Proposal (564) to South American Classification Committee

 

Merge Pipile into Aburria

 

Effect on SACC: This would merge the cracid genus Pipile into Aburria

 

Background: Our current footnote summarizes the situation:

 

The genus Pipile is merged by some (Delacour & Amadon 1973, Haverschmidt & Mees 1994) into AburriaPipile is currently treated as a separate genus in most classifications. Genetic data (Pereira et al. 2002, Pereira & Baker 2004) indicate that Aburria and Pipile are sister taxa, and this is supported by morphological data (Grau et al. 2005). New genetic data (Grau et al. 2005) indicate that Aburria aburri is embedded within Pipile, thus forcing the merger of Pipile into AburriaSACC proposal to merge Pipile into Aburria did not pass. Frank-Hoeflich et al. (2007) presented morphological and genetic data to support the merger of Pipile into Aburria. Eo et al. (2009) were unable to support the monophyly of Pipile + Aburria because A. jacutinga fell outside the group. Proposal badly needed.

 

The original proposal for the merger of the two genera failed because we were not impressed with the genetic data: N=1 individual of each species, all based on blood samples from aviaries.  So, not only are the identifications of the samples based on faith and cannot be checked, but as some who commented noted, cracids hybridize regularly in captivity.  Although Pereira, Wajntal, and Grau, in their comments to SACC, defended their identifications because two or more co-workers were present when blood samples were drawn, this assumes that hybrids would be phenotypically obvious.

 

Since then, two additional studies have appeared.  Frank-Hoeflich et al. (2007) analyzed 669 bp of cyt-b and found that jacutinga was the sister to the others including aburri, corroborating earlier genetic results.  However, because they analyzed one of the same two mitochondrial genes (Grau et al. sampled cyt-b and ND-2), such a result was expected.  More important from my view is that they evidently sampled different individuals, albeit from a single aviary (Estudillo Lopez, Mexico City) and still N=1 for the Pipile species, thus increasing confidence that the results of Grau et al. (2005) are valid.  Their analysis of 151 osteological characters revealed no differences between the 5 Pipile taxa analyzed and Aburria – they appear in their Figure 2 as a monophyletic group but as a 6-way polytomy within the group.  However, in their analysis of 74 morphological and behavioral characters, Aburria is the sister to all the Pipile, and within Pipile, jacutinga is the sister to the rest … in other words supporting the conventional, traditional view based on plumage etc.  When all three data sets are combined (their Figures 3 and 4), the tree places Aburria inside Pipile, i.e., jacutinga + (aburri + ((cujubi + (((cumanensis + grayi)))))).

 

Eo et al. (2008) used a supertree approach to look at the phylogeny of Galliformes and Anseriformes in general. Supertrees combine previously published phylogenetic trees, whether based on genetic, morphological, or behavioral information, to produce an overall phylogeny based on parsimony.  Thus, the outcome is derived from existing phylogenies (in this case, all the studies mentioned above), and conflicts among studies are resolved in a complicated but objective way.  Their supertree predicted that Aburria was sister to all Pipile except jacutinga, which appeared in a polytomy with all Penelope, i.e. is not certainly the sister to the other Pipile + Aburria.

 

Discussion:  The merger of Pipile into Aburria is not a new or controversial idea – Delacour & Amadon (1973) did it.  That they are minimally sister genera is also not controversial.  If the mtDNA tree is a true species tree, not just a gene tree (a serious problem not addressed by the previous studies), then merger of Pipile into Aburria is required.  The species tree vs. gene tree problem was not addressed by those previous studies because the pitfalls of single-locus phylogenies as representing the true historical branching patterns of populations was not appreciated by most workers until recently, in ornithology at least.  (In fact, I worry about some of the decisions that we have made in the past based on a single gene tree.)  I think the possibility of the mtDNA representing a gene tree rather than a species tree is a real one in this case because (1) mtDNA is probably more likely to have these problems because of matrilineal inheritance, and (2) the traditional view of relationships strongly favors Aburria and Pipile as separate; in fact, some authors have treated jacutinga as a member of a superspecies with the other Pipile, and some authors have even treated them all as conspecific.  It’s not just the plumage similarities among Pipile vs. the much less-patterned Aburria but also (A) the habitat (mostly lowland, often near water for Pipile, including jacutinga according to HBW; vs. strictly montane for Aburria), and (B) the vocalizations.  The piping guans are named for their very high-pitched, rising song, and the limited recordings of jacutinga indicate that it shares that vocalization with the Amazonian group, at least cumanensis (by someone named Mark Robbins) and cujubi.  In contrast, as noted by Thomas Donegan in the original proposal, the voice of Aburria aburri is bizarre and unlike that of any other cracids (my wife, listening to this next to me, refuses to believe it’s an actual bird).  (Does Aburria have the ”propeller” wing display of Pipile? <check>) Although we’re used to examples of plumage characters misleading us with respect to phylogeny, I am concerned that such severe vocal differences between Aburria and Pipile may reflect a species tree better than do 2700 bp of mtDNA.

 

Nonetheless, let’s just say, hypothetically, that a DNA-sequence study with additional, nuclear loci and multiple individuals of vouchered individuals of non-aviary origin found that Aburria was indeed sister to traditional Pipile?  Then, in the subjective world of generic limits, would we follow Delacour & Amadon in merging the two?  After all, those authors were familiar with these birds and assumed that jacutinga was a Pipile <need to check discussion in D&A 1973>.  On the other hand, Delacour & Amadon’s philosophy on generic limits included merging of Nothocrax into broadly defined Crax, a treatment no one follows.  However, note that Frank-Hoeflich et al. (2007) were unable to distinguish Pipile from Aburria in their osteological analysis.

 

Recommendation:  I have none – I look forward to the discussion and will vote after reading what others have to say.  Those with stronger backgrounds in DNA sequence analysis should speak out on the chances that the published results are just a gene tree.  Those who really know these birds should speak out on the subjective issue of whether the two genera should be recognized even if jacutinga is a true Pipile.

 

Literature cited (if anyone needs pdfs, just let me know)

 

EO, S. H., O. R. P. BININDA-EMONDS, AND J. P. CARROLL.  2009.  A phylogenetic supertree of the fowls (Galloanserae, Aves).  Zoologica Scripta 38: 465–481.

FRANK-HOEFLICH, K., L. F. SILVEIRA, J. ESTUDILLO-LÓPEZ, A. M. GARCÍA-KOCH, L. ONGAY-LARIOS, AND D. PINERO. 2007. Increased taxon and character sampling reveals novel intergeneric relationships in the Cracidae (Aves: Galliformes). J. Zool. Syst. Evol. Res. 45: 242-254.

GRAU, E. T., S. L. PEREIRA, L. F. SILVEIRA, E. HÖFLING, AND A. WAJNTAL. 2005. Molecular phylogenetics and biogeography of Neotropical piping guans (Aves: Galliformes): Pipile Bonaparte, 1856 is synonym of Aburria Reichenbach, 1853. Molecular Phylogenetics & Evolution 35: 637-645.

PEREIRA, S. L., A. J. BAKER, AND A. WAJNTAL.  2002. Combined nuclear and mitochondrial DNA sequences resolve generic relationships within the Cracidae (Galliformes, Aves). Systematic Biology 51: 946-958.

 

Van Remsen, October 2012

 

_________________________________________________________________________________________________________________

 

 

Comments from Nores: “NO. Although the molecular study of Frank-Hoeflich et al. (2007) supports merging both genera, the molecular analysis of Eo et al. (2009) does not. Moreover, they are not similar in plumage and the voice of Aburria aburri is bizarre and unlike that of any other cracid (Thomas Donegan). In addition, I do not agree with sampling animals from aviaries, especially cracids which hybridize regularly in captivity.”

 

Comments from Robbins: “Ugh, once again a subjective decision. Aburria has a unique voice and plumage compared to Pipile species, but does that merit generic recognition?  Even if there are issues with using just mitochondrial data, Aburria and Pipile are obviously closely related and thus this becomes an arbitrary decision.”

 

Comments from Thomas Donegan: "The new proposal is not based in particular on much new data which was not available on the previous proposal. The Frank-Hoeflich study was cited and discussed in the last iteration, although in comments and not in the proposal itself.  Eo et al (2009) subsequently found Aburria/Pipile not to form a monophyletic group consistent with previous studies (contra Nores' comment: see their table of non-monophyly and their phylogeny showing jacutinga in a polytomy involving some Penelope among others).  However, Eo et al. (2009) is a super-tree study based on essentially same data as used in the previous studies (Frank-Hoeflich et al. and Sergio Pereira et al.'s various papers).  As a result, Laurent Raty's comments on the last proposal as regards the particular mtDNA strand which supports the paraphyly of the Aburria/Pipile group (and the other, which apparently does not so much) remain applicable.  I don't totally share others' innate distrust of aviary samples and would also note that without these, a number of these species could only be sequenced using ancient DNA. Pereira and his team have produced some really interesting results especially as to issues like higher-level taxonomy, which are less likely to be affected by inter-species hybridisation. As long as one takes into account the possibility of aviary hybridisation and hence a conservative approach to making taxonomic changes based on the data, then that should be enough said.  For such an endangered group like cracids - and of big birds which are difficult to catch, and hence only sample-able by shooting - there are few other ethical alternatives. One does sometimes find cracid body parts (e.g. casques) in farms of local people in the field and taking samples from these should perhaps be encouraged in the future such that fieldworkers can help collate materials in order to some more universally acceptable study materials. 

 

“If further studies of other DNA parts continue to support jacutinga being basal to Aburria (and the latter being basal to the rest of Pipile), then there is a real issue, not a question of taste.  As Robbins notes, the solution to the issue would then be a subjective matter.  Personally, I would think it a great shame (and Latin misnomer) to give the Piping-Guans, which all pipe and are appropriately called "Pipile", a new name (Aburria), which is onomatopoeic for a totally different bird and type of song.  Although there are no osteological differences, there are differences in bare skin/cere extent/shape, which actually unite aburri with jacutinga (distinguishing both from the rest of Pipile) and massive vocal differences between Aburria and the rest.  Subjectively, and being familiar with all three 'groups' (if not all the Pipile taxa) in the field, I'd rather see "Yacutinga jacutinga" or similar as a new monotypic genus and maintain Aburria aburri and the other Pipile where they are than lump the lot of them into Aburria.  This opinion is not based particularly on genetic distance/divergence but also on subjective, aesthetic considerations.  But similar to Pearman on another live proposal, I am not keen on making available what would be a controversial a new genus name. For the avoidance of doubt in the event that these interesting discussions are published one day, this paragraph is not intended to constitute a genus description."

 

Comments from Stiles: “NO.  I´d rather see jacutinga in its own genus than lump Pipile into Aburria, which to me is a very different bird.  The problem here may be that with only data for mtDNA, we may be seeing a gene tree rather than a species tree, so we really need some nuclear data to better sort this problem out.”

 

Comments from Pacheco: “NO.  Undoubtedly, a subjective matter. Besides the glaring difference vocal between Aburria and Pipile, Wattled Guan apparently does not produce sounds with their wings (See Rios et al. 2005 at: http://www.galliformes-sg.org/cracids/bull21.pdf).

 

Comments from Zimmer: “NO.  I’m having a real hard time buying that Aburria is more closely related to all other Pipile than these are to P. jacutinga. I can certainly accept the latter as basal to other Pipile, but aside from differences in the extent of bare skin on the face and cere, everything about jacutinga screams Pipile to me, including its voice, its behavior, its overall plumage pattern and morphology, and its altitudinal distribution (predominantly lowland, marginally into lower foothills of the Atlantic Forest, i.e. < 1000 m), and its preference for riverine forest.  On the other hand, Aburria, with its completely divergent voice, even more extensively feathered head (relative to jacutinga versus all other Pipile), relatively unpatterned plumage, lack of mechanical wing display sounds, and montane Andean distribution seems like a fairly different beast.  I could support lumping all Pipile with Aburria into a single genus over segregating jacutinga in a separate genus, but I’d rather not do either.  Pipile (including jacutinga, which is most divergent and seemingly basal to the others), without Aburria, comprises a morphologically, vocally, behaviorally and ecologically cohesive group, and I would want to see some stronger evidence before messing with it.”

 

Comments from Pérez-Emán: “NO. This proposal is basically the same as previous 277, which did not pass. Grau et al. (2005) estimated phylogenetic relationships of Neotropical piping guans using three mitochondrial genes and osteological characters, analyzed independently. Support for Aburria as sister taxon to all Pipile except jacutinga was low. Frank-Hoeflich et al. (2007)’ s study increased taxon sampling and combined molecular with osteological, integumentary and behavioral characters. However, as Laurent Raty pointed out in proposal 277, they used, for Pipile jacutinga, the same Cyt b sequence than Grau et al. (2005). Raty also indicated that he ran independent analyses by gene (using GenBank sequences used by Grau et al.) and only Cyt-b sequences were the ones supporting placement of jacutinga as sister to Aburria + rest of Pipile. Such an analysis would favor a hypothesis of results showing a gene tree more than a species tree. In fact, I also downloaded ND2 sequences from this study and ran a maximum likelihood analysis with 1000 bootstrap pseudoreplicates. Results show a moderate support (bootstrap value, bv, = 76) for a sister relationship of P. jacutinga to the rest of piping guans (which are a strongly support monophyletic group, bv = 98). Additionally, all Pipile are sister to Aburria with strong support (99). It is important to say that Frank-Hoeflich et al.’s study showed that osteological characters did not resolve relationships between Aburria and Pipile and that integumentary + behavioral characters support the idea of two different genera. The combined data phylogenetic hypothesis of these authors is potentially biased toward the Cyt b sequence information. Finally, Eo et al. (2009)’s study used previous published data but without including Frank-Hoeflich et al.’s study as it is not cited in the Supplementary Material document. Thus, I don’t think we have more information to change our views than we had when proposal 277 was evaluated. Besides, morphology, behavior, ecology and even molecular information (at least for ND2 phylogenetic hypothesis) still support keeping both Aburria and Pipile.”

 

Comments from Cadena: “NO, for the same reasons as in proposal 277; I agree with additional points made by Jorge here.”