Proposal (1052) to South American Classification Committee

Revise species limits in South American Bubo owls: (A) Maintain broadly defined B. virginianus; (B) Treat magellanicus as a separate species from B. virginianus; (C) Treat nigrescens as a separate species from North American B. virginianus, irrespective of the former’s status with respect to magellanicus; (D) Treat nigrescens a subspecies of B. magellanicus; (E) Also treat B. nacurutu as a separate species from B. virginianus, restricted to North and Middle America.

Background: Proposal 328 was a first attempt laid out before SACC to change the taxonomy of Great Horned Owl (Bubo virginianus) in splitting off the austral form B. (v.) magellanicus as its own species. The complex had first been reviewed by Traylor (1958), who laid out the distribution and known taxonomy of the South American populations, with a northern Andean subspecies (nigrescens, type locality: “Cechce” <sic: Ceche, Chimborazo; Paynter 1993>, Ecuador), a southern Andean and Patagonian subspecies (magellanicus, type locality: Tierra del Fuego), a lowland savanna subspecies (nacurutu, type locality: Paraguay; including scotinus from Caicara, Rio Orinoco, Venezuela, and elutus from Lorica, Bolivar, Colombia), and a caatinga form (deserti from Salitres, near Joazeiro, Bahia, Brazil). As Mark Robbins has laid out the situation of the taxonomy of the group in the previous proposal, I won’t run over it much further here. However, largely due to a set of confounding recordings by Ted Parker from the northern Andes of Peru (Cerro Cruz Blanca, Piura) associated with a specimen at LSU (LSUMZ 97577) that has been identified as “Bubo virginianus nigrescens?” (see images), the SACC rejected the proposal at the time. These recordings seemed to represent a population of Bubo in the Andes of northern Peru that shared both “northern” and austral song types.

Since then, multiple authorities have supported the split, including Clements checklist (Clements et al 2025)—and by extension, eBird—, IOC checklist (Gill et al 2025). This split within the American Bubo has been proposed by various authors due to the distinctive song and smaller size of the austral B. (v.) magellanicus (e.g., Koenig et al. 1999, Jaramillo 2003, Pearman and Areta 2020) in comparison to the geographically nearby lowland form B. v. nacurutu, particularly because of no habitat or elevational overlap between the two despite very close proximity in Argentina (see comments by Jaramillo in Prop 328). The first publication on the genetics of the complex (Ostrow et al. 2023) showed a deep branch between B. (v.) magellanicus (including samples from as far north as the Andes of Peru) and the remainder of the B. virginianus complex, though not without some messiness. So, it is time for SACC to re-evaluate the situation.

Analysis: At the time that Mark drew up Proposal 328, we had been stymied by the absence of better voice and genetic specimen sampling along the northern Andes. Since that time, two papers have reviewed both the vocalizations of the American Bubo (López-Lanús 2015) and the genetics (Ostrow et al. 2023). Both have supported the split of B. magellanicus from B. virginianus but have not suggested any further changes to the taxonomy within the latter. In addition, new recordings, from Ecuador and Colombia, have been added to Macaulay Library (see below). I will discuss the voice and the molecular studies separately below.

Voice: So, to assess is each of these new papers. López-Lanús (2015) did an exhaustive analysis of the vocalizations of all populations of the Bubo virginianus complex from Alaska and Canada south to Tierra del Fuego using online sound archives (Macaulay Library and Xeno-canto). He concluded that there were five song types represented: (1) the widespread North American voice (hereafter called “virginianus”), (2) the Patagonian magellanicus, (3) savana nacurutu, (4) northern Andean nigrescens, and (5) also considered the Parker recordings mentioned above to represent an undescribed sixth voice type (which he called “Ñécu” and which he proposed to be an undescribed taxon). Interestingly, the virginianus voice is very strongly conserved throughout North America and is diagnosably different in pattern (sex for sex) from both nacurutu and nigrescens. I should point out, that in López-Lanús’ study, the least represented population in the sound collections he checked was nigrescens with N=4. Since that time, additional recordings have been archived, particularly in Macaulay Library, and these are key to the situation with nigrescens and magellanicus. Importantly, these recordings include several duets, which apparently are the circumstance when nigrescens gives the puttering series that Parker recorded in Piura and that is so similar to magellanicus. These recordings of the puttering duet span the distribution of nigrescens from the Eastern Andes of Colombia to Piura, Peru:

ML cuts of B. v. nigrescens giving puttering notes like B. magellanicus: ML616192787 (Cundinamarca, Colombia), ML307206661 (Azuay, Ecuador), ML582374191 (Cañar, Ecuador), ML573946951, ML617182690 (Pichincha, Ecuador), ML21879, ML21880, ML21890 (Piura, Peru). See map figure.

It is this song type that López-Lanús (2015) considered his “new” voice “Ñécu!” Apparently, it is actually a representative song type over the entire distribution of B. v. nigrescens, and this may require a new view on the relationships of that taxon.

Genetic analysis: Ostrow et al. (2023) used both nuclear UCEs and mtDNA over a large area of both the North and South American distribution of the Bubo virginianus complex, although I should note that North American samples greatly outnumbered South American; some specimens were strictly sampled from toepads (including 2 nigrescens and 2 magellanicus). The sample sizes of the South American taxa were reported as: nigrescens 3, nacurutu 1, magellanicus 6. However, I must point out that one of the “nigrescens” (a toepad sample) was from La Guajira, Colombia, and therefore was actually a nacurutu. This translates to an actual sample of nigrescens 2, nacurutu 2, and magellanicus 6. In Ostrow et al. (2023), this misidentification resulted in nigrescens being paraphyletic in their Figure 2 (all samples) which is a maximum-likelihood UCE tree, with the La Guajira sample being sister to the “single” nacurutu sample, although this now makes sense as both samples were nacurutu. I will point out that no southern nacurutu (which are closer to the type locality) were sampled, so it would be useful to know how much structure there is within the taxon, particularly between northern and southern populations.

On the “all samples” tree, the nacurutu branch comes out within the North American virginianus group among samples that are largely from the southwestern USA and Middle America. Sister to the North American/nacurutu group is nigrescens, with magellanicus sister to this entire clade. In the “fresh tissues only” tree of Figure 2, the North American clade is sister to nacurutu (so: the two are monophyletic with respect to one another) with nigrescens sister to them and magellanicus sister to that entire clade. Figure 3 of Ostrow et al. is a maximum-likelihood tree using mtDNA. This tree has the North American virginianus as a monophyletic clade with its sister being nigrescens and a specimen of magellanicus (from Lima, Peru). Sister to these clades is nacurutu (including the misidentified “nigrescens”), and finally, the branch with the remainder of magellanicus. Ostrow et al (2023) concluded that there was still gene flow between nigrescens and magellanicus (both Andean taxa) in Peru, and also (some gene flow between one of the nigrescens (Cauca, Colombia) and North American birds (!).

These genetic results have me scratching my head a bit: the trees presented in the main paper of Ostrow et al. (2023) seem not to have concordant branching among the North American birds, nacurutu, and nigrescens, but all or the bulk of magellanicus is always sister to that entire clade. But the positions of nacurutu and nigrescens with respect to the North American birds switch depending on the tree, and a Peruvian magellanicus is on the nigrescens branch on the mtDNA maximum-likelihood tree. I can’t imagine that nigrescens and North American birds are currently experiencing gene flow across the Darien (nor have they in a long time!).

Recommendation: The publications since Proposal 328 have made a bit of a hash of the distribution and taxonomy of the complex, in part because of errors in assigning names to populations, in part due to missing data that have since become available. The voice and phylogenetic placement of B. magellanicus with respect to the rest of the B. virginianus complex seems to make a strong argument that it should be recognized as a separate species. However, there appears to be molecular evidence of continuing gene flow between magellanicus and nigrescens in Peru AND the vocal repertoire of nigrescens actually contains the puttering component that makes magellanicus so distinctive as to result in its split by so many authorities! So, if the gene flow between the two taxa is real, why aren’t they on a single branch apart from the remainder of the complex? Could it be a sampling artifact?. Finally, even though nacurutu and nigrescens are both later branches off the main B. virginianus tree with respect to B. magellanicus, each clade has distinctive voices compared to the virginianus voice, and neither is actively experiencing gene flow with populations of the North American virginianus group (the separation between nigrescens and the nearest virginianus is between the Central Andes of Colombia and Costa Rica, and similarly between nacurutu in the lowlands of northern Antioquia, Colombia, and Costa Rica). My gut feeling is that we should divide up the B. virginianus complex into four species: B. virginianus (all taxa south to Panama), B. nacurutu (lowlands of South America, including the populations named scotinus, elutus, and deserti, the validity of which remain to be reviewed), B. nigrescens (Andes of Colombia to Piura, Peru), and B. magellanicus. Either this or continue to maintain the full group as one species, as it seems that nigrescens and magellanicus share the “distinctive” puttering that has been the main character used to separate the latter from the remaining complex, and that Ostrow et al. found evidence of continuing gene flow between the two in Peru. Comprehension of the molecular results of Ostrow et al is not exactly in my wheelhouse, so I will leave it to those on the committee who can address them better and how best to vote. So, these are the options I see:

A. Maintain Bubo virginianus with all of the South American populations within it until their relationships are better understood (this seems the least helpful, I recommend NO).

B. Separate B. magellanicus from the remainder of B. virginianus (this seems the favored stance by most other authors and lists, but leaves some mess swept under the rug, I recommend YES).

C. Separate nigrescens from North American B. virginianus, irrespective of the former’s status with respect to magellanicus (vocal and molecular datasets support this. I recommend YES).

D. Consider nigrescens a subspecies of B. magellanicus (there is evidence of some genetic flow between them, and vocally the two share the puttering vocalization, but it seems to be given under different circumstances within each. We don’t have much information on what sort of contact zone there is between the two in Peru. Given that Ostrow et al do not have any trees showing nigrescens to be monophyletic with respect to magellanicus, however, I think this may not be the right move. I weakly suggest NO).

E. Also separate B. nacurutu from B. virginianus, now restricted to North and Middle America (given the voice distinctions, especially compared to the surprising conservativeness within North American populations, I am inclined to recommend YES).

Literature Cited

Clements, J. F., P. C. Rasmussen, T. S. Schulenberg, M. J. Iliff, T. A. Fredericks, J. A. Gerbracht, D. Lepage, A. Spencer, S. M. Billerman, B. L. Sullivan, M. Smith, and C. L. Wood. 2024. The eBird/Clements checklist of Birds of the World: v2024. Downloaded from https://www.birds.cornell.edu/clementschecklist/download/

Gill F, D Donsker & P Rasmussen (Eds). 2025. IOC World Bird List (v15.1).

Jaramillo, A. 2003. Birds of Chile. Princeton Univ. Press, Princeton, New Jersey, USA

König, C., F. Weick, and J-H. Becking. 1999. Owls. A guide to the owls of the world. Pica Press, Sussex, England.

López-Lanús, B. 2015. Análisis domparativo de las vocalizaciones de distintos taxa del género Bubo en América. Hornero 30:69-88.

Paynter, R. A. 1993. Ornithological Gazetteer of Ecuador, second edition. Museum of Comparative Zoology, Harvard University, Cambridge, Massachusetts, USA.

Ostrow, E. N., L. H. DeCicco, and R. G. Moyle. 2023. Range-wide phylogenomics of the Great Horned Owl (Bubo virginianus) reveals deep north-south divergence in northern Peru. PeerJ 11:e15787 http://doi.org/10.7717/peerj.15787

Pearman, M. and J. I. Areta. 2020. Birds of Argentina. Princeton Univ. Press, Princeton, New Jersey, USA

Roesler, I. (2024). Lesser Horned Owl (Bubo magellanicus), version 1.1. In Birds of the World (S. M. Billerman and F. Medrano, Editors). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.grhowl2.01.1

Schulenberg, T.S., D. F. Stotz, D. F. Lane, J. P. O'Neill, and T. A. Parker, III. 2007. Birds of Peru. Princeton Univ. Press, Princeton, New Jersey, USA.

Traylor, M. A. 1958. Variation in South American Great Horned Owls. Auk 75:143-149.

Dan Lane, May 2025

___________________________________________________________________________________________________

Vote tracking chart: https://www.museum.lsu.edu/~Remsen/SACCPropChart1044-.htm

Comments from Stiles: “A. NO. Clearly at least two species are well defined genetically, vocally and phenotypically. B. YES: the easiest choice; magellanicus is clearly separate from all the rest. C [now E]. YES to also splitting off nacurutu and nigrescens as good species. In going over the distributions of all of these taxa in several recent guides and other treatments, I was surprised at the wide distribution of nacurutu, which extends from the entire Caribbean region of NE Colombia through NC Colombia into NW to NE Venezuela, and also extends through the entire region of the Llanos in both countries (if nothing else, this form was severely under-sampled in the genetic studies). It is nearly everywhere a lowland species (ca. 0-600m elevations, although it might reach higher in extreme NE Venezuela). On the other hand, nigrescens is entirely montane in its distribution in S Colombia in the Eastern and Central Andes (and the Western Andes as well?) at elevations of 2500-4000m, south in the Andes through Ecuador to NE Perú (and Bolivia?). This species is absent from Venezuela, and there might be some overlap with the range of nacurutu in Colombia but the two occupy widely different elevations (and given their different ecologies, vocalizations and genetics, gene flow between them would seem highly unlikely in the recent past.”

Comments from Krabbe: “My votes are:

A: NO

B:YES (split magellanicus from virginianus)

C: YES (split nigrescens from virginianus)

D: NO

E: YES (split nacurutu from virginianus).

“After listening carefully through the vocal material, I must agree with Dan that the Piura recordings (ML21879, ML21880, ML21890) are typical of nigrescens in both hoots and the context of the puttering series. Recordings from Cajamarca (XC139660) and Lima (ML54603211, XC215522) are like recordings of magellanicus from Chile and Argentina.

“Besides the Piura recordings, I thus concur with López-Lanús (2015) in that there are 4 vocal groups: virginianus, magellanicus, nigrescens and nacurutu. As voice appears to be similar throughout the ranges of each group, it makes sense to rank all four as biological species, which is also in general agreement with the phylogenetic trees. Despite the mitochondrial tree suggesting some gene flow between magellanicus and nigrescens, the fresh tissue UCE tree places all four in separate branches.”