Proposal (885) to South American Classification Committee

Revise the generic classification of the Xolmiini

Two recent studies proposed alternative generic classifications for the Xolmiini: Ohlson et al 2020 and Chesser et al 2020. Their proposals were largely coincident but differed in how broad the genus Nengetus should be. Our subproposals here broadly coincide with these two papers but differ in the treatment of Nengetus.

We recommend a YES vote to all of the subproposals.

A) Include pyrope in Pyrope (YES) or do something else (NO)

The highly distinctive Patagonian forest pyrope was found as stemming from a deep node in both studies, supporting the recognition of the old genus Pyrope, of which it is the type species.

B) Include rufipennis in Cnemarchus (YES) or leave in Polioxolmis (NO)

Both species in this clade are rather elongated, upright-perching, high-altitude specialists that deliver shrill whistles and share the same tail pattern. We favor merging them in a single genus, which would emphasize the common theme between them in plumage, habitat associations and vocalizations. This treatment has been followed in the past (see table from Chesser et al 2020).

C) Restrict Xolmis to irupero and velata (YES) or do something else (NO)

This is a rather straightforward decision, as both species formed a coherent clade in both studies and the type species of Xolmis is irupero.

D) Use Myiotheretes for 4 species (YES) or merge Myiotheretes under Nengetus (NO).

Myiotheretes is a very homogeneous genus comprising four Andean species of reddish/brownish birds that deliver simple monotone flat whistles and inhabit the tree-line zone. The continued recognition of Myiotheretes implies stability and is consistent with the recovery of this clade in both studies.

Ohlson et al. (2020) proposed to merge Myiotheretes in an expanded (and heterogeneous) Nengetus, stemming from the placement of cinereus in their phylogeny. However, this merger is not consistent with the position of cinereus in the work by Chesser et al. (2020) and would create a large and very heterogeneous grouping of birds. The best way out of this is to recognize Myiotheretes, which would imply no change in our classification.

E) Use Nengetus only for cinereus and Neoxolmis for coronatus, rubetra, salinarum, and rufiventris (YES) or use Nengetus for all the taxa (NO)

As in the case of Myiotheretes, the rather obscure genus Nengetus is again at the center of conflict. Both studies advocate a rather broad use for Nengetus, but we believe this is not good for two reasons 1) it creates too heterogeneous groupings and 2) it disrupts stability.

Reasons for restricting usage of Nengetus for cinereus

The type of Nengetus is cinereus, a widespread species with distinctive plumage that lives in open savanna areas. This distinctiveness is also evident in the deep genetic divergence with other species in the clade. Besides these aspects, cinereus does not walk on the ground regularly or at all (unlike the other species usually placed in Neoxolmis) and inhabits more humid and generally lowland areas (reaching higher altitude in southern Brazil, a pattern exhibited by many "Pampas" birds), whereas the other species are all exclusive to arid areas mostly in the extreme southern portion of the continent. Finally, the position of cinereus might be open to discussion (see contrasting results in Ohlson et al 2020 and Chesser et al 2020 discussed also in Subproposal D); thus, we do not recommend expanding Nengetus any further beyond its type species, cinereus.

Reasons for placing four species in Neoxolmis

The four species that would be included in Neoxolmis (rufiventris, rubetra, salinarum and coronatus) are all species of dry, extreme areas that regularly walk on the ground (less often in coronatus). Their nests are made of sticks on the ground or in shrubs are remarkably similar, as are their eggs. They form a monophyletic group in both the Ohlson et al. (2020) and Chesser et al. (2020) studies. Traylor (1979) and Lanyon (1986) already recognized a close relationship of rufiventris+rubetra/salinarum when placing them together in the genus Neoxolmis (see table from Chesser et al 2020 above). In sum, they form a far more coherent grouping when cinereus is excluded. Moreover, using Neoxolmis has the advantage of using a name that has already been in wide use and which recognizes that they are in some way akin to "Xolmis". Having direct field experience with all the Xolmiini species, and with 3/4 species being endemic breeders to Argentina, we are convinced that using Neoxolmis is the best possible solution, which is consistent with genetic evidence, supported by field data and which best serves stability.

We think that, based on the deep genetic divergence and plumage differences, the Argentine endemic breeder coronatus may merit a new genus. However, no such genus name is available and at present it seems easier and less disrupting to place it in Neoxolmis together with the other species in this clade of austral birds.

F) Recognize Syrtidicola for fluviatilis (YES) or keep in Muscisaxicola (NO)

A deep divergence time, possible sister relationship to Satrapa and morphological and ecological differences (e.g., very short tail, lowland inhabitant of riverine sand bars) lead Chesser et al. (2020) to describe a new genus for fluviatilis.

G) Sequence. The sequence we need to adopt would be this one (YES) or another one (NO):

Muscisaxicola fluviatilis Little Ground-Tyrant --- Syrtidicola

Muscisaxicola maculirostris Spot-billed Ground-Tyrant

Muscisaxicola griseus Taczanowski's Ground-Tyrant

Muscisaxicola juninensis Puna Ground-Tyrant

Muscisaxicola cinereus Cinereous Ground-Tyrant

Muscisaxicola albifrons White-fronted Ground-Tyrant

Muscisaxicola flavinucha Ochre-naped Ground-Tyrant

Muscisaxicola rufivertex Rufous-naped Ground-Tyrant

Muscisaxicola maclovianus Dark-faced Ground-Tyrant

Muscisaxicola albilora White-browed Ground-Tyrant

Muscisaxicola alpinus Plain-capped Ground-Tyrant

Muscisaxicola capistratus Cinnamon-bellied Ground-Tyrant

Muscisaxicola frontalis Black-fronted Ground-Tyrant

Cnemarchus erythropygius Red-rumped Bush-Tyrant

Polioxolmis rufipennis Rufous-webbed Bush-Tyrant --- Cnemarchus

Xolmis pyrope Fire-eyed Diucon --- Pyrope

Xolmis velatus White-rumped Monjita

Xolmis irupero White Monjita

Xolmis cinereus Gray Monjita --- Nengetus

Xolmis coronatus Black-crowned Monjita --- Neoxolmis

Xolmis salinarum Salinas Monjita --- Neoxolmis

Xolmis rubetra Rusty-backed Monjita --- Neoxolmis

Neoxolmis rufiventris Chocolate-vented Tyrant ---Neoxolmis

Agriornis montanus Black-billed Shrike-Tyrant

Agriornis albicauda White-tailed Shrike-Tyrant

Agriornis lividus Great Shrike-Tyrant

Agriornis micropterus Gray-bellied Shrike-Tyrant

Agriornis murinus Lesser Shrike-Tyrant

Myiotheretes striaticollis Streak-throated Bush-Tyrant

Myiotheretes pernix Santa Marta Bush-Tyrant

Myiotheretes fumigatus Smoky Bush-Tyrant

Myiotheretes fuscorufus Rufous-bellied Bush-Tyrant

H) Recognize Heteroxolmis. The Black-and-white Monjita was found to not even belong in the Xolmiini, so a return to the genus Heteroxolmis is warranted. It was found to be more closely related to other "warm" grassland specialists (Alectrurus spp. and Gubernetes yetapa) in the Fluvicolini. Ohlson et al 2020 stated:

"Recognize Heteroxolmis Lanyon, 1986 (type = Tyrannus dominicanus (Vieillot) for Xolmis dominicanus and remove it from Xolmini to Fluvicolini. The distinctiveness of dominicanus from other Xolmis species was recognized by Lanyon (1986) based on morphological characters of the nasal capsule and syrinx, but he regarded the two genera as close relatives. Among the character states that motivated a separation from Xolmis is a fully ossified nasal capsule, including alinasal walls and turbinals, which is also found in Alectrurus Vieillot, Gubernetes Such, Fluvicola Swainson and Arundinicola d’Orbigny (Lanyon 1986). As the gender of Heteroxolmis is feminine (see Lanyon 1986, fig. 24), the name of the species becomes Heteroxolmis dominicana.”

Xolmis dominicanus Black-and-white Monjita --- Heteroxolmis

A YES vote would resolve Xolmis dominicanus as Heteroxolmis dominicana and place it in the sequence between Alectrurus and Gubernetes.

Nacho Areta & Mark Pearman, September 2020

References

Chesser, R.T., M. G. Harvey, R. T. Brumfield and E.P. Derryberry (2020) A revised classification of the Xolmiini (Aves: Tyrannidae: Fluvicolinae), including a new genus for Muscisaxicola fluviatilis. Proceedings of the Biological Society of Washington 133:35–48

Ohlson J.I., M. Irestedt, H. Batalha Filho, P.G.P. Ericson and J. Fjeldså (2020) A revised classification of the fluvicoline tyrant flycatchers (Passeriformes, Tyrannidae, Fluvicolinae). Zootaxa 4747: 167–176

Tree from Ohlson et al 2020

Tree from Chesser et al 2020

Table from Chesser et al 2020

Comments from Robbins: “YES to all proposals except E. With regard to E, looking at the genetic data it does not make sense to break up this clade by separating cinereus from the others. If one was going to apply a consistent treatment (note the branch length of coronatus is similar to cinereus), one either treats these as three different genera or all the same. The latter seems more reasonable to me.”

Comments from Stiles: “YES to all. When behavior, ecology and biogeography are combined, I agree with Kevin that cinereus is best separated in a monotypic Nengetus.”

Comments from Lane: “A) YES. B) NO. These two species don't strike me as being all that similar in their behaviors, particularly the Kestrel-like hovering that is so characteristic of Polioxolmis. I would prefer to keep them as two monotypic genera. C) YES. D) YES. E) YES. F) YES (emphatically!). G) YES. H) YES.”

Comments from Claramunt: “YES to all.

“After going back and forth on E, I ended up liking the proposed solution: a monotypic Nengetus plus a diverse Neoxolmis. Morphology, plumage and behavior are not of much use here as there is no satisfactory way of subdividing the entire clade based on these. Excluding cinerea doesn’t make the remaining Neoxolmis a nice homogeneous bunch. And even separating coronata does not solve the issue, as rufiventris is fairly different from rubetra and salinarum. What persuaded me at the end was the argument of nomenclatural stability. This is because: 1) the resurrection of the obscure Nengetus would affect only a single species (cinerea); 2) At least rufiventris, and rubetra and salinarum according to some classifications, would retain their traditional generic names; 3) The nomenclature would not change even if cinerea is found to be closer to Myiotheretes (as in Ohlson et al.). So, at the end, stability prevailed over my dislike for monotypic genera.”

Comments from Jaramillo:

“A. YES – The loss of Pyrope never quite sat well with me; this new dataset correlates with the fact that this is quite a different bird from typical Xolmis.

B. No – I agree with Dan, that they should be left as monotypic genera.

C. YES – is it velata or velatus though?

D. YES – retain Myiotheretes for the 4 species, retaining a well-defined and morphologically cohesive genus.

E. YES – I hesitated, thinking that coronatus perhaps deserved to be separate. However, it is already in Xolmis with various of the other species that will now go to Neoxolmis and I was OK with that. Definitely cinereus is the odd one out here, a rather shrike like species, quite unlike the rest.

F. YES – and note that I would be amenable to a separate genus for maculirostris as well. Not only due to the different morphology of that bird, but a distinct voice, unlike the very vocally simple Muscisaxicola (in the strict sense)

G. YES – adopt the new sequence.

H. YES – rather unexpected in some ways. It was always different, but I would have expected it was a relative to true Xolmis as we knew it. The membership in the grassland tyrants makes a certain amount of logic as well.”

Comments from Pacheco: “YES to all proposals except E, for consistency with the adoption of expanded Nengetus implemented by CBRO.”

Comments from Remsen: “YES to all, except that using the Chesser et al. tree, Xolmis must precede Pyrope following standard conventions.”