Resurrection of the genus Uromyias
We propose the resurrection of the genus Uromyias based on recent molecular phylogenetic evidence. Uromyias should include the species currently classified as Anairetes agraphia and Anairetes agilis, including subspecies therein. The appropriate placement of A. agraphia and A. agilis in relation to all other Anairetes species (Anairetes sensu stricto) has been debated for nearly a century (Hellmayr, 1927; Smith, 1971; Traylor, 1977; Lanyon, 1988; Roy et al., 1999; DuBay and Witt, 2012). Hellmayr (1927) first placed A. agraphia and A. agilis into a distinct genus, Uromyias, based on morphological characters including a shorter, wider, and more depressed bill, more developed rectal bristles, a proportionately longer tail, and greater difference between the shortest and longest rectrices. Lanyon (1988) concurred with the validity of the genus Uromyias based on cranial morphology; specifically, Uromyias has a fully ossified nasal septum and lacks posterior forking in the trabecular plate that is characteristic of other Anairetes species. Anairetes agraphia and A. agilis are the only members of the Anairetes group that are restricted to extreme humid habitats (i.e. Chusquea dominated cloud forest), whereas species in Anairetes sensu stricto are all tolerant of drier and/or more seasonal environments.
Systematists who have favored merging Uromyias with Anairetes have done so using justifications that we now know to be misguided or incorrect. Smith (1971) rejected Uromyias on the basis of his opinion that ecological differences should not delimit generic boundaries. However, molecular phylogenies have since revealed that habitats tend to be conserved across the tyrannid phylogeny (Ohlson et al. 2008). It follows that environmental characteristics are subject to evolutionary inertia and are therefore appropriate for guiding classification. Traylor (1977) subsequently favored dissolving Uromyias because the most recently described species, A. alpinus (Carriker 1933), appeared to him to be morphologically intermediate between the two genera. We now know that A. alpinus is phylogenetically nested within Anairetes sensu stricto as sister to the clade containing A. flavirostris, A. parulus, and A. fernandezianus (Dubay and Witt 2012).
In 1999, Roy et al. provided the first molecular phylogeny of Anairetes/Uromyias and recovered the genus Uromyias nested within Anairetes, albeit with weak bootstrap support. Two separate analyses presented by Roy et al. (1999) produced different phylogenetic positions for Uromyias, but Uromyias was nested within Anairetes in both topologies. It may have been underappreciated that neither of these alternative analyses could statistically rule out the possibility that Uromyias is sister to a monophyletic Anairetes. Nonetheless, the analyses of Roy et al. (1999) provided the impetus to merge the genus Uromyias into Anairetes. The Roy et al. (1999) phylogeny used short mtDNA fragments and parsimony and distance based methods. In a recent study we revisited the phylogeny of this clade using improved taxon sampling, seven loci, 6407 base pairs, partitioned Bayesian analysis, species-tree methods, and appropriate outgroups (DuBay and Witt, 2012). The results decisively supported Uromyias and Anairetes as reciprocally monophyletic sister clades with 100% posterior probability. Increasing the number of mtDNA characters alone was sufficient to refute the Roy et al. (1999) hypothesis under parsimony, ML, or Bayesian methods. Although individual nuclear genes were unable to resolve the phylogeny, the multi-locus *BEAST species-tree method with seven loci recovered the same result as the mtDNA alone.
In sum, it appears that previous justifications for merging the two genera were guided by misconceptions about phylogeny and underestimation of the evolutionary significance of the ecoclimatic niche. The new phylogenetic results affirm the views of previous workers, including Hellmayr (1927) and Lanyon (1988), that morphological and ecological differences between the two clades warrant classification in separate genera.
DuBay, S.G., Witt, C.C., 2012. An improved phylogeny of the Andean tit-tyrants (Aves, Tyrannidae): More characters trump sophisticated analyses. Molecular Phylogenetics and Evolution 64, 285-296.
Hellmayr, C. E., 1927. Catalogue of birds of the Americas and the adjacent islands in Field Museum of Natural History. Field Mus. Nat. Hist. Publ. 242 Zool. Ser. 13, 1-517.
Lanyon, W. E., 1988. A phylogeny of the thirty-two genera in the Elaenia assemblage of tyrant flycatchers. Am. Mus. Novit. 1-57.
Ohlson, J., FjeldsŒ, J., Ericson, P. G. P., 2008. Tyrant flycatchers coming out in the open: phylogeny and ecological radiation of Tyrannidae (Aves, Passeriformes). Zool. Scripta 37, 315-335.
Roy, M. S., Torres-Mura, J. C., Hertel, F., 1999. Molecular phylogeny and evolutionary history of the tit-tyrants (Aves: Tyrannidae). Mol. Phylogen. Evol. 11, 67-76.
Smith, W. J., 1971. Behavioral characteristics of serpophaginine tyrannids. Condor 73, 259-286.
Traylor, M. A. Jr., 1977. A classification of the tyrant flycatchers Tyrannidae. Bull. Mus. Comparative Zool. 148, 129-184.
Shane G. DuBay and Christopher C. Witt, June 2012
Comments from Stiles: ÒYES. Genetic, morphological and distributional evidence all favor recognizing Uromyias.Ó
Comments from Pacheco: ÒYES. Published evidence supports the proposal.Ó
Comments from Robbins: ÒYES, although this comes down to subjective opinion, given that these (2 Uromyias taxa + Anairetes sensu stricto) are sister clades, one could treat them as members of the same genus.Ó
Comments from Cadena: ÒNO. The Dubay and Witt paper reports on a very nice study, and I have no issue with the data and analyses. But I agree with Mark that, considering that "Uromyias" and "Anairetes" are sister groups, one could still treat them in a single genus. Doing that, phylogenetic information is retained in classification by revealing a close affinity between all the species in these two clades. Having them as separate genera conveys information on relationships at a shallower phylogenetic level. What is better? One could argue either way. Considering morphology and ecology, I realize there might be good reasons for separating these species in two genera, but I tend to prefer stability over taxonomic changes in cases like this (unless both alternative classifications have been widely used off late, such as in the Diglossa/Diglossopis case). In other words, given the new data, changing the classification is not necessary so I say we keep it as is.Ó
Comments from Stotz: ÒNO. This is a case very similar to Pipromorpha and Mionectes, where I voted no as well. If anything these two sister taxa seem less well-differentiated than Pipromorpha and Mionectes. The lump dates from the same time period, so in the absence of a compelling reason to split Uromyias from Anairetes, I have to support the status quo.Ó
Comments from Zimmer: ÒYES. As pointed out by Mark and Daniel, this really comes down to a matter of taste, as to how broadly or narrowly one likes to define genera. In this particular case, I donÕt have really strong feelings either way. However, as a general rule, I prefer more internally cohesive, narrowly defined genera, and, on both morphological and distributional grounds, Anairetes sensu stricto would seem to be sufficiently distinct from agilis and agraphia to warrant generic separation. So, a somewhat tepid YESÓ.
Comments from PŽrez-Em‡n: ÒYES. This is another proposal when we face the issue of what to group within a genus and, as Daniel pointed out, what level of information we want to include at this taxonomic level. We donÕt have yet a solution to this issue and it comes down to be a matter of taste. In this case, previous to the Roy et al (1999) paper, most people agreed with keeping both Uromyias and Anairetes as separate genera based on bill, tail, and cranial morphology. Lumping these taxa into Anairetes was a response to Roy et al.Õs results, an important study, but which could be considered an earlier version of the molecular study done by DuBay and Witt (2012). As a more complete study with similar characters does not support previous results (which were not conclusive), I rather prefer to split these taxa into two genera than lumping somehow divergent taxa into one genus (as a set aside thought, this course of action might be leading us to consider a genus as a group of taxa with few morphological and ecological differences compared to a larger and more diverse inclusive genus).Ó
Comments from Jaramillo: ÒYES. I never understood why the two were lumped, Uromyias looks and acts quite unlike Anairetes. Minus the crest, Uromyias is as similar to Anairetes as Serpophaga to me, so resurrecting Uromyias sounds good to me.Ó
Comments from Nores: ÒYES. Although the reasons for joining or not joining these species are both valid, I prefer to separate them for three reasons: 1) they are quite different morphologically, 2) they dwell in very different habitats, and 3) new phylogenetic results suggest that morphological and ecological differences between the two clades warrant classification in separate genera.Ó