Proposal (562) to South American Classification Committee


Recognize newly described Thryophilus sernai


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


Lara et al. (2012) discovered a new wren from the dry Cauca River Canyon, an inter-Andean valley in Antioquia, northwestern Colombia, which is formed by the northern sectors of the Western and Central cordilleras and is enclosed at is mouth by the rainforests of the lower Cauca basin (Nechí Refuge). We described this population as a new species. Our proposal is based on an integrative study in the Thryophilus group, including the results of a comparative study of the distribution, vocalizations, morphology, and phylogeny. The details can be found in the paper, but in short, we found the following:


- Plumage and morphology: The new species is similar to T. rufalbus but is paler overall and more cinnamon-brown rather than rufous. It is also similar to T. nicefori, but the latter differs in having a darker, colder brown upperparts with faint black barring on the dorsum and upper tail-coverts. The barring of the wings and tail differs decidedly among these three species and not so much among different subspecies of T. rufalbus (Fig. 1). Thryophilus sernai is smaller in body mass and wing length than both T. rufalbus (all subspecies) and T. nicefori; it has a smaller bill than T. nicefori, and its tail tends to be longer than T. rufalbus.




Figure 1. Dorsal view (left to right) of: Thryophilus nicefori, Thryophilus sernai sp. nov., T. rufalbus cumanensis and T. rufalbus minlosi.


- Vocalizations: The new species fall within the “Thryophilus” vocal group identified by Mann et al (2009). Recordings of the new species can be examined in the Auk website or in here. Relative to other taxa in Thryophilus, the new wren’s songs have a richer repertoire of syllable types, shorter trills, lower number of trill syllables, a distinctive terminal syllable with more modulations, and higher spectral frequencies. Although there was some overlap among Thryophilus taxa as expected in their “acoustic space” characterized by multivariate analysis, songs of T. sernai are distinct but more similar to those of nicefori and most distinct from T. rufalbus. Two discriminant analyses revealed statistical significant differences in songs traits of T. rufalbus, T. nicefori, and the new taxon. Valderrama et al. (2007, 2008) previously documented the vocal behavior of T. nicefori and demonstrated its vocal distinctiveness in relation to T. rufalbus.


- Genetic differences: A phylogenetic analysis based on mitochondrial DNA (cyt-b) sequence data confirmed that the new taxon is a member of the genus Thryophilus, and that it is evolutionary isolated from all other taxa in that genus (Lara et al. 2012). Sequence divergence of T. sernai averages the following values in comparison to other taxa: T. nicefori (3.8%), T. r. castanonotus (3.8%), T. r. minlosi (3.6%), T. r cumanensis (2.7%), T. rufalbus rufalbus (5.4%), and T. rufalbus ssp. (3.7%). The new species, T. nicefori, and T. rufalbus taxa (except the last two) are reciprocally monophyletic (Lara et al. 2012, C. D. Cadena et al., in prep.).


Recommendation: Lara et al. (2012) concluded that the population from the dry Cauca Canyon in Antioquia, Colombia, was an unnamed taxon that merited species rank. The new species has unique vocal, plumage, morphological traits, it is evolutionarily isolated from closely related forms, and it occupies a small range, ecologically and geographically separated from subspecies of T. rufalbus (Fig. 2). Although reproductive isolation has not been assessed directly, some or all of the divergent traits of the new species are likely involved in territory defense, courtship, and mating, and given the current species-level taxonomy of the group, which recognizes T. nicefori at the species level, we believe species rank for the new form is warranted.




Figure 2. Distribution of Thryophilus wrens: the new species (black dots), T. nicefori (gray squares), and T. rufalbus (shaded area) in north South America and Central America.


Moreover, Lara et al. (2012) stated that: “In comparison to the differences in plumage and song existing between the two Thryophilus species pairs known to be sympatric (pleurostictus–sinaloa and pleurostictus–rufalbus), T. sernai is arguably less different from T. nicefori and T. rufalbus. However, greater divergence is expected between sympatric pairs of species than between allopatric pairs (Price 2008). Furthermore, in the context of currently accepted species limits among members of the T. rufalbus complex (Valderrama et al. 2007, Remsen et al. 2012), T. sernai appears to be just as divergent (or more so) from T. rufalbus and T. nicefori as these two good species are divergent from each other.”


We also realize that the widespread and polytypic wren T. rufalbus as currently defined is not monophyletic and likely comprises multiple species and is in need of a formal taxonomic revision. With that analysis pending, however, we believe that recognition of T. sernai is a step forward in terms of better describing the species-level diversity of this group of wrens.




Lara, C. E., A. M. Cuervo, S. V. Valderrama, D. Calderón-F. and C. D. Cadena. 2012. A new species of wren (Troglodytidae: Thryophilus) from the dry Cauca River Canyon, northwestern Colombia. Auk 129: 537–550 – and references therein.


Mann, N. I., K. A. Dingess, F. K. Barker, J. A. Graves, and P. J. B. Slater. 2009. A comparative study of song form and duetting in Neotropical Thryothorus wrens. Behaviour 146:1–43.


Valderrama, S. V., J. E. Parra, and D. J. Mennill. 2007. Species differences in the songs of the critically endangered Niceforo’s Wren and the related Rufous-and-white Wren. Condor 109: 870–877.


Valderrama, S. V., J. E. Parra, N. Davila, N., and D. J. Mennill. 2008. Vocal behavior of the critically endangered Niceforo's Wren (Thryothorus nicefori). Auk 125: 395–401.


Carlos E. Lara, C. Daniel Cadena, and Andrés M. Cuervo, October 2012




Comments from Remsen:  YES.  I reviewed the paper at pre-publication stage and strongly support the conclusions.  As the authors noted, there may be additional problems in rufalbus, but that should not affect the decision herein.”


Comments from Zimmer: “YES.  The extent of the morphological and vocal differences would appear to be consistent with species-level recognition, especially when compared against the “yardstick” of differences between other taxa currently ranked as species within the genus.  Genetic distances between sernai and other congeners are also consistent with such a ranking, and, recognition of sernai as a distinct species makes sense biogeographically.”


Comments from Stiles: “YES.  Morphology, vocalizations, and genetics are in accord with species status for sernai, regardless of what finally happens in rufalbus (where minlosi at least might merit species status).”


Additional comments from T. Donegan: “We recently assessed this species for the Colombian checklist (reference and link below).  The text of our conclusions is set out below:


‘Antioquia Wren Thryophilus sernai

Recently described from the Cauca valley in Antioquia by Lara et al. (2012) as a species.  This is clearly a new taxon, and we congratulate the discoverers.  Any decision to assign it species rank (separately from allopatric Niceforo's Wren T. nicefori and T. rufalbus) at the present time is moot but we follow Lara et al. (2012)'s approach on account of this being a plausible long-term treatment.  The new species is illustrated in Fig. 3 and its distribution is shown in Fig. 11.  There are specimens (including the type specimens) and published sonograms and photographs of T. sernai from Colombia, so it can clearly be considered "confirmed" in the country to which it is endemic.


‘Recognition of this species has proved to be one of the more controversial issues considered in this series of annual papers on the Colombian checklist.  In discussing species limits, Lara et al. (2012) considered that "it is likely that T. sernai has differentiated from T. nicefori and T. rufalbus to the point that they would behave as reproductively isolated units should they come into contact" citing differences in morphology, mtDNA and song.  They claim in the diagnosis section that the new species is "distinctive in nine acoustic variables" and that it has a "richer repertoire of syllable types, shorter trills, lower number of trill syllables, a distinctive terminal syllable with more modulations, and higher spectral frequencies".  However, there is no data available that would suggest that sernai is diagnosable to the usual 97.5% benchmark (Isler et al. 1999) used for supporting species rank determinations vocally.  Their vocal "diagnosis" is based on the Kruskal-Wallis test (in Lara et al.'s table 3), which determines the likelihood that data sets come from populations with different medians, but says nothing about the extent of differences or diagnosability. No standard deviation data is presented, so there is no way of reverse engineering the data for these purposes. In studies of other taxa, pairwise mean differences have sometimes been found consistent with miniscule differentiation and low levels of diagnosability (e.g. Donegan 2012a).  There is c.91% differentiation and considerable overlap based on recorded values in multivariate space (their figure 6) suggesting that voice is not diagnostic.  The authors' claim of shorter trills is not borne out by the illustrations in the paper (2, 5 or 9 notes in the trills in Fig. 3 for sernai, versus 2-8 in other taxa).  The claim of a richer repertoire of syllables is subjective.  Differences in modulation of the final note and overall maximum acoustic frequency are true of some but not all examples of songs in their figure 3, so again are not diagnostic.  Song can be learned in oscines such as wrens, so the possibility that differences may be cultural and perhaps could be eliminated by learning if populations were to come into contact cannot be easily dismissed.  No mention is made of whether sernai responds to playback of related species or how.

‘Lara et al. (2012) note that the rufalbus group requires revision but also consider that: "the paraphyly of species is an expected outcome of speciation processes in which differentiation occurs in peripheral populations".  There is at least one documented instance of this phenomenon in Troglodytidae (Troglodytes cobbi: Campagna et al. 2012).  However, T. cobbi is strikingly different in its ecology (absence from human modified habitats) to T. aedon, whilst T. sernai is a differently marked version of T. rufalbus / nicefori in a different dry valley.  It has a more proximate distribution to rufalbus and nicefori than T. cobbi does to T. aedon. New taxon sernai is less differentiated in its mtDNA than some other named populations in the rufalbus group are from one another (2.5-3.5% between nicefori, sernai and proximate rufalbus; compared to 6.8% between nominate T. rufalbus and subspecies castonotus).  In conclusion, it seems implausible that a rational treatment for the rufalbus group involves only T. nicefori and T. sernai being afforded species rank.  At least, nominate rufalbus and its relatives would also appear to need splitting from the southern rufalbus taxa. 


‘Despite these concerns, we recognise T. sernai on account of its broadly similar levels of vocal differentiation from rufalbus to that of nicefori, which is historically widely recognised as a species and shows similar vocal differentiation from other taxa (Valderrama et al. 2007).  Long-term, splitting sernai, nicefori and some other rufalbus taxa would seems a reasonable approach. In molecular phylogenies, sernai (like T. nicefori) is nestled within T. rufalbus and is similarly differentiated to T. nicefori.  Moreover, it would be a questionable outcome to see a potentially threatened taxon like this, with a unique distribution go unprotected whilst an open-ended taxonomic revision takes place.  A revision of species limits in the T. rufalbus group as a whole is urgently called for however.’



“For such an accessible population, Lara et al. (2012)'s lack of mention of playback experiments seems anomalous and raises most questions.  The results of playback studies were key in affording species rank to Henicorhina negreti, the only other recently described wren species occurring in western Colombia.  This new species was described by Salaman et al. (2003), a paper highly relevant to the sernai description but not cited.  It seems unimaginable that the authors would not have played sernai an MP3 of nicefori and/or rufalbus given the widespread and low-cost use of Ipods and similar devices for playback nowadays and easy availability of recordings of these other species.  Did sernai respond?  T. rufalbus has never been that difficult a bird, and now T. nicefori is also accessible in a few known localities.  Do they respond to sernai?  How? Is this relevant to species limits?


“There would not seem any reason of principle as to why a species should not be recognised in a paraphyletic group like this, even if the description increases overall paraphyly, where the treatment is defendable as a long-term approach and differentiation of the study taxon is consistent with that of other species.  The following might be a less speculative justification: "The paraphyly of species has been found in some better-studied groups and is an expected outcome of historical processes for nomenclature and taxonomy in groups which only recently have been subject to vocal and molecular investigations".


“Reference:  Donegan, T.M., Quevedo, A., Salaman, P. & McMullan, M. 2012. Revision of the status of bird species occurring or reported in Colombia 2012.  Conservación Colombiana 17: 4-14.


Comments from Cadena:  YES (I am a coauthor of the description). Regarding the "anomaly" noted above, I do not know if people have played the song of nicefori and rufalbus to sernai; I only saw the bird in life for the first time last month. I bet people probably have, however, and considering that nicefori responds to songs of rufalbus and that wrens may often be interspecifically territorial (good published examples in Thryothorus and Henicorhina), I bet that sernai likely responds. What would that mean? This is something one cannot tell without carefully designed reciprocal experiments (its not simply a matter of taking an iPod once to the type locality), which someone should probably carry out at some point when a much-needed revision of the whole rufalbus complex is undertaken. I will leave it to readers and to other committee members to evaluate whether the additional criticisms are valid. I stand by our conclusions that, taken together, and relative to the existing species-level taxonomy of the group, sernai is best ranked as a species-level taxon.”


Comments from Nores: “YES. Vocalizations, morphology, and especially the genetic differences match the species status for this taxon.”


Comments from Robbins: “YES.  All data support the recognition of T. sernai as a species.”


Comments from Stiles: “YES.  Recognizing sernai as a distinct species (at least pending a general in-depth study of the entire rufalbus complex) seems the best option.”


Comments from Pacheco: “YES.  I also agree that treatment as a species is the best choice for this complex.”