Proposal (124) to South American Classification Committee
Split Notharchus swainsoni from N. macrorhynchos
Effect on South American CL: This proposal would elevate a taxon to species rank that we currently treat as a subspecies on our baseline list.
Background: The subspecies swainsoni of the Atlantic forest region of southeast Brazil, eastern Paraguay, and northeastern Argentina was described as a separate species (Gray 1846) from the wide-ranging Notharchus macrorhynchos (White-necked Puffbird), a course that was followed by Cory (1919). Peters (1948) treated the two as conspecific, as have most subsequent authors. Rasmussen and Collar (2002), noting the distinctiveness of swainsoni, the highly disjunct nature of its range with respect to the nearest populations of macrorhynchos, and, citing Sibley (1996) regarding alleged vocal differences between the forms, went ahead and elevated swainsoni to species status, giving it the English name of Buff-bellied Puffbird. Alvarenga et al. (2002) provided support for that treatment, based on differences in several plumage, biometric and osteological characters.
Rasmussen and Collar (2002) treat swainsoni as a monotypic species. The same authors recognize three subspecies of N. macrorhynchos as follows:
N. m. hyperrhynchus (Sclater 1856) - S Mexico south to N & NE Venezuela, and south to Colombia, Ecuador, E Peru, N Bolivia and W Brazil (E to Rio Tapajós and S to Mato Grosso).
N. m. macrorhynchos (Gmelin 1788) - extreme E Venezuela, the Guianas, and extreme N Brazil south to the Amazon.
N. m. paraensis (Sassi 1932) - lower Amazon Valley in Brazil (Pará east of the lower Rio Tapajós and into N Maranhão).
The subspecies swainsoni is readily distinguished from all other subspecies of macrorhynchos by its distinctly buff-colored belly and vent (these areas white in all other subspecies of macrorhynchos). Further distinguishing plumage characters include gray (not black) barring on the flanks; lower face, sides of neck and hind collar washed with gray (these areas unmarked white in other subspecies of macrorhynchos); upperparts black with greenish gloss (black with blue gloss in other subspecies of macrorhynchos); and narrower black breast band.
Alvarenga et al. (2002) examined 58 specimens of swainsoni and 9 specimens of macrorhynchos (2 of nominate, 3 of hyperrhynchus, 4 of paraensis); and compared 4 skeletons of swainsoni with 3 skeletons of macrorhynchos (2 of hyperrhynchus and 1 of paraensis). They found swainsoni to be consistently smaller in size, with a proportionately smaller head and bill. N. macrorhynchos varied from 260 to 270 mm in total length, whereas swainsoni varied from 230 to 235 mm (about 13% smaller); wing chord in macrorhynchos varied from 114 to 120 mm, versus 103 to 106 in swainsoni (about 11% smaller); whereas bill length varied from 42-50 mm in macrorhynchos and from 32 to 36 mm in swainsoni (about 26% smaller). These authors give weights for 1 specimen of N. m. paraensis as 101 g; cite Burton (1973) for weights of 109 and 103 g for two specimens of N. m. hyperrhynchus from Panama; cite Hartman (1961) for a mean of 97 g for 5 females and 95.4 g for 12 males of N. m. hyperrhynchus; and cite Storrs Olson for a mean of 91.3 g for 6 specimens of nominate macrorhynchos from Guiana. Alvarenga et al. provided weights of three specimens of swainsoni as 72, 75.5, and 75.6 g respectively. These authors point out that the markedly smaller size of swainsoni contradicts Bergmann's rule concerning size variation with increasing latitude within a species.
Alvarenga et al. (2002) also point out several osteological differences between swainsoni and macrorhynchos. The main differences are in the breadth and shape of the temporal fossae, in the opening between the palatines and in the thickness and height of the maxillar process of the nasals. These differences are well illustrated in comparative line drawings.
There is no published analysis of vocal differences within this complex, although there are a few published qualitative descriptions. Rasmussen and Collar (2002) describe the song of macrorhynchos (presumably nominate, although this is not stated) as "a very high weak trill at variable speeds, usually descending "ui-ui-uiwi-di-dik wi-di-dik wi-di-dik" The same authors describe the song of swainsoni as "a descending sequence of whistles, varying in rhythm, "ui-ui---dibule-dibule", which seems to be taken from Sick (1993). Unfortunately, Sick does not distinguish between the voices of swainsoni and macrorhynchos, since he treated the two as a single species. However, given that Sick worked primarily in the Atlantic Forest, it seems likely that his vocal description would pertain to swainsoni. These voice descriptions do a poor job of conveying just how different the voices of swainsoni and macrorhynchos are from one another. In my experience, the songs of the wide-ranging N. m. hyperrhynchus and of N. m. paraensis are virtually identical and unvarying throughout their wide distributions. This song is described by Stiles and Skutch (1989) as "a long bubbling trill, at a constant pitch or rising slightly, then falling" and by Hilty (2003) from Iquitos, Peru as "a long, nasal, frog-like trill, prrrrrr (up to 15-20 seconds)". These descriptions fit my own tape recordings of hyperrhynchus from Chiapas, Mexico; Costa Rica; and E Ecuador; as well as my recordings of paraensis from Mato Grosso and Amazonas, Brazil. The song of nominate macrorhynchos is very different, and is described from SE Venezuela by Hilty (2003) as "a long series of rapid pree whistles (ca. 30 whistles in 8 seconds) on the same pitch" (I will deal with differences between nominate macrorhynchos and hyperrhynchus/paraensis in a separate proposal). By contrast, the song of swainsoni is essentially as described by Rasmussen and Collar in that it begins with a long series of 15+ "ui-ui-ui-ui" notes, which accelerate in pace and build in intensity before switching to a nearly as long series of 10+ disyllabic or trisyllabic "dibule-dibule-dibule" notes that descend in pitch but still accelerate toward the finish.
Analysis: Plumage differences between swainsoni and all other subspecies in the N. macrorhynchos complex are at least as great as between the latter and N. tectus (Pied Puffbird), N. ordii (Brown-banded Puffbird), or N. pectoralis (Black-breasted Puffbird). Although the study by Alvarenga et al (2002) suffers from a small sample size for the various subspecies of macrorhynchos, the biometric differences noted between all macrorhynchos and swainsoni (for which there was an impressive sample), are striking. The osteological differences were also striking. Although there is no good published analysis of vocal differences within this group, there are published qualitative descriptions that match well with my own tape recordings. The proposed split reflects an established biogeographic pattern of differentiation between Amazonian and Atlantic Forest taxa at the species level.
Recommendation: I recommend that we elevate N. swainsoni to species level, based on the preponderance of evidence from plumage, biometric, osteological, vocal and biogeographic characters. This would restore a taxonomy that was altered without justification by Peters. Based on my own field experience with each of the four taxa (macrorhynchos, hyperrhynchus, paraensis, swainsoni) involved, I have no doubts that the vocalizations alone are different enough to act as isolating mechanisms were swainsoni ever to come into contact with one of the other forms. The English name of "Buff-bellied Puffbird" as used in HBW, is appropriate for swainsoni, in that it highlights the most obvious plumage character separating that species from all other black-and-white Notharchus. If we can use a two-part proposal, I would propose first of all that we accept the split of swainsoni, and secondly, that we adopt the English name of "Buff-bellied Puffbird".
ALVARENGA, H.M.F., E. HOFLING AND L. F. SILVEIRA. 2002. Notharchus swainsoni (Gray 1846) (Bucconidae) é uma espécie válida. Ararajuba 10 (1):7377.
BURTON, P. 1973. Non-passerine bird weights from Panama and Colombia. Bull. British Ornithol. Union 93:116118.
CORY, C. B. 1919. Catalogue of birds of the Americas. Publications of the Field Museum of Natural History, Zool. Ser. 13(2):608 pp.
HARTMAN, F. A. 1961. Locomotor mechanisms of birds. Smith. Misc. Coll. 142:191.
HILTY, S. L. 2003. Birds of Venezuela. Princeton University Press, Princeton, New Jersey.
PETERS, J. L. 1948. Checklist of birds of the world, vol. 6. Museum of Comparative Zoology, Cambridge, Massachusetts.
RASMUSSEN, P. C. AND N. J. COLLAR. 2002. Family Bucconidae (Puffbirds). Pp. 102138 in: del Hoyo, J., Elliott, A., & Sargatal, J., eds. (2002). Handbook of the birds of the world, Vol. 7, Jacamars to Woodpeckers. Lynx Edicions, Barcelona.
SICK, H. 1993. Birds in Brazil, a natural history. Princeton University Press, Princeton, New Jersey.
SIBLEY, C. G. 1996. Birds of the World. Version 2.0. Thayer Birding Software. Naples, Florida.
STILES, F. G. AND A. F. SKUTCH. 1989. A guide to the birds of Costa Rica. Cornell University Press, Ithaca, New York.
Kevin Zimmer, May 2004
Comments from Stiles: "YES. The published evidence for this split is definitely much stronger than that for maintaining swainsonii within N. macrorhynchos, and Kevin's English name sounds OK to me as well."
Comments from Robbins: "YES, the published information along with Kevin's unpublished data support the recognition of swainsoni as a species."
Comments from Jaramillo: "YES. Data and logic looks solid to me. Elevation to full species warranted based on the published evidence."
Comments from Nores: "YES, aunque no con el énfasis que pone Zimmer en la propuesta. Para mí las diferencias morfológicas podrían perfectamente corresponder a una subespecie, como en el caso de Arremon flavirostris (polyonotus de dorso gris y flavirostris de dorso verde, entre otras cosas). Si hay diferencias osteológicas y de vocalizaciones, la cosa cambia y estaría más justificado el paso a especie."
Comments from Pacheco: "YES. O trabalho de Alvarenga et al. 2002 ofereceu razões suficientes para propor o restabelecimento do status de espécie ao taxon Notharchus swainsoni. O excelente arrazoado de Kevin reforça, com argumentos adicionais, esta proposição."