Proposal (#278) to South American Classification Committee

Transfer some Sakesphorus species into Thamnophilus (Thamnophilidae)


Effect on SACC: This would move three species of Sakesphorus into Thamnophilus

Background & New information: As currently recognized by the SACC, the genus Thamnophilus is composed of 27 species (T. bridgesi of Central America not covered by SACC, but recognized by AOU Checklist, 1998). The genus Sakesphorus, of which six species are currently recognized, has long been thought to be polyphyletic based on vocal and behavioral similarities of some of its species to Thamnophilus (Zimmer and Isler 2003). A recent molecular study (Brumfield and Edwards 2007) sampled individuals from 24 of the 27 described Thamnophilus species, plus Sakesphorus bernardi, S. canadensis, S. melanonotus, S. melanothorax, S. luctuosus, and S. cristatus. The data matrix analyzed contained almost 4,000 base pairs of mitochondrial and nuclear sequence data, although sequences of T. multistriatus, S. melanonotus, S. melanothorax, and S. cristatus were considerably shorter.

In phylogenies reconstructed from the ND3 gene, the ND2 gene, and a combined data matrix of three mitochondrial genes (ND3, ND2, cytb) and two nuclear genes, Brumfield and Edwards (2007) found that S. bernardi, S. melanonotus, and S. melanothorax were embedded within Thamnophilus. Bootstrap support for the inclusion of the three species within Thamnophilus was high (maximum likelihood bootstrap 100% based on analysis of combined data, 99% based solely on ND2). A value of 70% of higher is considered to be well supported, thus, at least from the perspective of mitochondrial data, these three species clearly fall within Thamnophilus.

In Brumfield and Edward's (2007) phylogeny, S. bernardi occurred as the sister taxon to a clade of T. bridgesi and T. atrinucha, the three forming a clade of species with distributions restricted to the lowlands west of the Andes. Vocalizations of S. bernardi also support its placement in Thamnophilus.

The likely phylogenetic relationships of S. melanonotus and S. melanothorax have long been problematic.

The proposal here is to change the genus of S. bernardi, S. melanothorax, and S. melanonotus to Thamnophilus.


AOU Checklist. 1998.

Brumfield, R. T., and S. V. Edwards. 2007. Evolution into and out of the Andes: a Bayesian analysis of historical diversification in Thamnophilus antshrikes. Evolution 61:346-367.

ZIMMER-ISLER. HBW chapter on Thamnophilidae.


Robb Brumfield, May 2007



Comments from Remsen: "YES. Having been close to this analysis for a while, I see no reason not to make this change. Solid genetic sampling, solid analyses, and sensible results produce a phylogeny that should be reflected in our classification."

Comments from Stiles: "YES. The data look good, and are consnstent with other evidence. A well-substantiated change."

Comments from Zimmer: "YES. The basis for these transfers is well substantiated. This also provides yet another example of the importance of vocalizations versus plumage as taxonomic characters in suboscines. On the basis of plumage pattern, bernardi is remarkably similar to S. canadensis and S. cristatus, but vocal characters align it with Thamnophilus, which is also clearly indicated by the molecular analysis."

Comments from Nores: "SI, a pesar de que mofológicamente me parece difícil que sea así., especialmente en el parecido de bernardi con canadensis y cristatus. No bstante, los análisis moleculares parecern ser indicadores de lo que se basa la propuesta, siempre teniendo en cuenta las incongruencias en análisis moleculares mostrados por Livezey & Zusi (2007). Tampoco me convence demasiado la aseveración de que el canto sea tan diferente de los otros Sakesphorus y sea tan similar a Thamnophilus. Incluso me pregunto a que Thamnophilus? ya que hay también variación entre las especies. Por ejemplo, el canto de T. caerulescens es bastante diferente que el de T. ruficapillus o el de T. doliatus."

Comments from Jaramillo: "YES - I see no reason to question the results, it all lines up nicely, molecular data, voice, biogeography ... "

Comments from Pacheco: "YES. Os resultados a partir da análise ampla de Brumfield & Edwards em combinação com os dados disponíveis conferem uma robustez a esta transferência."