Proposal (654) to South American Classification Committee



Elevate the subspecies Anthocephala floriceps floriceps and A. f. berlepschi to species rank


Effect on SACC: This proposal, if passed, would split the Blossomcrown Anthocephala floriceps into two species.


Background: The Blossomcrown (Anthocephala floriceps Gould, 1854) is the single representative of a monotypic genus of hummingbirds endemic to Colombia, where two sedentary subspecies are recognized based on plumage variation. They live in regions separated by more than 900 km. A. f. floriceps is restricted to the foothills and mid elevations of the Sierra Nevada de Santa Marta in northern Colombia (500-1700 m), whereas A. f. berlepschi is found in the Andes (1200-2300 m) in Tolima and Huila departments.


New information: Lozano-Jaramillo et al. (2014) found that the two subspecies are reciprocally monophyletic for mitochondrial and nuclear loci, and that their divergence occurred ca. 1.4 million years before present (95% credibility interval 0.7-2.1 mybp). These data suggest an older date than reported divergence times for phylogroups within some Neotropical hummingbird species and even older than divergence times between several lineages recognized as different species of hummingbirds. Also, ecological niche models suggest that populations of A. f. floriceps are divergent in their climatic niches from populations of A. f. berlepschi, suggesting they are not ecologically equivalent. The paper is open-access and available online at:


Recommendation: Evidence for marked divergence and reciprocal monophyly in mitochondrial and nuclear loci, in addition to differentiation in climatic niche and morphological diagnosability, implies that each population is a fully diagnosable phylogenetic species. Because the two populations are disjunct and occur in widely separated areas, their status as independently evolving units ("evolutionary" species) is likely to persist over the long term. Reproductive isolation cannot be assessed directly owing to their distributions, but phenotypic differences are marked and arguably comparable to those seen between some good species of hummingbirds, suggesting these are probably best considered separate biological species as well. Therefore, we recommend elevating both subspecies to species rank.


Vernacular names: If passed, this proposal would require new English names for the taxa, we propose:


Sierra Nevada Blossomcrown Anthocephala floriceps

Andean Blossomcrown Anthocephala berlepschi


This treatment emphasizes the most prominent geographic difference between the two taxa.


Literature Cited:


Lozano-Jaramillo, M., Rico-Guevara, A., Cadena, C.D., 2014. Genetic Differentiation, Niche Divergence, and the Origin and Maintenance of the Disjunct Distribution in the Blossomcrown Anthocephala floriceps (Trochilidae). PLoS ONE 9, e108345.


Maria Lozano, Alejandro Rico, and Daniel Cadena (Nov. 2014)


Photos by David Ocampo from Daniel Cadena, added 9 Feb. 15:







Measurements from Gary Stiles (added 9 Feb. 15):


Dimensions of males of two taxa of Anthocephala hummingbirds

(todas las medidas en mm excepto donde anotado); promedio, desviacin estndar y mbito)



A. floriceps __ (n=5)

A. berlepschi __ (n=3)

Masa corporal (g)

2,76 0,10 (2,65-2,90)

3,21 0,31 (3,0-3,6)

Exposed culmen

14,84 0,46 (14,1-15,2)

16,43 0,51 (16,0-17,0)

Total culmen

16,42 0,28 (16,0-16,7)

17,97 0,15 (17,8-18,1)

Ancho de la comisura

4,46 0,11 (4,3-4,6)

4,57 0,15 (4,4-4,7)

Altura del pico (a _ de la cera)

1,60 0,07 (1,5-1,7)

1,77 0,12 (1,7-1,9)

Largo del ala cerrada

50,36 1,18 (49,8-51,6)

52,40 0,79 (51,8-53,3)

Largo del ala extendida

55,74 1,03 (54,7-57,2)

58,30 0,90 (57,4-59,2)

Cuerda (ancho mximo) del ala

17,62 0,71 (16,6-18,5)

19,13 0,50 (18,6-19,6)

rea del ala (cm2)

8,158 0,239 (7,84-8,47)

8,830 0,308 (8,55-9,16)

Largo del tarso

4,18 0,23 (4,0-4,3)

4,70 0,26 (4,5-5,0)

Extensin de la pata

8,36 0,30 (7,9-8,7)

8,57 0,06 (8,5-8,6)

Cuerda de la ua del hlux

2,11 0,07 (2,0-2,2)

2,27 0,15 (2,1-2,4)

Largo de la cola

30,20 1,60 (27,8-31,7)

31,27 0,84 (30,3-31,8)

Carga alar (g/cm2)

0,169 0,002 (0,166-0,171)

0.185 0,011 (0,175-0,197)

Razn de forma

3,167 0,119 (3,049-3,295)

3,048 0,034 (3,020-3,186)

Razn de aspecto

7,619 0,161 (7,511-7,893)

7,700 0,046 (7,652-7,742)

Puntudez del ala

0,204 0.031 (0.158-0,232)

0,263 0,013 (0,249-0,267)

Largo de la cola

30,20 1,60 (27,8-31,7)

31,27 0,84 (30,3-31,8)


Aunque las muestras tan pequeas (especialmente de berlepschi) hacen poco viable un anlisis estadstico, es notable que en 11 de los 17 parmetros no hay traslape entre las medidas de los dos taxones.  En general, parece que berlepschi es ms grande, con alas ms anchas, pero ms puntiagudas, que floriceps.  La diferencia en tamao es especialmente en el largo del pico, pero en todas las dimensiones morfolgicas el promedio es ms alto para berlepschi.




Comments from Stiles: YES.  These two look so different that I only wonder why they werent split long ago by someone with Mulsantian tendencies ... but the genetic and ecological evidence really clinch the case.


Comments from Stotz: Still thinking about this; but I wanted to comment on the English name for A. floriceps when split.  Sierra Nevada Blossomcrown is suggested, but this is a Santa Marta endemic and I think all of the Santa Marta endemics with geographic modifiers are known as Santa Marta whatevers.  There are Sierra Nevadas in a variety of places, including the United States, Spain, Mexico, and Venezuela.  If we make this split, it should be Santa Marta Blossomcrown.


Comments from Nores: YES, but I agree with Stotz in to name to Anthocephala floriceps as Santa Marta Blossomcrown.


Comments from Zimmer: YES.  It strikes me that this is a similar case to the one presented by Oxypogon, in terms of morphological distinctions and range disjunction, and our vote there was to split.


Comments from Remsen: NO, tentatively, at least until I can see documentation that the plumages differ as much as is implied.  All I have to go on is HBW, which doesnt even illustrate Central Andean berlepschi and states that it differs only by amount of white in tail tip.  Usually HBW illustrates any and all distinctive subspecies, but perhaps they did not have specimens?  From the paper and from Garys comments, I take it that that is definitely NOT the case, but this needs to be documented in the proposal (because its not in the paper or did I miss it?).  We (SACC) just split Oxypogon mostly on the grounds that they never should have been lumped because their plumage differences were at least as great as taxa ranked as species in other genera of hummingbirds; therefore, if the Anthocephala case is analogous, we should be consistent. 

            That leaves the genetic data and the niche-modeling.  First, the genetic data.  As for reciprocal monophyly, the sample sizes of individuals (3 and 4) are too small to make any confident statements about reciprocal monophyly.  In fact, the paper states: We realize our sample sizes are not large enough to provide a robust test of reciprocal monophyly, but given the strong divergence and geographic isolation, we suspect our conclusions would be robust to analyses with larger sample sizes.  I am not comfortable with suspect.  As for the degree of differentiation, the paper states: Our divergence time estimates between populations of A. floriceps (1.4 mybp) suggest an older date than the reported divergence times for phylogroups within some Neotropical hummingbird species [60,6365] and even between several lineages recognized as different species of hummingbirds [66].  That might be the case but (1) Id like to see comparative data from other Andean hummingbirds with similar distribution patterns, not just lowland ones, and (2) even so I philosophically oppose (as do many others) the use of genetic distances in general as relevant to species limits because the variation (degree of differentiation with respect to taxon rank) overlaps almost completely.

            As for the niche modeling, these are biologically important and fascinating analyses.  However, whether differences in realized niches can be used in taxon ranking is open to discussion.  For example, the authors themselves note: We note, however, that estimates of potential historical distributions based on ecological niche modeling must be considered cautiously because the realized conditions under which species exist at present (i.e., those used to build ecological niche models) may not fully represent their fundamental niches and could lead to potentially misleading reconstructions of their geographic ranges at other times.  Therefore, I would be reluctant to interpret current realized niches to represent inescapable evolutionary trajectories.  If that were the case, then all montane taxa that occur in both the Santa Martas and the Central Andes would be ranked as separate species, but that is not the case (e.g. in Trochilidae, Colibri coruscans, C. thalassinus, C. delphinae, Metallura tyrianthina, Lafresnaya lafresnayi)

            These two disjunct populations represent a distribution pattern that I think is unique in Andean birds, namely Central Andes and Santa Martas only, and this naturally creates a perception, at least for me, that they must be different species.  But the anomalous nature of the distribution pattern applies no matter what taxonomic rank the sister populations are assigned.


Comments from Robbins: YES, this seems straightforward from genetic, ecological, and obviously morphology given that these are named taxa from a time when they used only plumage morphology, i.e., there are undoubtedly plumage differences. I also agree with Doug that Santa Marta Blossomcrown would be a more appropriate name.


Comments from Cadena: Van raises some good points. We used to have more on the phenotypic distinctness of these taxa in the paper but took it out following reviews. I'll try to get some good photos of specimens and have them posted here to support this part of the story.


Comments from Areta: NO, pending critical morphological analyses and biological data. It is unfortunate that no plumage descriptions, pictures or illustrations of berlepschi and floriceps were presented by the authors. Restall's illustrations in Birds of Northern South America show a white belly and white tips to tail in floriceps, and creamy belly and creamy tips to tail in berlepschi. Although easy to find in the illustrations, the differences seem relatively minor in comparison to those among several other recently suggested splits in hummingbirds (e.g. Oxypogon and Stephanoxis). The divergence estimate of 1.4 million years with a 95% credibility interval of 0.7-2.1 million years seems to have a large error, with a threefold difference between the lower and the maximum estimate, making me wonder how much time has really passed since their divergence. Modeling results do not seem to be showing highly diverging niches in these birds, and the results might change with the incorporation of more data points from both taxa or refined modeling. I also take the extreme overprediction of the range of berlepschi and the lack of prediction of the presence of floriceps in Santa Marta in the 21,000 years ago model as indicative of problems in the usefulness of the models. [Note: I wrote this before Daniel's post. I am eager to learn more on their morphological distinctiveness]


Additional comments from Stiles: To add to my comments on the Anthocephala proposal of Daniel et al. - floriceps and berlepschi WERE considered separate species by all authors since the description of the latter, until Peters lumped them - hence, this is simply another of Peters's unsupported lumpings and if nothing else, Daniel's paper puts the burden of proof on those who would continue to consider them conspecific.  I have examined Daniel's specimens of berlepschi and recently captured and measured my first berlepschi, and can state that the differences in size and plumage colors are really substantial - the extensive white feather tips of male berlepschi stand out like a flag, and surely could function as a recognition mechanism in reproductive isolation. My bird weighed 3.9 g, vs. 2.6-3.1 for a series of 5 male floriceps - etc. With regard to English names, I agree with Doug that Santa Marta is a better descriptor than Sierra Nevada for floriceps, and might also take issue with Andean Blossomcrown for berlepschi, as it implies that this bird is widely distributed in the Andes, which it certainly is not. Its tiny known distribution is centered in the Department of Tolima, with at most a couple of reports from adjacent Huila - hence, I could suggest Tolima Blossomcrown as a better English name, calling attention to its very restricted distribution.


Comments from Jaramillo: YES. This is another Peters lump, burden of proof should be to retain the lump. I think this paper does a lot to convince about the distinctiveness and level of differentiation between these two taxa that warrant separation as species. I too think that Santa Marta Blossomcrown should be the name; Sierra Nevada is not a place (good beer) but not a place. This is as informative as saying Estados Unidos, which one de America o de Mexico? You get my point, a Sierra Nevada is not the name of the place, but a modifier of the name; the place is really Santa Marta.


Comments from Steve Hilty: Just a couple of comments. First, I think Van raised a number of good points, and I wish some of them could be pursued, and probably should be at some point. On the other hand, Gary Stiles also raised some valid points that certainly argue in favor of separating these two forms. I was just observing this species in the field in Tolima a few days ago (and also a year ago) and have seen the Santa Marta form many times over the past three or four years. The plumage differences between the two are indeed striking in the hand, but seem somewhat less so (to me) in the field but this may have more to do with time/space observing differences and observer amnesia than anything else. I don't see much difference in behavior or habitat preference between the two. Both are lower to mid-montane species that seem to feed from small, mostly tubular corolla flowers rather low around forest edges and in gardens and partly cleared areas, but always close to forest or wooded areas. A small song lek I observed (in Santa Marta Mts.) was just inside forest and birds perched mostly in lower strata of vegetation.  The current geographical separation is indeed interesting and parallels that of the various Oxypogon forms now recently split on characters less striking than those between these two forms.


Based on published work and comments posted, I suspect the two should be split, or are eventually likely to be so regarded. Doug Stotz has a very good point regarding names. Santa Marta Blossomcrown would be a more helpful name and less confusing than "Sierra Nevada"; and Gary's suggestion of Tolima Blossomcrown is an excellent suggestion and better, I feel, than "Andean", which is too general. Most of the latter's distribution is in the department of Tolima. Even before I read Gary's suggestion Tolima Blossomcrown came immediately to mind.


Additional comments from Areta: After examining the photographs and measurements I am changing my vote to a YES (thanks Gary and Daniel for providing the necessary data). Morphological and plumage differences are well marked, and I think agree with those between other good hummingbird species (assuming the two birds shown in the photographs are good representatives of features of their populations). I saw berlepschi in Tolima some months ago, but missed nominate floriceps in my single Santa Marta incursion, so I am unable to provide a satisfactory field comparison. I still regret that morphological data was not included in the paper, which would have provided a much more convincing case from the start."


Comments from Stotz: YES.  I vote in favor of this split, but not by much.  On English names, I think floriceps should be Santa Marta Blossomcrown, and I like the idea of Tolima Blossomcrown for berlepschi as suggested by Gary.