Proposal (341) to South American Classification Committee

Reinstate Pipromorpha for the rufous Mionectes (2)

Summary:

In Proposal 202, six SACC committee members voted in favour of reinstating the genus Pipromorpha for the rufous Mionectes species, M. oleagineus, M. macconnelli and M. rufiventris and four against. Being just one vote short, the proposal failed to pass. A recently published molecular study mandates reconsideration of the previous proposal. Unpublished nesting data referred to in Proposal 202 have also now been published.

Nesting Data:

Greeney et al. (2006) have published a summary of nesting behaviour in Mionectes, finding Mionectes and Pipromorpha species to have similarly structured nests. This finding was made available to the SACC during the course of the Proposal 202 through a comment from Nacho Areta. The relevant data have now been published.

New Molecular Data:

Miller et al. (2008), with broad geographical sampling of all Mionectes species and many populations of such species that are candidates for species or subspecies rank, studied both mitochondrial and nuclear DNA variation within the genus. Consistent with the morphological and vocal analysis presented in Proposal 202, they found that Mionectes is monophyletic, but that the genus comprises two well-defined clades: (i) the green, quiet, montane core Mionectes; and (ii) the rufous, louder, lowland species often referred to as Pipromorpha. The two clades each had 100% posterior probability (100% ML bootstrap) support. Possible new species limits arising from that paper (and from vocal considerations) are not considered in this proposal.

Miller et al. (2008) noted that a monophyletic Mionectes was consistent with current classifications. However, referring to the two clades within the genus, they stated that: "This split is old: average model-correlated cyt b distance between the montane and lowland Mionectes clades was 14.3%, dating to ca 7 Myr ago." The deep nature of this split was also supported by nuclear DNA analysis. In their discussion, dating the split to the Late Miocene, Miller et al. (2008) stated that: "Within the genus, both mitochondrial and nuclear gene sequences identified a basal phylogenetic split between montane and lowland Mionectes species, providing support for earlier classifications that placed the three lowland species in the genus Pipromorpha (e.g. Traylor 1977)."

Ohlson et al. (2008)'s recent paper provides a useful yardstick to compare the depth of this split to that of other Tyrannid genera. Many generic divisions in Tyrannidae apparently occurred in the Late Miocene and several occurred more recently than that (Pliocene/Pleistocene) (see their Figure 2). As sampling of species (as opposed to genera) was not extensive in the Ohlson et al. study, it is not clear yet whether many species in other genera diverged at such an early point in time - the only examples in this study being in one of two Myiophobus clades and Myiopagis. It is, however, clear that recognition of Pipromorpha would be consistent with Tyrannidae genus limits based on molecular clock data. Recognition of Pipromorpha is also considered consistent with subjective morphological considerations of generic limits in Tyrannidae in the view of the author of this proposal and a majority of SACC committee members.

Latin Grammar:

I understand that Mionectes oleagineus would become Pipromorpha oleaginea if this proposal passes.

Recommendation:

SACC has shown itself to be open to overturning the result of a previous proposal following the publication of new data or new considerations (e.g. "Twistwing" for Cnipodectes; Dysithamnus leucostictus anyone?). One Committee Member suggested that he would change his vote on Proposal 202 if the division between Pipromorpha and Mionectes proved to be as deep a one as that in another Tyrannid genera - a point which the Miller et al. (2008) emphatically demonstrate to be the case.

Unlike other generic limit proposals considered recently by SACC, this split is not mandated by paraphyly. Mionectes and Pipromorpha form a monophyletic group and generic limits are therefore a matter for taste, consistency with other genera and nomenclatural stability. Those Committee members that voted against Proposal 202 based on stability concerns may therefore wish to maintain the current status quo (which has 25-30 years' history behind it). For others of the four original "no" voters, molecular results may now tip the balance in favour of recognition of Pipromorpha. I encourage a YES vote, again.

References not in Proposal 202:

Greeney, H.F., Dingle, C., Dobbs, R.C. & Martin, P.R. 2006. Natural history of the Streak-necked Flycatcher Mionectes striaticollis in northeastern Ecuador. Cotinga 25: 59-64.

Miller, M.J., Bermingham, E., Klicka, J., Escalante, P., Raposo do Amaral, F., Weir, J.T. & Winker, K. 2008. Out of Amazonia again and again: episodic crossing of the Andes promotes diversification in a lowland forest flycatcher. Proc. R. Soc. B. 275: 1133-1142.

Ohlson, J., Fjeldså, J. & Ericson, P.G.P. 2008. Tyrant flycatchers coming out in the open: phylogeny and ecological radiation of Tyrannidae (Aves, Passeriformes). Zoologica Scripta 1-21.

Thomas Donegan, April 2008

______________________________________________________________________________________________________

Comments from Remsen: "YES. I stand by my comments in #202, and as summarized by Donegan, genetic distance data roughly consistent with separation at genus level."

Comments from Cadena: "NO. Being a relative newcomer to Neotropical ornithology, I have no particular appreciation for the existence of the name Pipromorpha and was never affected "emotionally" by the merger of two genera. Aside from being a matter of taste as mentioned by others in earlier discussions, I think we ought to consider what would we gain (and lose) by resurrecting Pipromorpha. I would argue that by splitting Mionectes in two separate genera, linear classifications lose phylogenetic information: the fact that species of Mionectes and Pipromorpha form a clade that sets them apart from all other species in the Tyrannidae would disappear. A counter-argument could be that we would actually gain phylogenetic information by having a classification that highlights that species in Pipromorpha are more closely related to each other than they are to Mionectes and vice-versa. At what level do we want to retain phylogenetic information is indeed a matter of taste, but if we give equal weight to these two alternative positions, we are left with a tie. And given a tie, for the sake of stability, I'd say we should retain the status quo. Also, Pipromorpha and Mionectes could be recognized at the subgenus level to result in a classification that conveys information about relationships (is there a supragenus category?). Another issue is the fact that in the previous discussion a dominant theme was whether the division between Pipromorpha and Mionectes is a deep or shallow one, and some people mentioned that should that division turn out to be deep (i.e. similarly deep as divisions between other Tyrannid sister genera), the recognition of two genera would be warranted. I appreciate the idea of trying to have consistent definitions for genera based on degree of genetic differentiation (i.e. divergence times), but let's face it, avian genera in all families are very heterogeneous in their ages. I believe that unless we want to change all of the genus-level classification of birds to give this rank only to clades of particular ages (I guess that's not want we want!), we should tinker with genus-level decisions only if the genera recognized by traditional taxonomy turn out not to be monophyletic. I obviously do not question the results of Miller et al., which appear very solid, but I simply do not see them as sufficient to justify a change in the status quo; let's keep the classification stable unless it is definitively necessary to change."

Comments from Schulenberg: "NO. I agree with the arguments made earlier (Proposal 202) by Gary Stiles, Nacho Areta, Doug Stotz, and Mark Robbins, and in connection with the current proposal by Daniel Cadena. The differences between the two clades are not that great. In that connection, I have little experience with singing Mionectes striaticollis, which perhaps is very seasonal in its breeding (at least in Peru), but my recollection is that it too forms exploded leks. (Anyone else have similar experiences with striaticollis?) There's no compelling reason to resurrect Pipromorpha, other than "personal preference." My personal preference is not to make changes unless we need to."

Comments from Robbins: "NO. Daniel presents good rationale for maintaining the status quo."

Comments from Nores: "YES, por la nueva evidencia molecular aportada por Miller et al. (2008) y por las razones dadas en la propuesta 202: "aunque como puede deducirse de las diferentes opiniones emitidas al respecto, las diferencias no son tan marcadas como para pensar en algo definitivo. Yo veo ambos grupos bastante diferentes, particularmente en lo que se refiere al color. El color predominantemente rufo en el cuerpo es poco frecuente en Tyrannidae, y con pocas excepciones, es una característica genérica. Es el caso de Neopipo, Terenotriccus, Miyophobus, Pyrrhomyias, Myiotheretes, Cnemarchus, Hirundinea, Neoxolmis Cassiornis y Attila. En los pocos casos que no es así, Myiarchus semirufus, Pachyramphus castaneus, Rhytipterna holerythra, no difieren marcadamente del resto, son sólo más rufos. En Mionectes y Pipromorpha los colores son muy diferentes: los primeros amarillos o verdosos muy rayados y los otros rufos muy poco rayados. Además, es notable el parche blanco detrás del ojo presente en Mionectes y ausente en Pipromorpha".

Comments from Jaramillo: "YES - see comments in previous incarnation of this issue."

Comments from Stiles: "NO. In addition to my previous arguments on this point, I would question what we gain from this split vs. what we lose. Nest architecture tends to be a character most useful at the generic level in several suboscine groups and the Mionectes nest type is distinctive, only Leptopogon coming close. In the latter genus we have rufous-highland vs. yellowish-lowland species (referring to underpart coloration), which I don't see as any greater than a highland-streaky vs. lowland-rufous difference. All the Mionectes-Pipromorpha species share morphological and behavioral features (no rictal bristles, a similar "jizz" (pardon my British), leks etc. so I fail to see anything useful in splitting them."

Comments from Stotz: "NO. I am one of those people who at least implicitly suggested that if genetic data indicated that Mionectes and Pipromorpha was a reasonably deep split I would reconsider my vote on resplitting the two genera. Having reconsidered it, I think in the end I will maintain my vote against splitting Mionectes. I just don't see the benefit in this split. Because Pipromorpha and Mionectes are sister to one another, we don't have to make this split, and the benefit of recognizing a genus Mionectes with only 2 species seems pretty limited."

Comments from Zimmer: "YES. See my earlier comments on this one. Also, the recently published data on genetic distance adds weight to my earlier convictions for separating these at the generic level. Several members have expressed the desire to not make a destabilizing change in part because all that would be accomplished is creating a couple of genera with two species each. New information on vocalizations, displays and ecology of various populations of "Pipromorpha" make it likely that we will be seeing changes to recognized species-limits in this group sooner rather than later. If so, it means there will be one or more additional members of the very cohesive "rufous group" of Mionectes, which, to my thinking, would make the generic split even more desirable."