Proposal (27) to South American Classification Committee

Elevate subspecies huancavelicae and arequipae of Asthenes dorbignyi to species rank

Effect on South American CL: This proposal would split off two species from within our broad Asthenes dorbignyi: A. huancavelicae and A. arequipae.

Background: Asthenes dorbignyi (Creamy-breasted Canastero) as constituted in our current classification, which essentially follows the traditional Peters/Meyer de Schauensee classification, consists of at least three distinctive populations that almost certainly warrant species rank. Although Fjeldså & Krabbe (1990) did not accord each of the three official species rank, they demarcated them with separate English names: "Pale-tailed Canastero" of central Peru for huancavelicae, usheri, and an undescribed subspecies; "Dark-winged Canastero" of SW Peru for arequipae and an undescribed subspecies; and "Rusty-vented Canastero" for the nominate subspecies and consobrina of Bolivia and N Argentina. They provided qualitative descriptions of the voices of usheri, arequipae, and nominate dorbignyi and point out that they differ substantially. The three groups differ strongly in plumage, especially tail and wing color and presence or absence of gular patch. They stated that there were no signs of intergradation between any of the component taxa, although all are thought to be allopatric, with no certain cases of parapatry. Collar et al. (1992), based on a manuscript by Fjeldså and Schulenberg described as "in press" split dorbignyi into the three species as noted above, but did not relay any details of the manuscript, which in fact was not in press (fide Schulenberg). Ridgely & Tudor (1994) also split the three, citing Collar et al. for the same MS, and also described vocal and plumage differences; they did note, however, possible intermediate specimens between arequipae and consobrina.

To my knowledge, nothing further has been published.

Recommendation: Although it is almost certain that the broad species dorbignyi actually consists of three or more species, I recommend a NO on my proposal because:

(1) the Fjeldså-Schulenberg MS has never been published.

(2) the details of this complex situation do indeed need to be published; as of now, all we have is qualitative vocal descriptions to accompany the admittedly dramatic plumage differences. I worry about qualitative vocal descriptions of an unknown N of individuals and localities.

(3) Fjeldså and Krabbe (1990) list two "unnamed subspecies," one within Pale-tailed and another within Dark-winged, and perhaps I'm being too picky, but it seems these out to be described before rearranging species limits.

(4) the arequipae-consobrina situation does need more work, as noted by Ridgely & Tudor (1994). I also noticed this when doing HBW accounts and wrote: "Race arequipae like consobrina in lacking rufous wing band but larger with deeper bill; blacker lores and auriculars; darker upperparts (especially crown); larger, more conspicuous throat patch; more extensive rufous in outer rectrices; populations from Chile and W Bolivia are browner-backed and paler-faced and are probably intergrades between arequipae and either consobrina or nominate race."

Reasons to vote YES might include:

(1) two or more species are almost certainly involved, and so even if we don't get the details right, some sort of split is closer to "the truth" than current broad dorbignyi.

English names: The above names used by Fjeldså and Krabbe (1990) were also used by Ridgely & Tudor (1994), and so already have a tradition.

Lit Cit:

Collar, N. J., Gonzaga, L. P., Krabbe, N., Madroño Nieto, A., Naranjo, L. G., Parker, T. A. III & Wege, D. G. 1992. Threatened birds of the Americas. The ICBP/IUCN red data book. Third edition, part 2. International Council Bird Preservation, Cambridge, United Kingdom.

Fjeldså, J., & N. Krabbe. 1990. Birds of the high Andes.

Ridgely, R. S. & G. Tudor. 1994. The bird of South America. Vol. II. The suboscine passerines. University of Texas Press, Austin, Texas.

Van Remsen, 20 May 2003

________________________________________________________________________________________

Comments from Zimmer: "I'm torn on this one. My normal inclination is to wait until a proper analysis has been published. On the other hand, given the pronounced morphological differences, along with published descriptions (albeit qualitative) of pronounced vocal differences, I'm tempted to recognize the split along the suggested lines with the attitude (suggested by Van) that even if we don't have all of the details right, this new treatment would much more closely approximate reality than does the status quo. Retention of the suggested (Fjeldså and Krabbe) English names for the 3 populations makes sense. Record me as a somewhat reluctant "YES" on this one."

Comments from Robbins: "This is one of the toughest that we have dealt with. Because of the lack of a publication that clarifies everything, I'm leaning to vote "no" on this one.”

Comments from Schulenberg: "NO. It's all a mess. Likely more than one species involved, but which are they? It doesn't make sense to me (biogeographically) that the (unnamed) Ancash population would be conspecific with the interior valley huancavelicae and usheri. I guess it's not impossible that that would be the case, but it is not obvious that it would be either. Nothing has been published on the voice of the Ancash population; I've heard recordings from different sources that are reported to be usheri but that sound different to me (although I haven't studied these in detail); I don't know of certain vocal recordings of huancavelicae; etc. And if I remember correctly, somewhere in one of Fjeldsa's papers is a suggestion that the unnamed dark rumped, dark tailed population in Ayacucho intergrades with arequipae where these approach one another: does anyone else remember having read that? So, if two such different looking types *do* intergrade (and not having seen the specimens, I can't confirm that they do), then wouldn't we want to move a little more carefully on this?

"This would be a fascinating project for someone, someday to work through. It's been up in the air for decades and much as I'd like to see it resolved, I can afford to wait a bit longer in the hope that someone will *study* it so that we don't feel obligated to *guess* about it."

Comments from Stotz: "I am voting YES on 27, 35, and 40, and NO on the other four. I will attempt to explain my logic, and hope that others will weigh in on the philosophical issues that face us. I am in a position of being uncertain what we should do about the species status of allopatric taxa with differences but no serious analysis done. The standard we are applying is basically that which is used by the North American AOU committee--status quo in the absence of compelling published data to the for North America with a century of multiple ornithologists doing detailed species level taxonomic work, I think it makes much less sense in South America. Central America, surprise, surprise, is an intermediate case. In North America, for the most part we don't have species that are lumped or split simply because of the historical accident that somebody lumped or split them based on nothing more than a cursory examination of specimens or a thoughtless application of current tendencies in species concepts. In South America this happens all the time. We have seven such cases before us (pending proposals 27, 28, 33, 35, 37, 40, 41). Mostly because of historical accident, five of these are taxa that may contain multiple species within them, one (33, Cinclodes)has split two taxa that may be conspecific. The published data is not sufficient to make a well-reasoned argument for a particular treatment of any of these cases. But why should that cause us to choose the status quo as the appropriate treatment? In North America, the status quo typically has a significant history behind it. Given that in South America that is not true I think that we are best served at this early stage in codifying the base list in using our best judgment in answering the question. Are there multiple species here? If the answer is yes, then we should split them, and if no, lump them, even given that there are problems in assigning particular taxa within the unit to species. We need to remember that our decisions are not the last word one way or another.

"With that in mind, I vote to split Asthenes dorbignyi. It seems clear that there are multiple species here (as acknowledged even by those committee members voting no), probably more than two, but the suggested split at least clarifies on big discontinuity in the group."

Comments from Jaramillo: "YES. I am going with the position that the split is closer to the truth than keeping all of these lumped. I have seen and heard many dorbignyi and arequipae in my travels, and surely these are not the same species. Given the personal experience I have with these, I feel better splitting them with little data, rather than lumping things and assuming that they are even each other's closest relatives. I mean if we lump all of these, then why not just lump in steinbachi with this group? I feel better keeping them separate although I am well aware of the different set of assumptions, involved in this choice. There is no good choice with this one."