Proposal (621) to South American Classification Committee


Recognize newly described Dendrocolaptes retentus and split Dendrocolaptes certhia into six species


Effect on South American CL: If adopted, this proposal would add a newly described species of Dendrocolaptes to the list and recognize five current subspecies of D. certhia (radiolatus, juruanus, concolor, ridgwayi and medius) as species-level taxa.

Background: Over the past, the Barred Woodcreeper (Dendrocolaptes certhia) has been treated as a polytypic species, with six recognized subspecies (Marantz et al 2003).  Some authors (Peters 1951) have treated one of these subspecies as a species level taxon (D. c. concolor), whereas others have interpreted some taxa as merely representing hybrid swarms (D. c. polyzonus and D. c. ridgwayi; Zimmer 1934, Marantz 1997, Marantz et al 2003).


New information: A previous preliminary molecular study found D. c. radiolatus and D. c. concolor to be as equally or more divergent from each other than either is to D. certhia sister species (D. sanctithomae), suggesting that the former, as currently defined, may include more than a single species (Marantz et al. 2003).  Consistent with this finding, Batista et al. (2013) published a molecular phylogeny showing the existence of seven reciprocally monophyletic groups in the polytypic D. certhia, each corresponding to an already named taxon, except one including birds from the Xingu – Tocantins interfluvium, which had been considered until now a hybrid swarm between D. c. concolor and D. c. medius. The obtained molecular data does not support a hybrid status for this population, neither for that attributable to ridgwayi, which has also been submerged under concolor based on the same argument of representing a hybrid swarm (Marantz 1997, Marantz et al. 2003, Marantz and Patten 2010).


Analysis/Recommendation: The phylogeny obtained by Batista et al. (2013) was based on sequences of the mitochondrial genes cytochrome b (1016 bases pairs) and NADH subunit 2 (1045 bases pairs) of 73 individuals of all taxa grouped under the polytypic Dendrocolaptes certhia (sensu Marantz et al. 2003; i.e., certhia, radiolatus, polyzonus, juruanus, concolor, ridgwayi, and medius. The phylogeny estimated by Bayesian inference recovered with high statistical support a topology whereby seven main lineages not corresponding entirely to current subspecific limits within the polytypic D. certhia were found (for example, samples of juruanus and polyzonus clustered together in the same clade). Also, the fact that ridgwayi and the neighboring population endemic to the Xingu – Tocantins constitute non-sister, independent, and reciprocally monophyletic lineages separated by 0.9% of uncorrected sequence divergence from each other, and 1.7% from the taxon under which they have been submerged (concolor), argue for their validity as distinct taxa, supporting the recognition of the former group (for which no previous name is available) as a new taxon (D. retentus).


Three nuclear genes were also sequenced for most of the specimens and all taxa sequenced for the mitochondrial genes (Batista 2012); even though these markers revealed significant phylogeographic structure in the D. certhia complex, haplotype networks estimated independently for each gene showed main that main groups of alleles are shared among taxa, except for the MUSK gene, which included a particular group of alleles found only in individuals of nominate certhia (Batista 2012; Fig. 6). More importantly, when both nuclear and mitochondrial genes were analyzed together, the obtained species tree continued to support with overall high posterior probabilities the reciprocal monophyly among the seven main lineages / taxa of the D. certhia complex (Batista 2012; Fig. 3), mirroring the results of the mitochondrial tree showed in Batista et al. (2013).


Interestingly, the genetic data showed that weakly barred and barred taxa of the polytypic D. certhia are polyphyletic, thus demonstrating that degree of barring (traditionally used as a taxonomic character in this group) can result in misleading arrangements from an evolutionary perspective. Hence, based on the reciprocal monophyly and phenotypic diagnoses of the following taxa, we recommend the recognition of seven species and vernacular names in the D. certhia complex: Amazonian Barred Woodcreeper (D. certhia; distributed on the Guianan shield north of the Amazon and east of the Negro River in Venezuela, Brazil, Guyana, Surinam, and French Guiana); Napo Woodcreeper (D. radiolatus; found west of the Negro River in Amazonian Brazil and Venezuela westward towards the base of the Andes in Colombia, Ecuador, and Peru north of the Amazon / Solimões rivers); Juruá Woodcreeper (D. juruanus; distributed west of the Madeira and south of the Amazonas/Solimões rivers in Amazonian Brazil and towards the base of the Andes in Bolivia and Peru); Plain-colored Woodcreeper (D. concolor; found in the Madeira - Tapajós interfluve in Amazonian Brazil south to northern Bolivia in Dept. of Santa Cruz); Ridgway's Woodcreeper (D. ridgwayi; distributed in the Tapajós - Xingu interfluve in Amazonian Brazil); Xingu Woodcreeper (D. retentus; occurring in the Xingu - Tocantins interfluve in Brazil); and Todd's Woodcreeper (D. medius; found east of the Tocantins River in Pará towards western Maranhčo, with an isolated population in the Brazilian states of Alagoas and Pernambuco).



Literature Cited


Batista, R. (2012). Filogeografia e limites inter-específicos em Dendrocolaptes certhia (Aves: Dendrocolaptidae). Unpublished Master's thesis, 55pp. Pós-Graduaćčo em Zoologia, Universidade Federal do Pará / Museu Paraense Emílio Goeldi, Belém, Brazil. Available at:

Batista, R., Aleixo, A., Vallinoto, M., Azevedo, L., Sena do Rźgo, P., Silveira, L. F., Sampaio, I. & Schneider, H. (2013).  Molecular systematics and taxonomic revision of the Amazonian Barred Woodcreeper complex (Dendrocolaptes certhia: Dendrocolaptidae), with description of a new species from the Xingu-Tocantins interfluve.  In J. del Hoyo, A. Elliott, J. Sargatal & D. Christie (Eds), Handbook of the Birds of the World.  Special Volume: New Species and Global Index, pp.245-247.  Lynx Edicions, Barcelona.


Marantz, C., 1997. Geographic variation of plumage patterns in the woodcreeper genus Dendrocolaptes (Dendrocolaptidae). Ornithological Monographs. 48, 399 429.


Marantz, C., Patten, M. A., 2010. Quantifying subspecies analyzes: a case study of morphometric variation and subspecies in the woodcreeper genus Dendrocolaptes. Ornithological Monographs. 67, 123 – 140.


Marantz, C. A.; Aleixo, A.; Bevier, L. R.; Patten, M. A., 2003. Family Dendrocolaptidae (Woodcreepers). In: del Hoyo, J.; Elliott, A.; Christie, D.A. (ed.), Handbook of birds of the world, pp. 358-447. Lynx Edicions, Barcelona, Spain.


Peters, J. L., 1951. Check-List of Birds of the World. Massachusetts (U.S.A):  Cambridge Museum   of Comparative Zoology.


Zimmer, J. T., 1934. Studies on Peruvian birds No. 14. Notes on the genera Dendrocolaptes, Hylexetastes, Xiphocolaptes, Dendroplex, and Lepidocolaptes. American Museum Novitates 753.


Romina Batista and Alexandre Aleixo, December 2013




Comment from Thomas Donegan: “We considered these and other HBW splits relating to Colombian taxa for purposes of the Colombian checklist update this year. The taxonomic proposals here were not adopted for the following reasons: "[Batista et al. 2013] found the loudsong of newly described D. retentus not to 'differ constantly from those of other taxa in the D. certhia complex'. Moreover, all proposed split taxa demonstrate less than 2% mtDNA differences. We do not recognize proposed splits in this species in the absence of a detailed vocal study.” The lack of any noted vocal differentiation in this group is not mentioned in this proposal. Subspecies treatment would seem more appropriate for these different populations, which show some plumage differences.  These comments should not be taken as a criticism of the important molecular and morphological work that led to these taxonomic proposals, or of treatment of these taxa as species under phylogenetic concepts.  Reference: Donegan, T.M., McMullan, W.M, Quevedo, A. & Salaman, P. 2013. Revision of the status of bird species occurring or reported in Colombia 2013. Conservacion Colombiana 19: 3-10.”


Comments from Stiles: “NO. Here, such evidence as is presented justifies recognizing retentus – but as a subspecies, not a species, and especially given that the authors were apparently unable to find diagnostic vocal differences between forms in the D. certhia complex plus the very small genetic differences cited, the evidence that these forms represent biological (as opposed to phylogenetic) species just isnęt there.  Given that the most evident differences are in the degree of barring (pretty much either strongly or very faintly barred) and the fact that these differences show something like a leapfrog pattern, it may be that the inheritance of barred vs. nearly unbarred patterns might represent a difference of few (perhaps only two?) alleles, different ones having become fixed in different forms of quite recent divergence, without evidence (distinctive vocalizations are often the key in suboscines) of reproductive isolation.  Divergence and arguably speciation in this complex clearly has not proceeded as far as in the Lepidocolaptes case, and in my view there is insufficient support for splitting the complex into multiple species.”


Comments from Remsen:  “NO.  In my view, there are no data in the paper or the proposal relevant to species limits under the BSC.  The genetic data are based on fairly large N and involve both mitochondrial and nuclear loci; and gene tree vs. species tree concerns are adequately addressed.  Nonetheless, the species boundaries are based largely on reciprocal monophyly and fairly good concordance with plumage phenotypes.  Although I don’t think genetic distance data can be used to assign species limits in allopatric taxa, even so the genetic distances among the taxa are all very weak, even by temperate-latitude standards.  The tangential mention of lack of vocal differences suggests that these taxa have not diverged to the level associated with species rank under the BSC.”


Comments from Nores: “NO. There are no data in the proposal relevant to species limits under the BSC. Gary, Van and especially Thomas Donegan give good reasons for not splitting the complex into multiple species: 1) Batista et al. 2013 found the loudsong of newly described D. retentus not to 'differ constantly from those of other taxa in the D. certhia complex'. 2) all proposed split taxa demonstrate less than 2% mtDNA differences. 3) The lack of any noted vocal differentiation in this group is not mentioned in this proposal. 4) Subspecies treatment would seem more appropriate for these different populations, which show some plumage differences.”


Comments from Zimmer: “NO, for multiple reasons already cited by other committee members.  To me, vocal differences or lack thereof are the key to species-limits in most suboscines, and particularly in woodcreepers.  I just can’t believe that birds that spend virtually all of their time pressed tightly to tree trunks in the dim light of the forest interior are really sorting themselves out on the basis of subtle plumage characters (spots versus streaks on the breast, black bars on a dark brown background versus no black bars, etc., etc.).  It is much more plausible that recognition of rivals or potential mates is based on distinctive vocalizations that allow identification at a distance in poor light and dense vegetation.  The absence of a vocal analysis in this paper, combined with my own field experience and anecdotal evidence from others that there are no diagnosable differences in the songs of the various populations of the certhia complex, is a deal killer for me.  As noted by others, the genetic distances between the various populations are unimpressive, particularly in the absence of any known vocal differences.  To me, the differences noted should be reflected in the recognition of subspecies, but not species under the BSC.”


Comments from Robbins: “NO. The combination of little vocal and genetic differentiation among these forms indicates that they best be treated as subspecies.”


Comments from Pacheco: “NO. Considerando o conceito biológico de espécie - em voga no SACC - forćoso é admitir que as evidźncias apresentadas na excelente revisčo do complexo sčo suficientes para o tratamento em nível subespecífico.”