Proposal (398) to South American Classification Committee

Re-split Buteo poecilochrous from B. polyosoma

With this proposal I will present in more detail the arguments for considering Buteo poecilochrous a species distinct from B. polyosoma, because I do not feel that the evidence was fully appreciated in the preceding discussion of this question. Part of the problem is that much of the relevant data were published in Spanish in journals or publications not readily accessible to members of this committee, and one of the key publications in English (Cabot & deVries 2003) in BBOC definitely suffered for want of a more firm editorial hand – perhaps the reason Mark reacted so negatively to it (?). Moreover, a detailed discussion of distribution and habitat selection of the two species, with measurements of a smaller number of specimens, was published in BBOC (Cabot 1991) and merely cited but not discussed in detail by Farquhar (1998) in the most substantive publication in favor of lumping them. However, these problems do not affect the validity of the evidence itself.

The key piece of evidence advanced by Cabot & deVries (2003, 2004) and not available to Farquhar was the detailed data on complete plumage sequences of known-sex captive birds (sexed genetically as well as morphometrically). Pavez (1998) also published a detailed description of the plumage sequence of a captive pale-morph B. polyosoma. These studies clearly established that both polyosoma and poecilochrous require up to six years to attain definitive plumage. Cabot & deVries (2004) also emphasized that both species have distinct pale and dark morphs, a fact overlooked by Farquhar (1998) and earlier by Vaurie (1962) in a more superficial study of plumages of these hawks. The existence of pale and dark phases or morphs in the related B. albicaudatus has long been known, although lack of detailed information on plumage sequences has led some (e.g., Johnsgard 1990) to suggest “continuous variation” between the extremes. Cabot and deVries found that even taking individual variation into account, the plumage sequences of polyosoma and poecilochrous were specific to morph, age and sex in both species. In general, the corresponding plumages (age for age and sex for sex) in the pale morphs of the two are more similar than in the dark morphs, but diagnostic differences are nonetheless evident (Cabot & deVries, 2008 and in press). Complex age-related variations in coloration had been mentioned (albeit not in such detail) for polyosoma by Goodall et al. (1951), Housse (1945), and Johnson (1965); more detailed information for such changes were given for the related B. albicaudatus and B. (“Geranoaetus”) melanoleucus by Lehmann (1945) and Torres-Barreto (1986). Farquhar (1998) examined parts of the plumages separately but could not put together the whole picture and did not consider feather replacement in individuals in any detail but simply classified birds as “adults” vs. “juveniles”. The sequences documented by Cabot & deVries clearly showed that Farquhar’s “age” criterion – type and extent of tail barring – varies through several subdefinitive plumages according to sex, such that no such simple division is meaningful.

Especially interesting was the discovery by Cabot & deVries that males in subdefinitive plumage show a striking resemblance to adult females. In effect, when combined with morphometric data, Cabot & deVries could distinguish 48 different plumages (six per sex, thus 12 per color phase, thus 24 per species – with the critical difference that each plumage was unequivocally related to individuals of known sex and age. This information permitted them to allocate each of the 400+ specimens examined to species, sex, and age categories, as well as detecting geographic variation in both species. In the process, they also detected various misidentified and mis-sexed specimens (many of the older collectors, as well as many museum workers, apparently sexed specimens by plumage or used the now thoroughly discredited (by both Farquhar and Cabot & deVries) wing formula of Stresemann to distinguish species). In some cases, specimens had even been correctly sexed by the collectors and subsequently erroneously re-sexed by others examining the skins later (see also Jiménez 1995). Faced with such a diversity of plumages and lacking this key information, it is hardly surprising that Farquhar ascribed plumage variability in these forms to random variation. Moreover, I emphasize that regardless of the details, this information negates his simple “adult-juvenile” age classification.

Regarding the validity of such information, I note that much of what we know about plumage sequences in large raptors (especially tropical species) is due to observant falconers and less often, zoo-keepers: only they have the opportunity to make continuous, close-range observations on known individuals over many years. Museum specimens are analogous to single isolated snapshots; banded birds recaptured, to time-lapse photos of known individuals; but the data advanced by Cabot & deVries are the equivalent of videotapes – and with which would one prefer to document the exact sequence of events in, say, a store robbery? Mark questioned the provenance of the birds used by Cabot & deVries. These birds were taken from local nests in Peru and Ecuador by falconers, and subsequently imported legally into the zoo at Lima, where they are now reproducing in captivity under Cabot’s supervision, such that they and their offspring can be studied further.

With respect to the morphometric analyses of Farquhar and Cabot & DeVries, I have already noted that the oversimplified age criterion of the former is bound to cause problems. Cabot & deVries found that some measurements, especially wing and tail length, varied between age stages, (see also Housse 1945, Jiménez 1995), as does the form of the primary tips (Vaurie 1962). They also detected geographic variation in both polyosoma and poecilochrous. Birds from the southern portion of the range of poecilochrous (Peru through extreme northern Chile and Argentina) average smaller in a wide variety of measurements and differ in various details of plumage; they recognize these as a new subspecies fjeldsaai (Cabot & deVries, submitted). On the other hand, the southernmost populations of polyosoma, beyond the range of poecilochrous, are composed of larger individuals to the extent that subspecific distinction could be warranted (Cabot & deVries, submitted). A further complication is that the lowland birds of the Tumbesian region of SW Ecuador and extreme NW Peru (subspecies peruviensis) are smaller than highland birds in the same latitudes and apparently are separated by altitude (no known overlap). There are also island populations of polyosoma in the Malvinas (= Falklands) and the Juan Fernandez Islands; the latter in particular is considered a distinct subspecies exsul that might even warrant species status, as pointed out by Farquhar (1998). Raty has already noted that Farquhar’s use of residuals is based upon the assumption that each form varies in the same manner with respect to geography (longitude and altitude), age and sex; the data of Cabot & deVries conclusively demonstrate that this assumption is invalid. By lumping together all individuals of both forms into “adults” vs. “juveniles” of each sex (likely including mis-sexed and possibly misidentified birds) over the entire range of this complex in his analysis, Farquhar could scarcely fail to produce a picture of clinal variation suggesting conspecificity, especially given the overlap in overall size between measurements of male poecilochrous and female polyosoma. It would be desirable to repeat the analyses of Farquhar using refined sex and age criteria, taking geographic variation into account. For the present, the PCA of Cabot & deVries, which used such criteria and demonstrated nonoverlapping distributions in multivariate space, seems more satisfactory, and to date they have encountered no intermediates between the two. The fact that most of the factors in their analysis loaded most heavily on the first or “size” principal component simply reflects the fact that sex for sex and age for age, poecilochrous is indeed the larger bird. Here again, it might be desirable to repeat this analysis using more of the shape criteria they found, which would probably increase the separation between these two forms.

Field observers with extensive experience of both forms, including Cabot (1991), Koepcke (1994), Fjeldså & Krabbe (1990), Walker (2002) and Pavez (2005), have noted that the “jizz” of the two forms differs. This difference includes not only the shape of the wingtip but also the width of the whole wing, including especially lengths of the secondaries, a dimension not considered by Farquhar (see Cabot 1991), the kind of flight (Pavez 2005), especially stationary flight (Walker 2002); this information was dismissed by Farquhar. Also criticized by Farquhar were the differences in behavior and ecology cited by Cabot & deVries, which he considered were more likely due simply to adaptations to local habitats. This is likely true, but overlooks the fact that in regions where they co-occur, polyosoma and poecilochrous do indeed occupy mostly distinct habitats – open grassy paramo and puna with many cliffs for the latter, more xeric, brushy to wooded habitats for the former – but with polyosoma also occupying more open and higher-elevation habitats both north and south of the range of poecilochrous. In this context, it is worth noting that the “southern cone perspective” supplied by Alvaro may not bear directly upon the question of the specific distinction, because only polyosoma occurs in most of this region, occupying the entire range of elevations, and the larger size of southern individuals of this species has already been noted by Cabot & deVries. Other Chilean workers with experience of both forms accept the validity of poecilochrous as a species (e. g, Jaksic & Jiménez 1990, Jaksic et al. 1991). Of particular interest here is the recent discovery of a population of poecilochrous breeding in the Antofagasta region of northern Chile along river valleys with facing cliffs in which gallery forest has been removed (thus producing the more open situations with cliffs favored by this species); polyosoma also occurs in the region but in a different habitat (Cabot et al., a & b., submitted; see also Fjeldså & Krabbe 1990, Jiménez 1991, Buitrón-Jurado et al. in press).

Farquhar found little difference in the one ‘territorial-alarm’ vocalization he analyzed, stating that the variation between both forms was encompassed by that shown in the related B. albicaudatus. Whether this vocalization is critical to species recognition is debatable; one could also argue that it might function in exclusion of polyosoma by poecilochrous from its more specific preferred habitats. Convergence (or in this case, non-divergence?) of territorial vocalizations has been documented in a number of related and probably competing species pairs where their territories abut. Cabot & deVries considered this analysis “trivial” and asserted that experienced field observers could indeed distinguish two species on the basis of vocalizations (see Fjeldså & Krabbe 1990, Walker 2002, Pavez 2005). Here, I consider the published evidence to be weak on both sides of the fence: the vocalization analyzed by Farquhar might not be the most appropriate for the question of specific distinction, but detailed spectrographic analyses of a range of vocalizations of both forms is lacking. DeVries and his students are now actively recording both species, but for the present, I feel that published evidence on vocalizations is simply inconclusive to resolve this question.

The genetic data are also debatable. Farquhar reported no difference between the two forms in a 600 bp sequence of the cytochrome b gene, as reported by Reising et al. (2003), and criticized Cabot & deVries for not considering this data - a bit unfair, as this paper and that of Cabot & DeVries in BBOC were published nearly simultaneously. Incidentally, Cabot has recently confirmed this, as reported in Cabot et al., in press and Buitron-Jurado et al., in press. However, Lerner et al. (2008) found genetic support in the CR_1 control region for independent origins of polyosoma and poecilochrous from albicaudatus, thus supporting their status as distinct species. Lack of divergence in cytochrome b sequences need not automatically indicate conspecificity – this gene has little to do with species recognition characters, and were speciation to have been quite recent, there might not been time for accumulation of mutations – in effect, we are in the realm of lightning strikes rather than molecular clocks (the widely accepted value for this gene is 2% changes in sequence per million years). In fact, Cabot & deVries (2008) postulated that speciation is indeed recent, probably since the latest glacial maximum, associated with the retreat of glaciers making high-elevation habitat available. Moreover, a probably comparable case exists in the genus Buteo: B. swainsoni and galapagoensis show identical cytochrome b sequences, but are universally recognized as species based upon multiple morphological and behavioral criteria (Hull et al. 2008). I might also state a case in point from the Bogotá region – that of the hummingbirds Coeligena bonapartei and C. helianthea. These species differ strikingly in coloration but are quite similar in morphometrics. Studies by Cadena and others have demonstrated no differences in cytochrome b sequences of these birds, which in fact act as distinct species in the wild. In an area where I have been observing regularly over the last 15 years, I have noted replacement of bonapartei by C. helianthea and have never observed or captured an individual with intermediate characteristics. This replacement probably reflects interspecific competition in the context of forest succession, helianthea being more associated with forest while bonapartei occurs more often in successionally earlier matorral habitats. The fact that hybrids have been reported from the period of the “Bogota” skin trade probably reflects the decimation of local populations of both, resulting in a scarcity of conspecific mates during this period – no hybrids have been reported in at least the last 80 years (a situation found in many hummingbirds of the region) and rather indicates that species distinctions have been reinforced rather than weakened with the end of such intense persecution.

In conclusion, the poecilochrous-polyosoma situation is indeed very complex – but once the underlying pattern has been deciphered, the “mess” can be resolved. In this case, the resolution involved extensive experience with wild and captive birds over many years, as well as exhaustive examination of museum specimens. The documentation of phase-, sex-, and age-specific plumage sequences in each species was a key ingredient in the resolution of the problem, and this information was not available to Farquhar. There is now a mounting accumulation of field, museum and laboratory data consistent with species status for poecilochrous, increasingly published within South America. I consider that even without conclusive vocal differences (data on vocalizations are now being gathered), this evidence definitely tips the balance in favor of recognizing poecilochrous as a species distinct from polyosoma, and urge the SACC to consider this evidence in more detail. In short, I recommend a YES on this proposal.

References:

Buitrón-Jurado, G., J. Cabot, & T. deVries. (in press). Distribution of the Variable Buzzard (Buteo polyosoma) in Ecuador. Actas I Congreso Internacional de Aves rapaces y Conservación, Quito 1-5 de abril 2008, Simbioe-Pontificia Universidad Católica de Ecuador.

Cabot, J. 1991. Distribution and habitat selection of Buteo polyosoma and B. poecilochrous in Bolivia and neighbouring countries. Bulletin of the British Ornithologists’ Club 111: 199-209.

Cabot , J. & T. de Vries. 2003. Buteo polyosoma and B. poecilochrous are two distinct species. Bulletin of the British Ornithologists’ Club 123:190–207.

Cabot, J. & T. de Vries. 2004. Age-and-sex differentiated plumages in the two colour morphs of the Variable Buzzard Buteo polyosoma: a case of delayed maturation with subadult males disguised in definitive adult female plumage. Bulletin of the British Ornithologists’ Club 124: 272-285.

Cabot, J. & T. de Vries (in press). Comments on plumage patterns of Buteo poecilochrous and B. polyosoma depicted in some ornithological guides. Actas I Congreso Internacional de Aves rapaces y Conservación, Quito 1-5 de abril 2008, Simbioe-Pontificia Universidad Católica de Ecuador.

Cabot, J. & T. de Vries. MS. A new subspecies of Buteo poecilochrous.

Cabot, J., T. deVries & C. Márquez. (in press). Sobre la especiación de Geranoaetus melanoleucus y los Busardos de cola blanca. Actas I Congreso Internacional de Aves Rapaces y Conservación. 1-5 de Abril 2008. Simbioe-Pontificia Universidad Católica de Ecuador, Quito, Ecuador.

Cabot, J., S. Alvarado & T. deVries. MS. First breeding and new distributional and altitudinal records for the Gurney Buzzard Buteo poecilochrous in Chile.

Cabot, J., T. deVries & S. Alvarado. MS. Preferencias espaciales de las rapaces en un sector de la meseta precordillerana, desierto de Atacama, región de Antofagasta.

DeVries, T. 1973. The Galapagos Hawk. An eco-geographical study with specific reference to its systematic position. PhD Thesis. Free University of Amsterdam.

Fjeldså, J. & Krabbe, N. 1990. Birds of the high Andes. Apollo Books, Svendborg, Denmark.

Goodall, J. D., A. W. Johnson & R. A. Philippi. 1951. Las aves de Chile, su conocimiento y sus costumbres. Platt Establecimientos Gráficos, Buenos Aires.

Housse, R. 1945 Aves de Chile en su clasificación moderna, su vida y sus costumbres. Ediciones de la Universidad de Chile, Santiago.

Hull, J. M., S. K. Wesley, J. L. Bollmer, R. T. Kimball, P. G. Parker, N. K. Whiteman & B. E.Holly. 2008. On the origin of the Galápagos hawk: an examination of phenotypic differentiation and mitochondrial paraphyly. Biological Journal of the Linnean Society 95:779–789.

Jaksic, F.M, Silva S., Márquez P. & Contreras, L.C. 1991. Food Habits of the Gurney Buzzard in pre-Andean ranges and the high Andean Plateau of northernmost Chile. J. Raptor Res. 25: 116-119.

Jiménez, J. E. 1995. Historia natural del Aguilucho Buteo polyosoma: una revisión. El Hornero 14:1-9.

Jiménez, J. E. & F. M. Jaksic . 1990. Diet of the Gurney Buzzard in the puna of northernmost Chile. Wilson Bulletin 102: 344-346.

Johnsgard, P. 1990. Hawks, eagles & falcons of North America. Smithsonian Institution Press, Washington, D. C.

Johnson, A.W. 1965. The birds of Chile and adjacent regions of Argentina, Bolivia and Perú. Platt Establecimientos Gráficos S.A., Buenos Aires.

Koepcke, M. 1964. Las aves del Departamento de Lima. Museo “Javier Prado”, Lima.

Lehmann F. C., 1945. Rapaces Colombianas. Subfamilia Buteoninae. Contribuciones al estudio de la Fauna de Colombia. Revista de la Universidad del Cauca 6:73-124.

Lerner, H.R.L., M. C. Klaver & D. P. Mindell. 2008. Molecular phylogenetics of the Buteonine birds of prey (Accipitridae). Auk 125 :304-315.

Pavez, E.F. 1998. Observaciones sobre el patrón de coloración en machos y hembras de aguilucho (Buteo polyosoma, Quoy y Gaymard, 1924). Boletín Chileno de Ornitología 5: 21-23.

Pavez, E. F. 2004. Descripcion de las aves rapaces chilenas. Pp: 29-103 in: Muñoz, A., J. Rau & F. Yáñez Valenzuela (eds.). Aves Rapaces de Chile. Cea ediciones, Valdivia, Chile.

Reynolds, P.W. 1935. Notes on the Birds of Cape Horn. Ibis 13^th series 5: 65-101.

Torres Barreto, A. 1986. Cetrería Neotropical Colombiana, Manual de volatería experimental con rapaces suramericanas. Instituto Colombiano de Cultura Hispánica, Bogotá.

Vaurie, C. 1962. A systematic study of the Red-backed Hawks of South America. Condor 64:277-290.

Walker, B. 2002. Field Guide to the birds of Machu Picchu, Peru. PROFONANPE- National Trust Fund for Natural Protected Areas. Lima, Peru.

Gary Stiles, May 2009

Comments from William S. Clark, Sergio H. Seipke, & C. Craig Farquhar: “We read nothing in this proposal that changes our position on this taxonomic question as stated in comments in favor of the merger of these taxa in Proposal 317. We offer specific comments in rebuttal to the arguments presented in favor of this proposal.

“First, all studies on plumage sequences by age based on captive raptors are flawed. Raptors in captivity do not molt the same as they do in the wild, unless they are flown in falconry. WSC once saw a captive Bald Eagle (Haliaeetus leucocephalus) that had gone through two molts, yet it was still in juvenile plumage. Prout-Jones and Milstein (1986) flew captive African Fish Eagles (Haliaeetus vocifer) and studied their molts and plumages from juvenile to adult. Their results were the same as others on the plumage sequence and time to adult plumage of four years. Southern (1964) also proposed six years to adult plumage for Bald Eagle, again based on captive eagles. But McCollough (1989) and Gerrard and Bortolotti (1988) found, as WSC did from reading the band of an adult Bald Eagle, that they reach adult plumage after four annual molts, or at about 4 ½ years old (Clark 2001). It is bizarre to think of a buzzard taking six years to reach adult plumage. A much better technique for determining the age classes of raptors that take more than one year to reach adult plumage is to use the molt of the remiges (Miller 1941, Clark 2004). WSC is studying White-tailed Hawk (Buteo albicaudatus), a closely related buzzard, and from noting the remige molt he found that they reach adult plumage in three years. Seipke (2007) established that the Black-chested Buzzard-Eagle (Geranoaetus [Buteo?] melanoleucus), another closely related species (but substantially larger-bodied than polyosoma-poecilochrous, 501-1280 g vs. 1700-3200 g; Dunning 2008), reaches Definitive Basic plumage after four years. Again using primary molt, SHS is currently studying the plumage sequence in Variable Hawk (B. polyosoma) from over 140 museum specimens in museums of South America and U.S.A., and he found that the species reaches the Definitive Basic (adult) plumage in three years. There is no reason to think that either of the taxa in question takes longer than three years to reach maturity. WSC was fortunate to have looked at the large number of specimens of the two taxa in question that CCF had amassed at the American Museum of Natural History in New York. It was very obvious to us that there was an almost complete overlap of plumages between them. All of the discussions in this proposal and by Cabot and de Vries (2004) on ages and plumages are without substance.

“Second, we strongly disagree with the statement that any information or discussion in Lerner et al. (2008) in any way suggests that these two taxa are separate species. On the contrary, their DNA results, as those of Riesing et al. (2003) (cited but not included in proposal references), strongly imply that the two are extremely close, if not the same species. In Lerner et al (2008) a Bayesian Posterior Probability of 1.00 leads to the nodes for each grouping of polyosoma and poecilochrous, suggesting, again, extremely close phylogenetic relationships. Stiles (SACC #398) stated that, “Lerner et al. (2008) found genetic support in the CR_1 control region for independent origins of polyosoma and poecilochrous from albicaudatus.” However, CR-1 (mitochondrial pseudo-control region) was not used in Lerner et al (2008); but data from that gene were referred to from results in Riesing et al (2003). In the latter paper it was determined that an average distance of 1.28% was found in the pseudo-control region (CR1) section of a group consisting of Buteo polyosoma and B. poecilochrous. The conclusion was that these two should not be considered separate species. In short, there is no published work on the phylogenetics of these taxa that would in any way support taxonomic distinction between B. polyosoma and B. poecilochrous.

“Third, Stiles (SACC #398) takes issue with the ‘debatable’ nature of Farquhar’s (1998) use of the species-typical territorial/alarm calls in lending support for conspecificity in polyosoma/poecilochrous. Those sounds are likely the communications most likely under selection to influence reproductive isolating mechanisms. Much has been written in the scientific literature on this subject in other species. While further data would be useful, there is no evidence that those loud, long-distance calls could be used to diagnose taxa, which is all that Farquhar (1998) meant to suggest).

Fourth, we find issue with the use of Cabot and de Vries (2003) result that discriminant function analysis (DFA) supported distinct taxa because in that paper it was apparent that the authors used a priori diagnostic criteria to place the birds into separate groups which were then validated by the very statistical technique (DFA) used to determine group membership; this is circular and erroneous for that technique. CCF (in prep.) re-analyzed data from Farquhar (1998) using principal components analysis (PCA) on correlation matrices of several morphometric variables (culmen chord, P8, P7, P6, tail, and tarsus), separated by sex, and found a continuum along the major contributor to explaining variation (e.g., P1; the ‘size’ vector) for the PCA score plot. Thus, taxonomic boundaries were not discernable.

“Finally, as there is no way to distinguish the two taxa for certain, they should be considered as one species, unless and until someone comes up with a way to distinguish them. One has to assume in light of their similarities that they would readily interbreed should they come into contact. The problem is: how will we know when that will happen? Or if it hasn’t already happened? Besides, the ability to successfully interbreed is but one (basal) character used to diagnose species.

“In summary, there is no information in this proposal to counter the solid arguments for treating these two taxa as the same species. We suggest the SACC reject proposal #398.

“Literature cited:

Cabot, J., and Tj.de Vries. 2003. Buteo polyosoma and Buteo poecilochrous are distinct species. Bulletin of the British Ornithologists’ Club 123:190-207.

Cabot, J. & T. de Vries. 2004. Age-and-sex differentiated plumages in the two colour morphs of the Variable Buzzard Buteo polyosoma: a case of delayed maturation with subadult males disguised in definitive adult female plumage. Bulletin of the British Ornithologists’ Club 124: 272-285.

Clark, W. S. 2001. Aging Bald Eagles. Birding 33:18-28.

Clark, W. S. 2004. Wave molt of the primaries of accipitrid raptors, and its use in ageing. Raptors Worldwide (eds. R. D. Chancellor, and B.-U. Meyburgh) proceedings of the V World Conf. on Birds of Prey, held in Budapest, Hungary, May 2003), World Working Group on Birds of Prey, Berlin, Germany.

Dunning, J. B., Jr. 2008. CRC Handbook of Avian Body Masses. Second Edition. CRC Press, Taylor & Francis Group, Boca Raton, London, and New York.

Gerrard, J. M. and G. R. Bortolotti. 1988. The Bald Eagle. Smithsonian Institution Press, Washington and London.

Lerner, H.R.L., M. C. Klaver & D. P. Mindell. 2008. Molecular phylogenetics of the buteonine birds of prey (Accipitridae). Auk 125:304-315.

McCollough, M. 1989. Molting sequence and aging of Bald Eagles. Wilson Bulletin 101:1-10.

Miller, A. H. 1941. The significance of molt centers among the secondary remiges in the Falconiformes. Condor 43:113-115.

Prout-Jones, D. V., and P. L. Milstein. 1986. Sequential moult with age class establishment in the African Fish Eagle Haliaeetus vocifer. S.-Afr. Tydskr. Natuurnav. 16:17-26.

Riesing, M.J., Kruckenhauser, L., Gamauf, A., Haring, E. 2003. Molecular phylogeny of the genus Buteo (Aves: Accipitridae) based on mitochondrial marker sequences. Mol. Phylogenet. Evol. 27:328-342.

Seipke, H. S. 2007. Aging Black-chested Buzzard-Eagle (Geranoaetus melanoleucus). Pp. 220-228, in Bildstein, K. L., D. R. Barber, & A. Zimmerman [eds.] Neotropical Raptors, Proceedings of the Second Neotropical Raptor Conference, Iguazu, Argentina 2006.

Southern, W. E. 1964. Additional observations on winter Bald Eagle populations: including remarks on biotelemetry techniques and immature plumages. Wilson Bull. 45:121-137.”

Comments from Dan Lane: “I find Gary's comments in Proposal 398 interesting and, to a degree, persuasive, but I fear that several important points have not been addressed with regard to the nomenclature of the Buteo polyosoma/poecilochrous complex:

“The type specimen of poecilochrous (from Yanayacu, Ecuador) reportedly (I believe this was mentioned in Farquhar 1998, which I do not have before me at the moment [VAN: PERHAPS YOU COULD CONFIRM THIS?]) does not show the primary formula pattern that has been regularly attributed as "characteristic" (indeed, it is widely touted as the primary [pardon the pun] field character for identification!) of Puna Hawk. If this is true, then can we be certain that the name is being applied to the proper taxon? Could it be that the name poecilochrous is synonymous with "polyosoma" and the "Puna Hawk" is actually not named? The holotype must be reviewed in detail in this respect.

“The name polyosoma belongs to birds on the Falklands/Malvinas (there are other names available, albeit that they are considered synonyms, from the South American mainland: erythronotus from "Strait of Magellan", aethiops from "central provinces of Chile" and peruvianus from "Eten, Lambayeque, Peru", all fide Mayr and Cottrell 1979 [Peters' Checklist, Vol I]). It seems to me that an important question that has yet to be addressed is whether or not mainland birds and Falklands birds are best considered conspecific (or at the very least monotypic) or not, since there are other very distinct taxa (many now considered species) endemic to the Falklands.

“Finally, it seems to me that until a well-supported molecular phylogeny (preferably including samples from near the type localities of the various names) is performed among the various populations within this complex to assess the presence of shared haplotypes or lack thereof (and thus determining the relative relatedness of these populations and the possibility of interbreeding between the Puna and Red-backed groups), much of this discussion is academic at best...

“Proposal 398 does not address the three points above, which are important ones with regard to what names should be applied to what taxa. Should these points not be satisfactorily resolved in a publication, I think it may be premature to make any new decisions regarding the status of the two (?) forms.”

Comments from Jaramillo: “NO – I fear that this issue is becoming the Neotropical equivalent of Larus thayeri, perhaps a grey area where clarity is elusive and strong opinions build on either side of the argument, partly due to the grey. I think thought that we are moving in the right direction and after reading all of the Cabot and deVries articles I was impressed by the picture they are painting, and the way they are teasing out some interesting patterns in what at first seemed like a random jumble of plumages. I am now convinced that there are two entities, or at least leaning in that direction, while before I was well opposed to this idea. But not all of this is new, I mean our Chile field guide is already many years old and we illustrate three ages there, and note that younger males may have reddish backs – so some of this has been known for a while. I will also note that the presence of the second stage (what we called first basic in our book), which are birds that are older than juveniles and have a barred tail with a wider and darker subterminal tail band is an age group that we had not noticed until Craig Farquhar brought it to our attention based on the work that he had done on White-tailed Hawks where he saw this intermediate age group earlier on. It may be a simplification, but it seems to me that after these two age groups the birds become essentially adult like with added barring here and there and it would not surprise me if in the field these hawks reach a terminal plumage after three complete molts, or something like that rather than after 6 years which seems excessive to me. I don’t know if Farquhar included this information in his choice of how to age birds for his paper by the way.

A few comments on the proposal:

“Other Chilean workers with experience of both forms accept the validity of poecilochrous as a species (e. g, Jaksic & Jiménez 1990, Jaksic et al. 1991).”

Note the dates of these papers -- this preceded the real controversy over these taxa. At that point the authors were going with the status quo I would assume, but I could be wrong.

“Of particular interest here is the recent discovery of a population of poecilochrous breeding in the Antofagasta region of northern Chile along river valleys with facing cliffs in which gallery forest has been removed (thus producing the more open situations with cliffs favored by this species); polyosoma also occurs in the region but in a different habitat (Cabot et al., a & b., submitted; see also Fjeldså & Krabbe 1990, Jiménez 1991, Buitrón-Jurado et al. in press).”

This is intriguing, but very confusing biogeographically. For one, there never were gallery forests in that dry part of Chile, bushes, maybe short trees yes, but no forest. So that is misleading, this part of the world is very dry, and it never had anything that could be called gallery forest, or at least the mental image of what a gallery forest is to most readers. But these sites are lower elevation spots, yet farther north in Putre the birds are polyosoma (based on the plumage features Cabot and deVries are using in their papers), and that site is at 3500m! At the latitude of Arica to Iquique, polyosoma reaches nearly to sea level in some of the valleys, and in mid altitudes in places such as the pampa del Tamarugal. Now why would these more highland specialist poecilochrous come down from the altiplano into the mid altitudes just a bit farther south of this – I just find this odd. I would also want to see exactly how they are identifying those birds in the field, just shape and size, dark morph plumages – what exactly?

But really, those are nit-picks and I apologize. The crux of the issue for me is that even accepting that there are two entities, that they have different plumage types, and average different size, and average different habitats and all that – what is the isolating mechanism? In other words, could these not be subspecies? There is mention of sympatry, but is there documentation of nesting of one type near a nest of the other type? To me it seems that sympatry in the context of the Cabot and deVries work is the presence of non-breeding polyosoma in the range of poecilochrous, which is interesting, but not really what I am looking for. Plumage differences are great to document, but in polymorphic species such as hawks it seems to me that plumage is not really all that useful of an isolating mechanism. So what are the displays, and the voices that keep these birds apart? This is the missing bit now. Farquhar says voice is not different, but he had a tiny sample size and perhaps we should be conservative and dump that bit of data, even so, no one has data that says voices are different and consistently so. In hawks that visually look alike and are sympatric (Accipiter in North America to give an example), their voices are quite different, so if these are two good biological species I would expect their voices to be different too. But I need that data to accept that two species are involved.”

Comments from Robbins: “NO. Despite the lengthy exchange, nothing has been presented that offers a definitive resolution. In fact, it appears that additional confusion has been added to this mess. Given that Clark et al. have ascertained that Buteo albicaudatus and B. polyosoma both reach definitive plumage in three years, one can’t help but question the “six years to attain definitive plumage” statement about B. poecilochrous and polyosoma. The available genetic data demonstrate that these two “taxa” (if poecilochrous is even real) are extremely similar and that there may be nothing more than minor population variation. The “jizz” comment can be dismissed for a number of reasons, and Alvaro clarifies other statements. Finally, Dan points out the same concerns that I expressed in proposal 317 regarding the holotype of poecilochrous. Given that there is little to nothing to support species recognition at this point, I’m for subsuming poecilochrous within polyosoma.”

Comments from Cadena: “NO. Interesting discussion, but I don't see anything really new and convincing to justify changing my vote on proposal 317. The situation is very confusing, and it seems that reaching a definitive decision on this case will be very difficult in the long run. I think we are best served by having these taxa lumped for now, if only for the sake of stability (how many times can we go back and forth with little to no new data?). The beauty of the online SACC list is all interested users of our classification can see all the complexities in this group by reading our discussions on them.”

Comments from Zimmer: “NO. As was the case when we previously reviewed this issue, I am not convinced by the arguments from either side. Gary has done a fine job of reviewing this exceptionally complicated case, and gives us reason to be skeptical of the methodology of the pro-lumping camp. But, I am ultimately persuaded by Alvaro’s argument that vocalizations and displays in these raptors are likely to carry more weight when it comes to species-recognition and reproductive isolation. Until we have that data, it seems we are expending a lot of time and energy in sorting out something (minute differences in biometrics and plumage sequences) that may have little bearing on whether these two populations treat one-another as separate species. In the absence of that data, I think it is better to hold off on changing our species-limits yet again. And just let me add that I, for one, am really glad that this is a taxonomic and field identification problem that we don’t have to deal with in Brazil!”

Comments from Stotz: “NO. I agree with Alvaro that it is starting to look like there might really be two taxa here, something of which I have never been convinced. However  I don’t see how even if poecilochrous and polyosoma are distinct taxa that any of the data presented here suggest that treating them as distinct species is currently the best option. As others have noted, in taxa that are so similar morphologically, you’d really expect to see vocal differences. In the absence of definitive evidence of vocal differences or clear sympatry without intergradation I can’t see treating these as distinct species.”

Comments from Stiles: “A few additional comments with reference to those of Clark et al. First, the fact that the complete sequence to definitive plumage in these hawks might take 3, 4 or 5 years is correct, and Cabot & DeVries acknowledge this – evidently the difference, at least in males, may reflect the dominance status of the males in multi-male breeding groups. However, the very fact that this is a multi-year sequence disqualifies the simple “juvenile-adult” dichotomy underlying Farquhar’s analysis – especially as a key character he used, extent of tail barring, may take more than one plumage to attain definitive and this differs between the sexes of both taxa. Much of the rest of their arguments seem to be saying that if they could not distinguish these two hawks, and did not detect any geographic variation in either, thus they are perforce indistinguishable and invariable. However, the evidence – published evidence – to the contrary is mounting; I understand that the papers by C&D recognizing subspecies in both taxa are now accepted for publication (or have been published?). Taking such geographic variation into account effectively negates another major part of Farquhar’s argument for lumping them.”

Comments from Nores: “YES, pero no porque esté convencido por lo que dice esta propuesta o por los comentarios que la apoyan. Voto SI para volver a la situación original o sea con las dos especies separadas. El hecho de que haya tanta controversia en las opiniones indica que la situación no es nada clara y me parece que SACC no puede eliminar una especie en base a situaciones como esta. Considero ahora que para eliminar una especie los fundamentos tienen que ser mucho más consistentes que para cualquier otro cambio. Alvaro dice que quisás se trate de una buena subespecie pero no de una especie, pero los fundamentos que da de lo que pasa en Chile y Argentina para mi indican que se trata de una especie o variaciones de B. polyosoma, pero no de una subespecie: “I have not looked into the specimens, but I doubt that there are that many from these areas where lowland and highland populations are entirely linked. In any case my field experience here is that there is no break, no barrier, no clear difference between highland and lowland birds”. En el análisis de Riesing et al. (2003) aparece B. poecilochrous separado de B. polyosoma, pero ellos consideran que a nivel de subespecie. Pero si no son subespecies, de acuerdo a lo expresado por Alvaro, serían especies.”

Comments from Remsen: “NO, but based almost entirely on not wanting to reverse our earlier decision until the issue is truly nailed convincingly one way or another. I now lean towards a two-species treatment based on the additional new information, but given the volatility of the entire situation and the existence of intriguing information on both sides, my view at the moment is stick to the (most recent) classification until someone demsonstrates that two species are involved by finding an isolating mechanism between the two and evidence that they do not represent a single intercommunicating gene pool along elevational gradients. That the two forms have a potential contact zone thousands of km in length provides ample opportunity for someone to study the situation directly in the field and to generate conclusive data one way or another.”

Comments from Pacheco: “NO. Os argumentos apresentados pelos dois lados que participam desta contenda não são menosprezáveis. Todavia, eu mantenho o mesmo voto dado na Proposal #317 em decorrência, sobretudo, dos resultados em Lerner et al. (2008) e Riesing et al. (2003).”