineages are in a particular clade (family in this case) as a way to document its evolutionary diversification relative to other lineages.

“A. YES. Silvicultrix has been evolving as a lineage separate from other group for an estimated 9 MY, which is older than many or most tyrannid lineages treated as genera, and the differentiation in syrinx structure, nest structure, and voice is consistent with that long isolation.

“B. NO. Tumbezia is evidently very distinct in all respects from Ochthoeca other than its syringeal structure, and it has been evolving separately from the latter for an estimated 5+ MY, comparable to lineage ages for most groups ranked as genera in suboscines.

“C. NO. If Tumbezia is retained, then so must we retain Colorhamphus, a lineage evolving separately from the others for an estimated 8 MY.”

Comments from Niels Krabbe (voting for Pacheco):

“A: YES. Do place jelskii, pulchella, frontalis and diadema in Silvicultrix. Silvicultrix is monophyletic, the four species are very similar to each other and differ from the rest in behavior, plumage, voice, and gizz. To me, their domed rather than open nests are sufficient grounds to warrant generic separation.

“B: YES. Do lump Tumbezia with Ochthoeca. Apart from a single netted individual, I have no experience with Tumbezia. For what it is worth, its head seemed to be shaped so much like Ochthoeca that I was not surprised to learn that it is closely related to that genus. That its syrinx differs in no way from the highly derived and otherwise unique Ochthoeca syrinx shows a close relationship. Most Ochthoeca are fairly alike, but, as pointed out by Dan, the cinnamomeiventris group is rather different, in fact in both behavior, shape, coloration, and habitat, and, in my experience, does not duet. Tumbezia is the only low elevation species, true, but I do not find that enough for a generic separation, even if 5 my old.

“C: NO. I have no experience with the species, but the distinctive syrinx described by Lanyon (1986) indicates an isolated position, further cemented by its very deep and basal phylogenetic position.”

Additional. comments from Claramunt: “Regarding Van’s comments, there would be nothing unusual in the age of a comprehensive Ochthoeca. There are several Tyranni genera that are as old or older (crown age):

Ochthoeca (sensu lato): 9.2

Phylloscartes (sensu stricto): 9.3

Pogonotriccus: 9.0

Inezia: 8.6

Myiopagis: 9.1

Ramphotrigon: 9.9

Elaenia: 10.5

Platyrinchus: 14.9

“So, clade age is clearly not a useful argument here. In particular, finding arguments for splitting a genus in the fact that it is “older than many or most tyrannid genera” doesn’t make sense, as it is trivially true (mathematically guaranteed) that 50% of the genera are older than the other half. Even the extreme case of splitting a genus because it is older than all the other genera in the group does not make sense as such a method would result in a runaway towards an atomized taxonomy of monotypic genera. There will always be genera that are older than “most other genera”!

“Regarding stability, since our current classification does not recognize Silvicultrix, stability would be most affected if we change the generic names of four species from Ochthoeca to Silvicultrix, than the alternative, as only two species would be affected (salvini and parvirostris). The use of a broad Ochthoeca would reduce instability by half!”

“Just to clarify my vote, I am pro-lumping here, so my vote would be:

“NO to A

“YES to B: Merge Tumbezia salvini into Ochthoeca

“YES to C: Merge Colorhamphus parvirostris into Ochthoeca

Comments from Bonaccorso:

“A. Yes to changing O. frontalis, O. pulchella, O. diadema, O. jelskii to Silvicultrix. The morphological data of Lanyon (1986) support this monophyletic group for the three first species, and then the molecular data of Fjeldså et al. (2018*; mitochondrial + nuclear), and Harvey et al. (2020) support this group with the inclusion of O. jekskii. I understand Santiago´s point, but for me the Silvicultrix have always been a more homogeneous group (plumage wise) than the Ochthoeca.

“B. YES. It is consistent with all the molecular data available so far, and its plumage differences are within the variation that could be expected with Ochthoeca.

“C. NO. I would maintain Colorhamphus parvirostris because of its molecular and morphological differences with Ochthoeca.

“* Fjeldså, J., Ohlson, J. I., Batalha-Filho, H., Ericson, P. G., & Irestedt, M. (2018). Rapid expansion and diversification into new niche space by fluvicoline flycatchers. Journal of Avian Biology, 49(3), jav-01661.”

Comments from Robbins: “This comes down to philosophy, so either lumping these all together or splitting into multiple genera are both acceptable options (in my opinion). I’ve got field experience with all of these taxa, and one could argue that they (the 4 proposed genera) are not only different, but there are dramatic differences in behavior and vocalizations even within Ochthoeca (sensu stricto), i.e., cinnamomeiventris is distinct from the other Ochthoeca. Santiago Claramunt makes good points in that age isn’t a good yardstick (when compared to other tyrannid genera) and if one wants to minimize instability then everything can be lumped into a single genus.

“So, I’m on the fence, but fine with going with the majority on this one.”

Comments from Zimmer:

“A) YES to recognizing Silvicultrix for frontalis, pulchella, diadema and jelskii.

“B) NO to merging Tumbezia into Ochthoeca. They are obviously closely related, based upon genetic data and morphology, but, drawing upon my own field experience, which corroborates the distinctions highlighted by Dan, I think Tumbezia is enough of an outlier in its behavior and ecology/habitat/elevation to warrant maintaining the monotypic genus.

“C) NO to merging Colorhamphus parvirostris into Ochthoeca. I do not have field experience with Colorhamphus, but, given my position on maintaining Tumbezia, and, given the syringeal distinctions noted by Lanyon, and its deeply rooted basal position in the phylogeny, it follows that we would need to maintain Colorhamphus as well.”