Cercomacra, SCLATER, 1858

Tello, Jose G., Raposo, Marcos, Bates, John M., Bravo, Gustavo A., Cadena, Carlos Daniel & Maldonado-Coelho, Marcos, 2014, Reassessment of the systematics of the widespread Neotropical genus Cercomacra (Aves: Thamnophilidae), Zoological Journal of the Linnean Society 170 (3), pp. 546-565 : 555-556

RELATIONSHIPS WITHIN CERCOMACRA

Cercomacra comprises manu , brasiliana , cinerascens , melanaria , ferdinandi , carbonaria , and nigricans ( Fig. 3 ). Cercomacra manu is sister to the rest of Cercomacra , which comprises two clades: one clade formed by the circum-Amazonian taxa ( melanaria , ferdinandi , carbonaria , and nigricans ), in which carbonaria and nigricans are sister to ferdinandi , and then to melanaria ; and a second clade formed by the Amazonian cinerascens and the south-eastern Atlantic Forest brasiliana . All nodes received high posterior probability and ML bootstrap support, with the exception of the node uniting cinerascens and brasiliana ( Fig. 3 ). Resolution of this latter node has important biogeographical implications (see below) and needs to be investigated further. Vocal similarities between cinerascens and the circum-Amazonian taxa support a close relationship between these taxa ( Fitzpatrick & Willard, 1990; Zimmer et al., 1997; Isler & Whitney, 2002), as supported (albeit weakly) by the mtDNA tree ( Fig. 4A ).

Contrasting with previous hypotheses (Fig. 2), our molecular analyses found that the ‘ nigricans ’ group, as delimited by Fitzpatrick & Willard (1990), Silva (1992), or Zimmer et al. (1997), does not constitute a natural group, because brasiliana and cinerascens are embedded within this clade ( Fig. 3 ). Similarity in vocalizations, i.e. a two-element call of males and a single element in the female call during the duet, supports the inclusion of cinerascens and brasiliana within Cercomacra senso stricto ( Vielliard, 1995; Isler & Whitney, 2002; Zimmer & Isler, 2003). The phylogenetic results also agree with the previous suggestion of a close relationship between carbonaria , nigricans , and ferdinandi ( Fitzpatrick & Willard, 1990; Zimmer et al., 1997; Fig. 2A, C). The species in this subclade possess plumage (heavy streaking on throat of females) and vocal (male–female duet song pattern) similarities that support their close relationship, particularly between nigricans and carbonaria ( Fitzpatrick & Willard, 1990; Zimmer et al., 1997; Isler & Whitney, 2002; Zimmer & Isler, 2003). Our results confirm Zimmer et al.’s (1997) conclusion that characters suggesting a close relationship between melanaria and manu ( Fitzpatrick & Willard, 1990; Silva, 1992), i.e. the overall similarity in female plumages, are probably due to sharing ancestral or convergent features within the group.

Figures

Figure 3. Cladogram of the Bayesian consensus tree of Cercomacra from the ten-partition model analysis (−lnL = 37320.48, 1stND2[GTR+G], 2ndND2[TVM+I+G], 3rdND2[HKY+G], 1stND3[SYM+G], 2ndND3[TIM3+I+G], 3rdND3[TIM1+G], 1stCYTB[GTR+I+G], 2ndCYTB[TIM2+I+G], 3rdCYTB[GTR+I+G], FIB5[TIM3+G]). Support values correspond to Bayesian posterior probabilities and bootstrap values of the single-partition model ML tree (−lnL = 39246.13, HKY+I+G), respectively. Nodes that uncovered ingroup relationships are numbered 1–14.

Figure 4. Resulting phylograms of the Bayesian consensus trees of Cercomacra from the separate analyses of concatenated mitochondrial (A) and FIB5 (B) data sets. A, phylogram of Bayesian consensus tree from the concatenated mitochondrial nine-partition model analysis (1stND2[GTR+G], 2ndND2[TVM+I+G], 3rdND2[HKY+G], 1stND3[SYM+G], 2ndND3[TIM3+I+G], 3rdND3[TIM1+G], 1stCYTB[GTR+I+G], 2ndCYTB[TIM2+I+G], 3rdCYTB[GTR+I+G]). B, phylogram of Bayesian consensus tree from the FIB5 single-partition model analysis (TIM3+G). Support values correspond to Bayesian posterior probabilities and ML bootstrap values, respectively. Numbered nodes represent those also found in the concatenated analysis. Incongruent nodes between the mitochondrial (node 15) and FIB5 (nodes 11 and 12) trees are marked with asterisks.