Proposal (545) to South American Classification Committee

Recognize newly described Capito fitzpatricki

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

Background: Seeholzer et al. (2012) discovered a new barbet from the east slope of the southern Cerros del Sira, Ucayali, Peru, that they described as a new species. Their conclusion that this taxon represents a new species is based on phylogenetic and population genetic analyses of mitochondrial DNA sequences of the new species and C. wallacei from which the authors concluded that they are reciprocally monophyletic sister species. The new species is diagnosable by plumage and morphology from C. wallacei and is apparently endemic to a small region of montane cloud forest in the southern portion of the Cerros del Sira.

Recommendation: Based on the information provided, I think that Capito fitzpatricki is a valid new species. I recommend a "yes" vote to add this newly described barbet to the South American list.

References:

O’Neill, J. P., D. F. Lane, A. W. Kratter, A. P. Capparella, and C. Fox J. 2000. A striking new species of barbet (Capitoninae: Capito) from the eastern Andes of Peru. Auk 117: 569-577.

Seeholzer, G. F., B. M. Winger, M. G. Harvey, D. Cáceres A., & J. D. Weckstein 2012. A new species of barbet (Capitonidae: Capito) from the Cerros del Sira, Ucayali, Peru. Auk 129: 1-9.

Manuel A. Plenge, September 2012

Comments from Zimmer: “YES. I am somewhat on the fence as to whether this new taxon should be recognized at the species-level or merely as a new subspecies of the recently described species C. wallacei. Genetic distance is small relative to other species-pairs in the family, although I don’t consider that a deal-breaker by itself. The morphological distinctions primarily involve flank color and the width and tone of red of the breast band. These are clearly diagnostic enough that the two taxa satisfy PSC criteria as separate species, but the question is one of whether or not these two allopatric taxa are sufficiently distinct as to be recognized as species under the BSC. Unfortunately, we have no vocal samples that can be analyzed, and only the assertion that the calls and songs of fitzpatricki are qualitatively similar to those of wallacei. That leaves us with the plumage differences, which, although striking, are limited. If we look at the Capito auratus group, we can see taxa with similar plumage distinctions that are recognized only as subspecies. The split of C. niger from C. auratus was justified by marked vocal differences and lack of evidence of hybridization in the contact zone; neither sort of evidence is available to us in evaluating the present case. In the case of the currently constituted C. auratus, we are talking about a lowland species with a relatively continuous distribution in which the various subspecies replace one another across river barriers, but seemingly, with some intergradation in the contact zones. Conversely, in the current case, we are dealing with two montane taxa that are isolated by some 400 km from one another, with broad, intermontane gaps in between. The genetic data indicate that fitzpatricki and wallacei are relatively recently separated from one another, but the separation is real – they do appear to be on independent evolutionary trajectories at this point. Perhaps the best “yardstick” to gauge them by is provided by a pair of tanagers, Tangara argyrofenges and T. phillipsi, the latter of which, like C. fitzpatricki, is restricted to the Cerros del Sira (albeit the northern part of the range). Plumage distinctions between the two tanagers are comparable to those between the two barbets, and I’m guessing that vocal differences are also comparable in extent. Given that we treat these two tanagers as distinct species, I’m willing to do the same with the barbets.”

Comments from Stiles: “A weak YES, for much the same reasons as Kevin. The plumage differences could go either way (species or subspecies) and the genetic differences are decidedly borderline; in the absence of analyses of vocalizations, the decision comes down to whether one considers the distance between fitzpatricki and xxx sufficient to assure that they are on separate evolutionary trajectories - a sticky point, to be sure. Although the two forms are basically foothill or lower mid-elevation forms, they are fairly strong-flying canopy birds such that ca. 400(?) km of continuous lowland forest might be less of a barrier than, say, the few km of open water separating Venezuela and Trinidad for a weak-flying forest-interior motmot. Like Kevin, I come down to precedent - there are a number of fairly comparable cases in tanagers, warblers etc. where we upheld species status for two forms showing differences of similar magnitude.”

Comments from Cadena: NO, although I believe this is a spectacular finding and I very much congratulate the people involved in the discovery. This is definitively a borderline case, and the authors clearly indicate in the paper that fitzpatricki indisputably is a new phylogenetic species (i.e., a fully diagnosable population), but that the evidence that it is a previously unknown biological species (i.e., a reproductively isolated population) is far less certain. Like Kevin and Gary said, the evidence is not overwhelming, and there are multiple cases in our baseline taxonomy in which marked geographic variation in plumage between allopatric populations is recognized at the subspecies and not the species level. There is some (small) genetic differentiation, but this is not surprising considering the geographic isolation of wallacei and the new form, and this tells us little about the odds of reproductive isolation. We have no quantitative data on vocal variation. What else could we use? Kevin rightly emphasized in his analysis the degree of phenotypic differentiation between other Sira endemics and their closest relatives, but I suggest we should also look at the degree of phenotypic variation that we "tolerate" within species of barbets. An example that immediately comes to mind is Eubucco bourcierii, in which there is considerable geographic variation in the extension of red plumage in the underparts among other things, and we still treat populations as conspecific. This, however, may be a bad example because of known differences in vocalizations and because work in progress by A. Cuervo shows strong genetic structure, which may lead to recognition of several species. Consider then the case of E. versicolor, in which forms glaucogularis and versicolor are well differentiated phenotypically (males are quite distinct), yet they are treated as subspecies of a single species and they hybridize in Ayacucho/Cusco. The case of E. richardsoni is also noteworthy, with (as per illustrations in the HBW) quite striking plumage differentiation within a single recognized species. Take also the case of the Capito niger complex. With the recognition of auratus, brunneipectus, and niger as separate species, within-species plumage variation decreased substantially relative to earlier classifications lumping them in a single species, but it is still quite considerable within the polytypic auratus in terms of carotenoid and melanic coloration. In addition to plumage variation, genetic differentiation between supposedly conspecific populations of auratus separated by rivers is quite large (much higher than between wallacei and fitzpatricki). This all suggests to me that assigning subspecies rank to fitzpatricki following the BSC would be a better course of action (despite these populations being on quite clearly separate evolutionary trajectories); hence my tentative vote for NO. I wonder whether people who have worked extensively on barbets (e.g., Dan Lane) could contribute to the discussion with some additional insights I might be missing.”

Comments from Pacheco: “NO. Aceitando que a descrição claramente indica o tratamento específico apenas sob o PSC e, evidentemente, pelo arrazoado de Cadena.”

Comments solicited from Dan Lane: “To be honest, I was hoping to sit on the sidelines on this case and see what the SACC would do with it, but several folks have asked for my opinion… so here it is. As the previous commentators have said, the taxonomic status of Capito fitzpatricki vis-à-vis the BSC is about as difficult to resolve as they come. The taxon certainly is a valid named taxon (and quite an attractive one at that!), and as such is clearly a species under the PSC, but its allopatric distribution, similarity of plumage, general habitat and elevational range, and (based on anecdotal reports by the authors) voice would suggest that it is rather closely related to C. wallacei … not to mention the sister relationship recovered by the molecular phylogeny in Seeholzer et al. The most obvious differences in plumage are carotenoid-based (red, orange, yellow) pigments, with minor melanin-based pigment differences; some morphometric differences (primarily in bill dimensions) also exist. When compared to other New World barbets, it is clear that perhaps the easiest plumage characters to change are carotenoid pigments: witness the subspecies of Capito maculicoronatus, C. auratus (also see the next paragraph), and Eubucco richardsoni, and E. bourcieri and E. versicolor. Thus, based on these pigmentation differences between fitzpatricki and wallacei, I’d say that they fit best with a subspecies relationship. It is a shame that Seeholzer et al. were not able to get recordings of song of fitzpatricki, as it would be instructive to see just how similar it is to voice of wallacei, and this perhaps would be the best way to judge reproductive isolation using phenotypic characters.

“Seeholzer et al. showed a molecular divergence estimate of 1.4% (using two mtDNA genes and a nuclear gene) between fitzpatricki and wallacei; less than that between C. squamatus and C. maculicoronatus (a sister species pair from Trans-Andean South America/Panama that also share a similar song to C. wallacei, both have relatively small world ranges, and which differ from one another by 2.2%), and considerably less than the divergence (4.9%) within C. auratus, a lowland species with wide distribution and large population size. I suspect that effective population sizes may play a larger role in the importance of these divergences, so that the rates themselves may not really be comparable (as is basically suggested in Seeholzer et al.). In this last case, I think the results of Armenta et al. (2005) will show that one of Stiles’ arguments above for species status under the BSC is not quite right: Armenta et al.’s results showed that Capito auratus seems incapable of crossing the largest trunks of the Amazon/Solimões drainage, distances considerably smaller than the ca. 400 km separating C. wallacei from fitzpatricki. These are not what I would consider “strong fliers” by any means, certainly not when compared to other canopy species! Furthermore, the results of Armenta et al. (2005) also suggested that carotenoid-based plumage variation does not stop introgression among populations of Capito auratus, because red- and orange-throated populations were not separable based on the mitochondrial genes they used, suggesting that they introgressed freely where they came into contact without geographic barriers (not surprising, as there are specimens of intermediate plumage to corroborate this). Interestingly, where barriers (large Amazonian rivers) occurred, populations on opposite sides that looked identical (and have been historically considered members of the same subspecies!) showed no recent gene flow. Returning to the C. maculicoronatus/squamatus clade, perhaps more comparable to the wallacei/fitzpatricki group in having smaller distributions, the two former taxa differ from one another not in the saturation of carotenoid pigments so much as in their placement, as well as considerably greater differences in melanin pigment patterns, which I suspect may require more divergence time to happen. That their divergence value is greater (again, their population sizes must be larger, as well), I think demonstrates that they are considerably farther along their independent evolutionary trajectories than are wallacei and fitzpatricki. To make comparisons of the wallacei/fitzpatricki complex to other members of Capitonidae seems to me to be the only path we can take presently in trying to make sense of their taxonomic ranking with respect to one another under the BSC. In conclusion, I would be inclined to rank fitzpatricki as a subspecies of C. wallacei under the BSC, but again reaffirm that fitzpatricki is clearly a species by the PSC.

“Literature cited

Armenta, J. K., J. D. Weckstein, and D. F. Lane. 2005. Geographic variation in mitochondrial DNA sequences of an Amazonian nonpasserine: The Black-spotted Barbet complex. Condor 107:527–536.”

Comments from Jaramillo: “YES. I find barbets and toucans troubling. Plumage coloration and voice sometimes tell different stories in these groups, and it confuses me. Various woodpeckers are in this camp too. In this particular case it boils down to a question of personal opinion, it becomes a subjective call. I am comfortable with calling this a new species, and hope that further work gains more data that confirms that this is a good choice.”

Comments from Nores: “NO. It seems to me a subspecies of C. wallacei, for three reasons: 1) The color patterns of the two species are very similar; only on the flanks and lower chest the yellow color is replaced by red, a similar difference to that between Capito niger punctatus and C. n. aurantiicinctus. 2) They are allopatric, despite the short distance between their ranges. 3) The average P-distance (1.14%) is small for species.”

Comments from Robbins: “YES. I could go either way on the treatment of fitzpatricki as a species. Clearly, based on both plumage and limited genetic data these two barbets are in the early stages of speciation, and unless there are dramatic differences in vocalizations (doubtful), I suspect there would be gene flow if they came together. However, given that it was published as a species and there are no strong data to refute that treatment (yes, it is a subjective decision), I’ll support species designation for now.”

Comments from Pérez-Emán: “My vote on this one is a tentative NO. The work by Seeholzer et al. (2012) is another great example of the need of continued exploration for understanding our biodiversity and its distribution. It also adds to the knowledge of the avifauna of these isolated Andean ridges. C. fitzpatricki is clearly a species under the Phylogenetic Species Concept: isolated montane populations that are reciprocally monophyletic (genetically) as well as diagnosable using morphological characters. Vocalizations are not known to a degree to support a particular taxonomic rank, although both taxa seem to have similar vocalizations. I think the evidence is clear showing that this population is on an independent evolutionary trajectory in comparison to C. wallacei; however, level of divergence on both plumage and molecular characters, as well as vocal similarities, do not suggest reproductive isolation to consider them species under the Biological Species Concept. Additionally, species definition in this group is not consistent, finding levels of intraspecific divergence highly variable and larger in some species than others even when considering different type of characters (e.g., DNA and plumage coloration).”