Proposal (112) to South American Classification Committee



Split Pteroglossus sanguineus and P. erythropygius from P. torquatus


Effect on South American CL: This would elevate three taxa to species rank that that we treat as one species.


Background: Three distinct taxa of Pteroglossus toucans replace each other along in northwestern South America; mostly Middle American torquatus along the Panama border in n. Colombia, followed by sanguineus in most of w. Colombia to nw. Ecuador, followed by erythropygius in w. Ecuador. They are very similar in plumage, iris color, and orbital skin color with no known differences in voice. Nominate torquatus has a rust-colored nuchal collar that is absent in the other two, and it also has a completely red loral area (vs. slaty in the other two). They differ strongly, however, in bill pattern and color (see illustrations in Ridgely & Greenfield 2001, Short & Horne 2001, 2002), and also in bill structure (differing in configuration of nasal canals and number of bill serrations; Short & Horne 2001). The three taxa were traditionally (e.g., Meyer de Schauensee 1970) treated as separate species.


Then, Haffer (1967) examined study skins from the contact region between torquatus and sanguineus in n. Colombia, and showed that plumage characters strongly suggest free interbreeding between the two wherever in contact. Using hybrid index scores of plumage characters, he found, for example, that a series of 20 specimens from the Río Cope area has no "pure" phenotypes and is highly variable. The hybrid zone is narrow, perhaps only 20 km wide. Haffer (1967) illustrated four convincingly intermediate bill types between "pure" torquatus and "pure" sanguineus.


Haffer (1974) treated sanguineus and erythropygius as subspecies of torquatus, with the following remark:


"P. t. torquatus and sanguineus are known to hybridize freely along a narrow zone where they meet in northwestern Colombia (Haffer 1967a). Since the other two forms do not differ more conspicuously than the hybridizing populations, they are here ranked also as subspecies, although proof of intergradation is lacking. The zones of contact of frantzii/torquatus in the Aguacate Mountains (Slud 1964) and of P. t. sanguineus/erythropygius in northwestern Ecuador have not been studied.”


Short & Horne (2001) examined specimens from northwestern Ecuador and commented as follows:


"... (in SW Colombia and NW Ecuador situation complex, but, e.g., all six specimens of a series from Gualea, N Pichincha, Ecuador[,] are hybrids, see Plate 34, and nearly half of the specimens designated 'erythropygius', and a like proportion of SW Colombian-NW Ecuador 'sanguineus' converge in the culmen and mandible characters) ... "


Short & Horne (2002) treated the three taxa as conspecific. The AOU (1998) treated sanguineus as a subspecies of torquatus but treated erythropygius as a species. Hilty & Brown (1986), Sibley & Monroe (1990), and Ridgely & Greenfield (2001) acknowledge the narrow hybrid zone between torquatus and sanguineus but nonetheless treated them all as separate species without explanation.


Analysis: Although increased sample sizes of localities and specimens would of course be desirable, all signs point to unrestricted gene flow between these three populations where they meet; at least the burden-of-proof at this point, in my opinion, would be on making the case for treating the three taxa as species. Some regard the existence of a "narrow" hybrid zone itself as evidence of a barrier to free gene flow, but I don't like this because (1) what's "narrow"? how narrow is "narrow"?, and (2) a genetic study would probably reveal more extensive intergradation in non-plumage characters. More broadly, differences in bill pattern among Pteroglossus taxa seem of little consequence to gene flow in contact zones in Amazonia; see Haffer (1974).


Recommendation: Although I do not have any direct experience with these taxa and their contact zones, I think that the published evidence strongly favors treating them as a single species (under BSC), and therefore will vote "NO."


Literature Cited:

HAFFER, J. 1967. Speciation in Colombian forest birds west of the Andes. Amer. Mus. Novitates 2294: 1-57.

HAFFER, J. 1974. Avian speciation in tropical South America. Publications of the Nuttall Ornithological Club, No. 14.

HILTY, S. L., AND W. L. BROWN. 1986. A guide to the birds of Colombia. Princeton University Press, Princeton, New Jersey.

MEYER DE SCHAUENSEE, R. 1970. A guide to the birds of South America. Livingston Publishing Co., Wynnewood, Pennsylvania.

RIDGELY, R. S., AND P. J. GREENFIELD. 2001. The birds of Ecuador. Cornell University Press, Ithaca, New York.

SHORT, L. L., AND J. F. M. HORNE. 2001. Toucans, barbets and honeyguides. Oxford University Press, Oxford.

SHORT, L. L., AND J. F. M. HORNE. 2002. Family Ramphastidae (toucans). Pp. 220-273 in "Handbook of the Birds of the World, Vol. 7. Jacamars to woodpeckers." (J. del Hoyo et al., eds.). Lynx Edicions, Barcelona.

SIBLEY, C. G., AND B. L. MONROE, JR. 1990. Distribution and taxonomy of birds of the World. Yale University Press, New Haven, Connecticut.


Van Remsen, March 2004




Comments from Stiles: "NO. Although a thorough vocal analysis would be nice, the three sound alike to my ear, their behavior and ecology are similar and the evidence for hybridization and intergradation seems strong; the differences in bills and plumage, while notable, are likely not genetically complex. I really do not see much justification for maintaining them as separate species at this point. So, YES lump them [keep them lumped]."


Comments from Nores: "NO, considero que dada la ausencia de estudios genéticos, las razones dadas por Haffer y Short & Horne son suficientemente válidas por el momento como para no separar como especies a P. sanguineus and P. erythropygius."


Comments from Jaramillo: "YES. There are distinct morphological differences between these taxa and the hybrid zone is narrow. I am one of these people that does think a narrow hybrid zone is in fact a feature that suggests a barrier to gene flow. A broad morphological cline would be another story. A hybrid zone of 20 km seems darn narrow to me, and is presumably a typical step-cline situation. The zone of hybridization of Bullock's and Baltimore oriole in western Kansas is also a step cline, and this zone is much wider than 20 km. How narrow is narrow is a good question, and it depends on the situation, how large the range is of the taxa in question is, and I do think that it is subjective in most cases. I don't think that anyone could argue that a change from one type to another in 20 km is not abrupt. So, in this situation I do think that by most people's opinion this would be considered a narrow hybrid zone. In addition, the fact that some well-known and studied hybrid zones are now known to involve taxa that are not each other' sisters (Baltimore x Bullock's Orioles; Western x Glaucous-winged Gull _ I can look up references if anyone is interested) I would rather keep these taxa as separate rather than lump."


Comments from Schulenberg: "YES. "As far as I am aware, these taxa are very similar vocally and in "every other way".


"But I also am impressed by narrow hybrid zones. I think the question to ask is not, "how narrow is narrow?" (and 20 km is as narrow as these things get), but, "if interbreeding is unrestricted, then why is the hybrid zone so narrow?"."


Comments from Zimmer: "YES. Although I don't feel strongly about it. I agree that the differences in bill color/pattern are not that impressive given other taxon pairs in the family that are treated as conspecific (for example, flavirostris/mariae/azara), and the hybridization in the contact zones is also troubling. But, like Alvaro and Tom, I am impressed with how narrow the hybrid zones seem to be, which really does suggest that there is an impediment to gene flow. This doesn't seem to be yet another ramphastid example in which two or more forms intergrade over a broad front wherever they come into contact.”