Proposal (610) to South American Classification Committee

Split Sirystes into two (A) or four (B) species

Background

The cross-Andean split of Western Sirystes S. albogriseus from Eastern Sirystes S. sibilator is widely recognised and supported by long-known and striking differences in vocalizations. Donegan’s (2013) study of vocal variation in the genus supported this split but also a further three-way split of eastern populations.

Part A: Split Western Sirystes S. albogriseus from Eastern Sirystes S. sibilator

Ridgely & Greenfield (2001), followed by Jahn et al. (2002), Hilty (2003), Gill & Wright (2006), Ridgely & Tudor (2009), McMullan & Navarrete (2013) and others treat S. albogriseus ("Western Sirystes") as a species separate from a broadly defined S. sibilator including all remaining taxa ("Eastern Sirystes"), based mainly on differences in vocalisations discussed in Ridgely & Tudor (1994).

Part A would change AOU-SACC treatment to reflect widespread existing treatments in the ornithological literature and can be considered independently of the merits of the recent publication cited below.

Western Sirystes gives only short chips whilst Eastern Sirystes vocalisations involve longer notes and far more complex.

Western Sirystes: http://www.xeno-canto.org/species/Sirystes-albogriseus?view=3

Eastern Sirystes: http://www.xeno-canto.org/species/Sirystes-sibilator?view=3

I know not of two more vocally different but lumped taxa on the SACC list. This proposal is long overdue for consideration and its acceptance should be strongly recommended.

Part B: Split "Eastern Sirystes" into three species

In Donegan (2013), voice of Sirystes throughout its range was studied in detail, as well as some specimens and other sources of records. The abstract includes the following:

"Vocal differentiation in the genus was studied using quantitative methods and the determination of primary, secondary and other kinds of vocalizations for different populations. The widely recognized split of Western Sirystes S. albogriseus of the Chocó (Panama, Colombia and Ecuador) from the more widespread Eastern Sirystes S. sibilator is strongly supported by differences in vocal repertoire and quantitative vocal differentiation. Eastern Sirystes is itself a superspecies comprised of three allopatric species which are vocally differentiated (quantitatively and in repertoire) to such an extent that species rank would be afforded under a modern biological species concept for: (i) White-rumped Sirystes S. albocinereus of Western Amazonia (in Venezuela, Colombia, Ecuador, Peru, Bolivia and western Amazonas and Acre states, Brazil); (ii) Todd's Sirystes S. subcanescens of North Amazonia (in northern Amazonas, northern Pará and Amapá states in Brazil, French Guiana and Guyana); and (iii) Sibilant Sirystes S. sibilator (including subspecies atimastus) of the Atlantic and Cerrado regions (in southern Pará state South through most of the rest of Brazil to northern Argentina and eastern Paraguay). Considerable variation in the kind of primary and secondary vocalizations are evident between these four proposed species, with primary vocalizations of particular taxa appearing as rare or secondary vocalizations or not being given at all by others. The four proposed species also show considerable morphological differences in the context of Tyrannidae."

Vocal differentiation between the Eastern races is further detailed in the paper as follows. See the discussion, appendices and large number of sonograms for more information.

· The most geographically distant pair: Atlantic-Cerrado versus Western Amazonian populations, differ: (i) in the kind of primary vocalization, which in the Atlantic-Cerrado are chattering songs and in Western Amazonia are whistles (App. 4); (ii) diagnosably in the song structure and note shape of whistles, which in Western Amazonia are always followed by one or more short notes, a chattering song or downstroke song but in the Atlantic-Cerrado population, to the extent fragments of downstroke songs can be considered the equivalent of a whistle, are delivered alone or in sequences of such notes and have a less symmetrical and more sharply peaked note shape (Figs. 1-2); (iii) to the extent fragments of downstroke songs can be considered the equivalent of a whistle, diagnosably and significantly in the maximum acoustic frequency of whistles (App. 3; Fig. 4A); and (iv) significantly but not diagnosably in other vocal variables set out in Appendix 3.

· North Amazonian and Atlantic-Cerrado populations differ: (i) in the kind of primary vocalization, which North Amazonia are whistles and in the Atlantic-Cerrado are chattering songs (App. 4); and (ii) in the kind of secondary vocalization, which in North Amazonia are chattering songs and in the Atlantic-Cerrado are downstroke songs (App. 4); (iii) diagnosably and significantly in the length of whistles (App. 3; Fig. 4A); (iv) provisionally (based on a single recording from North Amazonia), in the note shape of downstroke songs, with the North Amazonia recording having an initial upstroke which is slower in reaching the peak; and (v) significantly but not diagnosably in other variables set out in Appendix 3.

· Western Amazonian and North Amazonian populations differ: (i) diagnosably in the song structure of whistles, which are the primary vocalization for both populations, which in Western Amazonia are always followed by one or more short notes, a chattering song or downstroke song but in North Amazonia are delivered alone or in sequences of whistles; (ii) significantly in maximum acoustic frequency of whistles (App. 3; Fig. 4A), which narrowly missed the statistical test of diagnosability perhaps due to sample sizes from North Amazonia, but showed no overlap; (iii) in secondary vocalization type (which is the chattering song in North Amazonia versus downstroke song in West Amazonia) (App. 4); (iv) provisionally (based on a single recording from North Amazonia), in the note shape of downstroke songs, with the North Amazonia recording having an initial upstroke which is slower in reaching the peak; and (v) significantly but not diagnosably in other variables set out in Appendix 3.

Distributions and country lists

Most textbooks show a widespread range for Sirystes, but as noted in Donegan (2013) there is almost no evidence for the species occurring in southern Amazonia or much of western Amazonia, whilst all other parts of the genus' distribution are corroborated by multiple specimen, sound recording and trip report based/reserve list records. (See map below). Populations appear restricted to particular habitats/biogeographical regions (the Choco, western Amazonia (both sides of the Amazon river), North Amazonia and Atlantic & Cerrado) rather than to interfluves. The only population limited by a river is Todd's Sirystes in the southern part of its range, which is limited by the lower Amazon River (in its widest part).

Vernacular names

The vernacular names would be those set out in the Abstract quoted above, which are those of Hellmayr (1927) except that for "Western Sirystes", which is Ridgely & Greenfield (2001)'s preference over Ridgway (1907) and Hellmayr (1927)'s "Panama Sirystes". If Part B passes, then the "Western/Eastern" nomenclature becomes less attractive and "Choco Sirystes" could be used instead of "Western" for albogriseus. Given what happened to the Schiffornis, compass direction-based names seem better avoided. I will do a separate proposal on the three 'available' English names for albogriseus in the event that Part B passes.

References

Donegan, T.M. 2013b. Vocal variation and species limits in the genus Sirystes (Tyrannidae). Conservación Colombiana 19: 11-30.

The paper can be accessed from this link:

http://www.proaves.org/birds-of-colombia-2013/?lang=en

Other references are cited in the above.

Thomas Donegan, December 2013

Comments from Remsen: “YES to both A and B. Donegan’s published analyses shows that multiple species should be recognized.”

Comments from Stiles: “A definite YES to part A: this split is well documented and widely accepted. For part B, given their vocal distinctiveness and relatively less extreme plumage differences as well as the convincing analogy with the closely related Myiarchus, I think that the burden of proof has shifted towards those who would treat all three as a single species, so YES.”

Comments from Pacheco: “Yes (A) due to of reasons given by Donegan's paper, in accordance with the existing literature data. A Yes (B) also based on that article and my own experience.”

Comments from Nores: “A: YES. B: NO. I repeat here what I put in proposal 49: “Separar especies por suaves diferencias en el canto no me parece bien. Recientemente estuve en el noreste de Brasil y me llamó la atención lo diferente que son los cantos de algunas subespecies de allí con respecto a las poblaciones del sur de Sudamérica. Por ejemplo, Thraupis sayaca tiene un canto mucho mas potente y mas variado que las razas del sur y Turdus rufiventris emite un llamado permanente que nunca se lo escuché a la subespecie de esta latitud. Otro notable ejemplo es Vanellus chilensis, del cual la raza del sur de Argentina y Chile emiten un canto bastante diferente (parece un loro) que la raza que habita el norte y centro de Argentina hasta Amazonas. Esto no significa para mi que haya que elevar las subespecies a especies."

Comments from Zimmer: “YES to both A and B. The vocal distinctions between the various populations of Sirystes have long been obvious to those with the required geographic breadth of field experience, and Donegan has done a good job of detailing that in his paper. He has also done a good job of highlighting the apparent patchy distribution of Sirystes as a whole, something that squares with my experience, but which flies counter to most published range descriptions. These are suboscine birds (unlike the examples of Thraupis and Turdus mentioned by Manuel), and I see no reason not to treat their diagnosably different vocalizations as hard-wired, genetically based characters that are at least on a par with any plumage characters as being reflective of species-level relationships (or lack thereof).”

Comments from Robbins: “YES, to recognizing four Sirystes species. I did not appreciate that the distribution of the cis-Andean taxa was far more restricted than what the literature has indicated. Differentiation in plumage and vocalizations are indeed analogous to Myiarchus species.”