Proposal (874) to South American Classification Committee

Split Arremon flavirostris into two or three species and add English name(s)

The taxonomy of the Saffron-billed Sparrow (Arremon flavirostris) was briefly discussed by Silva (1991), who examined 55 specimens and tentatively suggested the recognition of three species: two green-backed ones, A. flavirostris and A. dorbignii, and a grey-backed one, A. polionotus (including devillii as a synonym). Silva also proposed that the olive-green back color shared by flavirostris and dorbignii was plesiomorphic whereas the short eyebrow length was a synapomorphy that indicated a closer relationship between flavirostris and polionotus. Two recent works have discussed the taxonomy of A. flavirostris and are used as the main sources for this proposal (for more historical information, refer to references therein). Buainain et al. (2016) studied specimens and vocalizations while Trujillo-Arias et al. (2017) focused on potential distributions and phylogeographic analyses. We describe the evidence and offer our view in three subproposals.

A) Split Arremon dorbignii from A. flavirostris

Both studies agree in affording species rank to the Andean green-backed taxon dorbignii. Genetic data show that dorbignii is a fairly recently diverged monophyletic clade (also recovered as a distinct population in 3/4 genes; see Figures 2 and 5 from Trujillo-Arias et al. 2017 below), morphologically it is diagnosed by a narrower pectoral band, a longer supercilium that usually begins in front of the eye and a complex song that does not include repeated notes (see Figures 3 and 5 from Buainain et al. 2016 below). The song differences are remarkable, and our more extensive (unpublished) dataset indicates that structural vocal differences described above uphold with larger sample sizes.

For the sake of completion, we also paste here the only (infinitesimal) piece of evidence that may go against recognition of dorbignii as a full species. Buainain et al. 2016 (p. 558-559) wrote: "One individual (MN11845) from Goiás has a green back and long superciliary stripes, which are diagnostic characters of A. f. dorbignii. It is plausibly either a hybrid specimen between A. f. flavirostris and A. taciturnus, or an aberrant individual. In addition, this individual is the only one with this character, thus the diagnosability of the taxon is still guaranteed by other plumage features, vocalizations and its geographic separation from A. f. dorbignii by more than 1900 km."

Trujillo-Arias et al. (2017) summarized their findings as follows "We suggest to split A. flavirostris into two full species (A. flavirostris and A. dorbignii). The current species is divided in two divergent genetic lineages that are completely isolated: subspecies dorbignii (central Andes rainforests) and the lineage of flavirostris, devillii and polionotus (Atlantic Forest and gallery forests in the Cerrado and Chaco). These two lineages are morphologically diagnosable (Fig. 2a); i.e. dorbignii has longer eyebrow starting at the nape-forehead (not at the upper-eye), a narrower pectoral band (Hellmayr, 1938) and a unique song (Buainain et al., 2016). These two lineages could be considered full species according to different species concepts. For example, according to the phylogenetic concept (Cracraft, 1983), they are full species because represent monophyletic populations (Fig. 5b). Also, according to the general lineage species concept (de Queiroz, 1998), they represent independent evolutionary lineages (i.e., migration between them M < 1) that are morphologically diagnosable. Finally, it is not clear if both populations could be separated according to the biological species concept. Even though there is very low gene flow between the Andean and the Atlantic Forest populations, indicating no interbreeding, and they are phenotypically diagnosable (plumage and song), it is impossible to show that both populations would not hybridize if their geographic ranges met (but see Buainain et al., 2016)."

We recommend a YES vote to A. Differences in genetic, plumage and vocal data suggest that it is a good species by any criteria.

B) Assign English name to Arremon dorbignii

We propose to use Moss-backed Sparrow for A. dorbignii. It is a memorable name, describes its rich green dorsal coloration, and evokes the Andean cloud-forests in which it is found.

We recommend a YES vote on B. The distribution of A. dorbignii is small in relation to A. flavirostris, and we do not recommend a name change for the latter.

C) Split Arremon polionotus from flavirostris

T

he studies differed in this regard: Buainain et al. (2016) proposed splitting polionotus as a separate species, while Trujillo-Arias et al. (2017) proposed to keep it within flavirostris. In plumage, devillii + polionotus differ from flavirostris in their grey backs, but are otherwise structurally identical, while vocally, devilli + polionotus are very similar (or identical) to flavirostris as was described above in subproposal B. Buainain et al. (2016) mention the existence of greenish feathers on the back of the grey-backed taxa as an indication of female sex and as a feature of juvenile/immature birds, proposing that these are not intermediates (hybrids?) between grey-backed birds and nominate flavirostris (for males they mention that "the hypothesis that this is a remnant of immature plumage and not hybrids is much more parsimonious"). So, this was left as an open question while clearly indicating their preferred interpretation. However, in part answering this question, some individuals of devilli were recovered as more closely related to some flavirostris than to polionotus in phylogenetic trees and in COI population structure analyses (see Figs. 2 and 5 from Trujillo-Arias et al. 2017 below) and the genetic divergence in the group comprising devillii, polionotus and flavirostris is shallow. Finally, the three taxa deliver a short introduction followed by a repetitive series of piercing notes (see Figures 3 and 5 from Buainain et al. 2016 below).

Again, Trujillo-Arias et al. (2017) summarized their findings as follows "Regarding the other subspecies (flavirostris, devillii and polionotus), our analysis did not support them as independent evolutionary entities (Figs. 2 and 5b) (but see Buainain et al., 2016). These subspecies are similar morphologically, with the only difference between them the color of their back plumage (e.g. green in flavirostris; grey in devillii and polionotus) (Hellmayr, 1938; Silva, 1991). However, since they are not genetically divergent (Fig. 5b), we do not suggest considering them as full species. "

We recommend a NO vote on C. Lack of vocal differences, meager genetic differentiation, lack of monophyly, and existence of allegedly intermediate looking birds do not support the recognition of polionotus as a separate species.

Figures from Buainain et al. (2016)

Figures from Trujillo-Arias et al. (2017)

References

Buainain, N. R., Brito, G. R., Firme, D. F., Figueira, D. M., Raposo, M. A. & Assis, C. P. (2016) Taxonomic revision of Saffron- billed Sparrow Arremon flavirostris Swainson, 1838 (Aves: Passerellidae) with comments on its holotype and type locality. Zootaxa 4178: 547–567.

Silva, J.M.C. (1991) Geographical variation in the Saffron-billed Sparrow Arremon flavirostris. Bulletin of the British Ornithologists’ Club 111: 152–155.

Trujillo-Arias, N., Dantas, G. P. M., Arbeláez-Cortés, E., Naoki, K., Gómez, M. I., Santos, F. R., Miyaki, C. Y., Aleixo, A., Tubaro, P. L. & Cabanne, G. S. (2017) The niche and phylogeography of a passerine reveal the history of biological diversification between the Andean and the Atlantic forests. Molecular Phylogenetics and Evolution 112: 107–121.

Juan I. Areta and Mark Pearman, August 2020

Comments from Stiles: “YES to A: split dorbignyi from flavirostris and B: for E-names, Moss-backed is fine for the former, retaining Yellow-billed for flavirostris. C: NO evidence for the further split of polionotus from flavirostris is insufficient.”

Comments from Zimmer: “A) YES” to splitting dorbignyi from flavirostris based upon genetic, plumage and vocal data. (B) YES to using “Moss-backed Sparrow” for dorbignyi – this name is descriptive, novel, and evocative of the Andean cloud forests that the species occupies. (C) NO to splitting gray-backed polionotus + devilli from green-backed flavirostris, given that the two “groups” differ only in back color, but have very similar vocalizations, a broad pectoral band (relative to that of dorbignyi), a supercilium that starts above the eye, not at the forehead, and a lack of genetic divergence between them.”

Comments from Pacheco: “A – YES. The splitting of dorbignyi from flavirostris is well corroborated. C – NO. There is no evidence to support this change, as recommended by Nacho and Mark.”

Comments from Bonaccorso: “A. YES. Phylogenetic and vocal differences are solid. Geographically and ecologically, it also makes a lot of sense. I am not so comfortable with the plumage differences since it seems that the back plumage color is quite labile in this group, and differences in eyebrow are subtle. C. NO. Arremon flavirostris polionotus is not monophyletic, and both clades that contain it also contain A. f. devilli, which may indicate ongoing hybridization; the suspicion of intermediate plumages supports this possibility. If anything, the logic proposal from the plumage point of view would be separating A. polionotus + devillei from A. flavirostris, if faster markers recover them as monophyletic in future studies.”

Comments from Jaramillo:

“A – Yes, the genetic, vocal and biogeographic data conforms to plumage details that confirm a species level separation here in my mind.

“B – Moss-backed is a great name. I would maintain Saffron-billed for the more widespread taxon.

“C – No, although back coloration is quite different, there seems to be little else here.”

Comments from Schulenberg: “B. YES to 'Moss-backed Sparrow’, fine name. in this case it's probably safe to retain Saffron-billed Sparrow for the (slightly reduced) new Arremon flavirostris concept, so sure, yes to this as well.”