Proposal (808) to South American Classification Committee



Species limits in Oreotrochilus


A. Elevate Oreotrochilus estella stolzmanni to species rank


Effect on South American Check List: This proposal would add a species to the list.


Background: In their phylogeny of hummingbirds, McGuire et al. (2014) found that Oreotrochilus estella stolzmanni (from extreme southern Ecuador to central Peru) is more closely related to O. melanogaster than to O. estella (from southern Peru to northern Bolivia). Samples of O. estella were from Ayacucho (LSUMNS B103835) and from an unknown location (MR 649). The study was based on mitochondrial (ND2, ND4) and nuclear (Bfib, ODC, AKI, Musk) genes, but resolution at the species level was given mainly by mitochondrial characters. Then, another study on the phylogeography of O. chimborazo (Rodríguez and Bonaccorso 2016), based on ND2 and ND4, included two newly collected samples of O. e. stolzmanni from southern Ecuador and found the same result. Finally, in their description of a new species of Oreotrochilus (O. cyanolaemus) from southwestern Ecuador, Sornoza et al. (2018) found that O. e. stolzmanni is more closely related to this new taxon than to O. e. estella, and that O. cyanolaemus and O. e. stolzmanni from a clade with O. melanogaster.


- Plumage: Fjeldså and Boesman (2018) treated O. stolzmanni as a different species from O. estella based on plumage differences, as follows: “male having metallic (not glittering) green vs matt buff-brown crown and upperparts (3); male having blackish vs chestnut belly stripe and whiter vs buffier belly sides (2); both sexes having stronger blue metallic tail (ns[1]); female having bronzy crown (ns[2]); female having green-spotted vs brown-spotted throat (2)”. Numbers in brackets refer to the “Tobias Criteria” (Tobias et al. 2010). We found considerable individual variation in tail color across Oreotrochilus specimens examined, with some individuals having steely blue-green upperside of tail and others steely green, contradicting observation of stronger metallic blue tail as a diagnostic character for O. stolzmanni versus O. estella. Still, the other plumage differences described by Fjeldså and Boesman are consistent with our observations. The overall similarity between O. stolzmanni and O. estella that has led to their treatment as conspecific is in fact quite superficial, likely ancestral and we believe that the phenotypic differences reflect evolutionary independence between these two taxa, underpinned by physical isolation and corresponding genomic divergence. Low genetic distances with striking plumage differences among allopatric forms characterize this genus, consistent with late-Pleistocene speciation and natural or sexual selection-driven differences in signaling traits. The mtDNA history of these species shows that O. estella is polyphyletic as long as it includes stolzmanni. It also shows that sister taxa of Oreotrochilus are always geographically adjacent along the latitudinal axis of the Andes, with one exception: O. estella subspecies occur both north and south of O. melanogaster, suggesting inconsistent taxonomy and the need to elevate stolzmanni to species status. Based on the body of emerging studies on genome-scale nuclear-DNA phylogeography, we can infer that it is likely that the mtDNA patterns of diversification in Oreotrochilus are largely consistent with the nuclear DNA diversification history; therefore, there is no reason to expect that the elevation of stolzmanni to species status would be overturned after the completion of genome-scale, nuclear-DNA studies.


 - Measurements: PCR-based morphometric analyses performed by Sornoza et al. (2018) could not separate O. e. stolzmanni from other samples of O. estella. However, the only two species of Oreotrochilus that were clearly separated in these analyses were O. leucopleurus and O. adela. These differences may have arisen from adaptations to different life history traits (living in drier environments or in a broader altitudinal range; feeding on columnar cacti and Puya; or not being dependent on Chuquiraga, as it happens with the other taxa).


In conclusion, O. e. stolzmanni is diagnosable based on plumage and, according to mitochondrial data, is not most closely related to O. estella.  The latter result is consistent with geographic patterns of proximity in these isolated allotaxa, and it is bolstered by at least four samples of O. e. stolzmanni coming from at least two different locations in southern Ecuador and northern Peru, indicating that it did not result from retention of ancestral haplotypes or introgression.


Thus, available data suggests that O. stolzmanni should be ranked as species. It is evolutionarily independent and it exhibits substantial phenotypic divergence, on par with other species in the genus and with hummingbird species in general.


B. Recognize newly described Oreotrochilus cyanolaemus


Effect on South American Check List: This proposal would add a newly described species to the list.


Background: We recently described a new species of Oreotrochilus, which was discovered in April-May 2017 (Sornoza-Molina et al. 2018), in the Andes of southwestern Ecuador. Seven specimens were secured (four adult males, three females), upon which the description is based. As will be summarized below, this new species is clearly diagnosable by plumage from all known taxa in the genus Oreotrochilus. Differences in vocalizations, DNA, and morphology are subtler, but are indicative of a close relationship with O. (estella) stolzmanni, for which a separate proposal to recognize as a full species is presented above.


-Plumage: The following combination of characters are diagnostic for male Oreotrochilus cyanolaemus: 1) ultramarine blue throat, 2) emerald green head tinged blue green, 3) emerald green upperparts with blue green sheen, 4) narrow and faint emerald green terminal tips to throat feathers. Additionally, females are also distinguished from all other Oreotrochilus by the following characters: 1) dusky grayish chin and upper throat contrasting with whitish lower throat, 2) emerald green head with blue green sheen, 3) emerald green upperparts with faint blue green shine, especially on rump.

Its emerald green upperparts are greener (brighter green) and bluer than all recognized Oreotrochilus taxa, the closest being O. (estella) stolzmanni and, to a lesser extent, O. melanogaster. However, O. cyanolaemus stands out for the blue sheen of its upperparts, a unique character in the genus. More importantly, throat color is unique and remarkably different from throat color in all known Oreotrochilus. Green to lime green is the predominant throat color in the genus, the exception being O. chimborazo, which has a deep violet throat (as the entire hood) in one subspecies, and deep violet throat with a turquoise green central patch.

Interestingly, the female in the new species is also diagnosable from other Oreotrochilus, being the only one with dusky olive-gray background to chin and throat, which have profuse emerald green to blue green glittering spots. All other females in the genus have whitish background with glittering green to dusky green spots.


-Morphology: There is no diagnosable character in measurements (exposed bill length, wing length, tail length, width of rectrix 1, width of rectrix 2) for the new species, but a PCA analysis placed our small sample of O. cyanolaemus within the variation range of O. (estella) stolzmanni, and close to O. melanogaster and O. chimborazo.


-Vocalizations: Again, small sample size precludes strong conclusions on voices. Yet, we suggest that the new species voice is not appreciably different in voice from O. (estella) stolzmanni. Further, limited and not-systematic playback experiments elicited strong response of three individuals of O. cyanolaemus to the song of O. (estella) stolzmanni.


-Genetic analysis: Our phylogenetic analysis based on gene ND2 indicate that the new species is closely related to O. (estella) stolzmanni and O. melanogaster, although nodal support for this relationship was relatively low. This group was recovered as sister to O. chimborazo with strong support. Genetic distances were smaller between the new species and O. (estella) stolzmanni (0.10%), O. (estella) stolzmanni and O. melanogaster (0.41%), and O. cyanolaemus and O. melanogaster (0.42%).


-Distribution: The new species has a very narrow (in elevation) and restricted (geographic) distribution, being confined to the small and rather isolated mountain ranges Chilla-Tioloma-Fierrourcu. To date, it is known from five localities. These ranges are isolated from the more continuous mountain chain of the eastern and northwestern Andes by the low and dry Jubones-León valleys, and from lower and more rugged Andean ranges in southern Ecuador by the low and dry Puyango-Catamayo drainages. Ecological niche modeling performed predicted the species presence in other mountainous areas further north, east and south, where there are no records of the new species [actually, within the known ranges of O. chimborazo and O. (estella) stolzmanni]. Still, it should be borne in mind that sample size is small (N=5). Yet, ENM was congruent in the geographic and ecological isolation of the new species. The distribution of the new species was possibly shaped by the availability of Chuquiraga jussieui and other possible factors, including soil conditions. Our Figure 2 shows that areas to the east and south of the predicted distribution of O. cyanolaemus have rather low suitability for the species, and actually lack stands of Chuquiraga (from field and herbaria data). In fact, the only site where Chuquiraga is found is at Jimbura, in extreme S Ecuador, where only O. (estella) stolzmanni has been recorded.


Conclusions: Even though reproductive isolation was not assessed directly, we suggest that the unique throat color acts as a social signal character (i.e., mate recognition by throat color). Although display behavior in the genus Oreotrochilus is poorly known, evidence available to date suggests that throat color plays a role in displays (interactions) between male and female.

Genetic distances in the genus Oreotrochilus are likely not reflected in plumages. On the one hand, O. (estella) stolzmanni and O. estella are rather similar in plumage, but show relatively large genetic distance (>2.8%). On the other hand, O. cyanolaemus and O. melanogaster are strikingly different in plumage, but have a short genetic distance.

Small genetic differences might be explained by the recent separation of the new species from the ancestral population. The clade formed by the new species, O. melanogaster and O. stolzmanni diverged about 0.2 myr.


Recommendation: Because O. cyanolaemus is clearly diagnosable from all other species in the genus Oreotrochilus, we recommend accepting this new species and placing it next to O. melanogaster until recommendation for species status of O. (estella) stolzmanni is revised and accepted. A thorough phylogeny of Oreotrochilus using genomics seems to be needed to resolve linear sequence within the genus.




Fjeldså, J. & Boesman, P. (2018). Green-headed Hillstar (Oreotrochilus stolzmanni). In: del Hoyo, J., Elliott, A., Sargatal, J., Christie, D.A. & de Juana, E. (eds.). Handbook of the Birds of the World Alive. Lynx Edicions, Barcelona. (retrieved from on 11 October 2018).


McGuire, J. A., C. C. Witt, J. V. Remsen, A. Corl, D. L. Rabosky, D. L. Altshuler & R. Dudley (2014). Molecular phylogenetics and the diversification of hummingbirds. Current Biology 24: 1–7.


Rodríguez-Saltos, C. A. & E. Bonaccorso (2016). Understanding the evolutionary history of a high Andean endemic: the Ecuadorian Hillstar (Oreotrochilus chimborazo). Neotropical Biodiversity 2: 37–50.


Sornoza-Molina, F., J. F. Freile, J. Nilsson, N. Krabbe & E. Bonaccorso. 2018. A striking, critically endangered, new species of hillstar (Trochilidae: Oreotrochilus) from the southwestern Andes of Ecuador. Auk 135(4): 1146–1171.


Tobias, J. A., N. Seddon, C. N. Spottiswoode, J. P. Pilgrim, L. D. C. Fishpool & N. J. Collar. (2010). Quantitative criteria for species delimitation. Ibis 152: 724–746.


Juan Freile, Jonas Nilsson, Christopher Witt, and Elisa Bonaccorso

December 2018




Comments from Stiles: “YES to part A: this is the split best supported by the genetic evidence, and plumage distinctions also exist. I find the very different female plumage of stolzmanni especially convincing (this concurs with what I'm finding in Heliangelus). Part B is more difficult, because plumage and genetic data conflict. There are two ways to go - treat stolzmanni, melanogaster and the new species as different species based on plumages, or treat all three as subspecies of stolzmanni given the minute genetic differences (albeit with only one gene); this could be favored by the playback between stolzmanni and melanogaster. I suspect that more thorough genetic analyses with a variety of genes and better sample sizes could resolve the problem, and a more complete set of reciprocal playback experiments might also be enlightening. I get the impression that plumages in Oreotrochilus could either be subject to strong sexual selection or show considerable homoplasy. The genetic evidence presented suggests a very recent radiation on different mountains and rapid differentiation of plumage (especially in melanogaster) not accompanied by divergence in vocalizations - which may or may not predominate in mate choice. Given the geographic isolation and divergence in ecological niche models, I will tentatively abstain on this part of the proposal, in the hopes that more evidence will be forthcoming.”


Comments from Zimmer: “A. YES.  The taxon is diagnosable on plumage characters, and genetic data show that stolzmanni is more closely related to melanogaster and cyanolaemus than it is to nominate ostella from multiple locales.

“B. NO, at least until we have more information.  As it stands, it seems to me that there is at least as strong an argument for treating cyanolaemus as a diagnosable subspecies of a polytypic stolzmanni, giving the minimal genetic distance between them, and the suggestive results of (albeit limited) playback trials.”


Comments from Robbins: “A.  YES.  B. NO, based on the current evidence, like Kevin, I’m not convinced that cyanolaemus shouldn’t be treated as a subspecies of stolzmanni.”


Comments from Areta:

“A. YES. Genetic and morphological data coincide in supporting the split.

“B. YES. Even though genetic evidence is not definitively convincing, I have trouble in considering the three highly divergent and allopatric melanogaster, stolzmanni and cyanolaemus as subspecies of the same species. As far as I can see, vocalizations seem of limited value, in general, to assess species limits in the group. I also note that within-stolzmanni genetic distance (0.18%) is larger than between stolzmanni and cyanolaemus (0.10%). However, until solid evidence on the contrary appears, I prefer to recognize cyanolaemus. melanogaster and stolzmanni as three valid species.”


Comments from Stotz:

“A. YES. This looks pretty straightforward, with plumage  and molecular data all supporting this split.

“B. NO. Currently I have to vote against recognition.  I can’t see any evidence that really argues against treatment of cyanolaemus as a subspecies of stolzmanni.”


Comments from Schulenberg: “There is resistance to recognizing cyanolaemus as a species, although not everyone who has voted to date is explicit about the basis of their reservations. Gary, however, mentioned that he could envision a single species that included cyanolaemus, stolzmanni, and melanogaster; and Kevin also mentioned the low genetic divergence between cyanolaemus and stolzmanni as a key factor in his vote.


“To balance against the low levels of genetic differentiation between cyanolaemus, stolzmanni, and melanogaster, it may worth looking in more detail at the contact zone between stolzmanni and melanogaster in Peru:


“These two taxa are in close geographic proximity (much closer than the allopatry between cyanolaemus and stolzmanni); see Zimmer 1951, the arch lumper who maintained melanogaster as a species on this basis: 


“There are no reports of introgression between stolzmanni and melanogaster. In fact, these two taxa are reported to be locally sympatric, see Short and Morony 1969: 


“Although these factors may be relevant to a proper consideration of the taxonomic status of cyanolaemus, I'm not sure how familiar all members of SACC are with these details.”


Additional comments from Areta: “After reading Tom´s comments and the papers by Zimmer (1951) and Short and Morony (1969) I reaffirm my YES vote to recognizing cyanolaemus as a valid species. Note, however, that there is no critical evidence of breeding overlap between stolzmanni and melanogaster, since they seem to overlap outside of the breeding season, and at least as far as I can see there is no evidence of breeding sympatry or syntopy. Still, the fact that they overlap at some time of the year and not, apparently, during the breeding season is informative itself. To me, the burden of proof should lie in those advocating for the merger of cyanolaemus, stolzmanni and melanogaster under a single species. I would also like to stress that if votes are going on this direction, then this proposal would have consequences in the treatment of melanogaster by SACC."


Comments from Claramunt: “A. YES. B YES. Difficult case but this does not seem to be a case of an isolated individual gene changing throat color on a uniform genomic background as other phenotypic differences are evident. Also, throat color differences provide a basis for inferring reproductive isolation despite the similarity in vocalizations.”


Additional comments from Stiles on B: “I abstained here, but in view of Tom's info and my experience with the related genus Heliangelus (in which female plumages are sometimes more indicative of relationships than those of males), I change my vote to YES.”


Comments from Jaramillo: “A. YES. This seems straightforward.

B. YES.  Additional to other comments, I think we have to treat these within the context of the entire Oreotrochilus genus. To me this form is as valid a species as others we currently recognize. For example, the morphological differences between estella and leucopleurus are slight, essentially nothing for the females. However, the two are different ecologically, particularly important is that leucopleurus is migratory whereas estella is not, and likely the two are sympatric during the non-breeding season (perhaps even breeding season depending on how long the breeding season for estella is). To me the slight morphological differences are important between these two species, even though they are slight, and the ecological separation is key and clarifies for me that it is reasonable to consider leucopleurus as separate from estella. Within the genus the males differ in head color, gorget color, tail color, breast color and extent of coloration there. So, given that this new species differs in some of these features, which are presumably important in display and mate choice, I think it adds up to accepting this as a new species.”


Comments from Pacheco: “A. YES. Multiple evidence allows supports this arrangement.

B. YES. The information brought by Tom was providential.  I agree with Nacho that the burden of proof should lie with those who advocate the lump of cyanolaemus, stolzmanni, and melanogaster as representatives of the same species.”


Comments from Remsen: “A. YES. The evidence places burden-of-proof on their treatment as conspecific, in my opinion, and if . B. YES, somewhat reluctantly, because as indicated in many of the comments above, this one is not as straightforward.  I like Alvaro’s extrapolation from the estella and leucopleurus situation.”


Additional comments from Robbins:  “I wasn’t aware of some of the information that Tom had provided after I had voted.  Now, having read what he provided, especially the Short and Morony (1969) note concerning sympatry between stolzmanni and melanogaster, I’ll change my vote to “Yes” for species recognition.  Clearly, one either has to treat these all as separate species or lump stolzmanni, melanogaster, and cyanolaemus.  To be consistent with my support of recognizing stolzmanni, I change my vote to “YES” for recognizing cyanolaemus as a species.”