Proposal (753) to South American Classification Committee:

Treat Poospiza whitii as a separate species from P. nigrorufa

Effect on South American CL: This proposal would add Poospiza whitii to the SACC list, by splitting it from Poospiza nigrorufa.

Background: Three subspecies are currently recognized in this complex: nigrorufa, whitii, and wagneri, the latter of which has been considered doubtfully distinct from whitii. Both nigrorufa and whitii were historically treated as full species until Hellmayr (1938) lumped them by treating whitii as the western subspecies of P. nigrorufa. Thereafter, nigrorufa and whitii were treated as a single geographically variable species but by some authors as full species; see Jordan et al. (2017) for details.

Genetic data indicate that nigrorufa and whitii are well differentiated sister taxa with levels of mtDNA divergence (ca. 2.5% for cyt-b and ND2) similar to those of other Poospizinae sister species (Shultz and Burns 2013).

SACC proposal 79 failed to pass due to lack of detailed published analyses, especially on vocal differences.

New information: Jordan et al. (2017) showed that nigrorufa and whitii differ in 1) plumage coloration and degree of dimorphism, 2) morphometric traits, 3) habitat, distribution and ecological niche models, 4) vocal characters, and that 5) in reciprocal playback experiments they ignore the other taxon while answering strongly to their own vocalizations.

1) Plumage: In the slightly dimorphic nigrorufa, males have tawny rufous throat, breast and flanks, and are brownish grey with slate tinged upperparts (crown, neck, back and rump), while females differ in the orange tinge of ventral parts and in the more olivaceous upperparts. Sexes of nigrorufa are hard to distinguish both in field and museum specimens. On the contrary, in the markedly dimorphic whitii, males have dark chestnut throat, breast and flanks, and slate upperparts, while females exhibit tawny pale orange throat, breast and flanks, and olivaceous light-brown upper parts (Fig. 1). In the field, females of whitii exhibit paler ventral colors, and more olivaceous upperparts than both sexes of nigrorufa. The key to correct identification seems to be the extent of the white tip of the tail, which is much greater in whitii than in any sex of nigrorufa.

Jordan et al. (2017) also proposed that the subspecies wagneri be treated as a synonym of whitii because they were unable to find any consistent plumage differences.

2) Morphometrics: Specimens of nigrorufa had significantly higher and longer bill, longer tarsus and wings, and were~10% heavier than whitii (n=106 nigrorufa and n=91whitii). Both species were sexually dimorphic, with males exhibiting longer wings and tails than females. Within-sex comparisons between species show that nigrorufa has longer wings, but not longer tails, than whitii (Fig. 2).

3) Habitat, distribution and ecological niche models: Locality records (781 for nigrorufa and 322 for whitii) and models of potential distribution indicate that ranges of both species are narrowly allopatric, approaching closely in central Córdoba province (Argentina) without overlapping (Fig. 3). P. nigrorufa inhabits shrubby open areas in wetlands with reeds (Typha, Schoenioplectus) and bulrushes (Scirpus, Rhynchospora) and grassy plains with Pampas Grass (Cortaderia selloana), whereas P. whitii inhabits closed to semi-closed xerophytic to semi-humid scrub (Prosopis, Geoffroea) and woodlands (Podocarpus, Alnus) far from wetlands. Finally, P. nigrorufa is generally a lowland species (except in southern Brazil), whereas P. whitii is a montane bird (Fig. 3).

4) Vocalizations: Songs (108 individuals; nigrorufa n= 81, whitii n= 27) and calls (18 individuals; nigrorufa n=14, whitii n= 4) differed dramatically (Fig. 4). The simple song of nigrorufa consists of one phrase with three pure whistled notes, that is repeated a variable number of times. This phrase is usually transliterated as pleased to meet you in English, quem te vestiu in Portuguese, or bichi-bichi in Spanish. The general shape and order of the three notes of the song of nigrorufa are very consistent throughout the range of the species.

The complex song of whitii is composed of a succession of a variable number of notes whose quality and order varies greatly from individual to individual. In stark contrast to nigrorufa, the number of notes performed by each analyzed individual of whitii varied between 8 and 12. A complete song is a rhythmic series of single melodious notes, among which paired melodious notes and a few trills are irregularly interspersed that might be represented as choo we, tip-tip, sweet peer, tweak, trrree, sweet peer.

The numerous calls of both species are easily distinguished with the aid of spectrograms, but are virtually impossible to distinguish in the field (Fig 4.).

5) Reciprocal playback experiments: 20 reciprocal sandwich-playback experiments were performed in Argentina during the breeding season. In all experiments, the ten males of nigrorufa and the ten males of whitii responded aggressively to conspecific vocalizations by approaching to the sound source and singing, while ignoring heterospecific ones, regardless of stimulus order (n=15 conspecific and 15 heterospecific stimuli for each species).

Recommendation: We recommend a YES vote. All lines of evidence clearly show that nigrorufa and whitii belong to different species under any species concept. This long overdue split is now fully justified with solid integrative evidence, including behavioral responses to mating cues.

Fig.1. A: P. nigrorufa male (Ph: Sebastián Preisz) B: P. nigrorufa female (Ph: D. Lins). C: P. whitii male (Ph: G. Núñez), D: P. whitii female (J. La Grotteria).

Fig. 2. Morphological measurements and weight of Poospiza nigrorufa and Poospiza whitii showing mean and standard deviation. Asterisks denote significant differences with alpha=0.05

Fig. 3. Presence localities and potential distribution of Black-and-rufous Warbling-Finch (Poospiza nigrorufa) and Black-and-chestnut Warbling-Finch (Poospiza whitii).

Fig. 4. Spectrograms depicting songs (A, B, D, E) and calls (C, F) of adult males of Black-and-rufous Warbling-Finch (Poospiza nigrorufa) and Black-and-chestnut Warbling-Finch (Poospiza whitii). See Jordan et al. (2017) for further explanation of symbols.

Literature cited

Hellmayr (1938) Catalogue of birds of the Americas and adjacent islands. Field Museum of Natural History Publications in Zoology 11, 1–662.

Shultz AJ & Burns KJ (2013) Plumage evolution in relation to light environment in a novel clade of Neotropical tanagers. Molecular Phylogenetics and Evolution 66, 112–125. doi:10.1016/j.ympev.2012.09.011

Jordan EA, Areta JI & Holzmann I (2017) Mate recognition systems and species limits in a warbling-finch complex (Poospiza nigrorufa/whitii). Emu https://doi.org/10.1080/01584197.2017.1360746

Emilio A. Jordan and Juan I. Areta

_________________________________________________________________________________________________________________________

Comments from Stiles: "YES. The combination of genetic, morphological and ecological data clearly shift the burden of proof to those favoring conspecifity of whitii."

Comments from Zimmer: “YES. Multiple data sets confirm the distinctiveness of these two taxa, and the playback experiments confirm the importance of the vocal differences, which were noted by Ridgely and Tudor (citing R. Straneck) as far back as 1989.”

Comments from Remsen: “YES. The playback trials in particular are convincing evidence that these taxa have diverged to the point that free gene flow no longer likely.”

“If this passes, then I suppose we should use the names in Ridgely & Tudor that are already fairly entrenched and included in the proposal above. However, this is one of those cases in which the usual policy of creating new names for both daughters should have received serious consideration.”

Comments from Claramunt: “YES. The evidence is solid, in my opinion.”

Comments from Cadena: “YES. The work by Jordan et al. is an an excellent, integrative taxonomic study. I do not think that morphometric analyses revealing differences in measures of central tendency are very useful for species delimitation (especially for allopatric populations) and I think that ecological differences as revealed by niche models are especially revealing in cases unlike the present one where potential distributions suggest there is potential for populations to come in geographic contact yet they remain distinct (we described this with Andrés Cuervo in a paper in 2010 but the idea has not gained much traction). Nonetheless, all the evidence points in the same direction that there are two different species here.”

Comments from Pacheco: “YES. The evidence gathered is convincingly strong.”

Comments from Robbins: “YES, for recognizing Poospiza whitii as a species based on multiple data sets.”