Proposal (584) to South American Classification Committee

Recognize subfamilies in the Cracidae

A. Recognize Cracidae subfamilies Cracinae and Penelopinae (II)

B. Recognize Cracidae subfamily Ortalidainae

This is essentially a re-run of Proposal 285A, to recognise Cracidae subfamilies, which ended up 5-5 in the voting last time. Several of those committee members who voted against the proposal felt it would be preferable to have a higher-level name for the chachalacas instead of placing them incertae sedis or in a subfamily to which they are not very closely related. A higher-level name is now available for Chachalacas, so this issue requires reconsideration.

Molecular studies have shown the chachalacas to be a more or less equally ancient lineage to guans and curassows. Pereira et al. (2002), using ‘molecular clock’ techniques, hypothesised the following periods (95% confidence interval) for major divisions in the Cracidae: 'core' Guans vs. other Cracids - 26.9-40.6 million years ago (Early Oligocene); Oreophasis from the remainder: 26.6-36.1 mya (Early Oligocene); and Ortalis from Curassows: 25.8-36.5 mya (Early Oligocene). These hypothesised divisions are as deep or deeper than those between some passerine and other subfamilies recognised by AOU-SACC.

A subfamily name for the chachalacas is now available (Donegan 2012). The description was not straightforward, given issues with the generic name Ortalis (originally described as "Ortalida") and the need to avoid homonymy with higher-level names for invertebrates. The tribal name Ortalidaini is now described, making the subfamily name Ortalidainae available. In the description of the tribe, this rank was expressly not a recommendation for rank vis-a-vis subfamily treatment.

Most authors on cracids over the last few decades have recognised divisions below the family level in this group. Per Donegan (2012):

"Vaurie (1968) recognised three major divisions within the family based on morphological data: the Guans and Chachalacas (Penelopini: Chamaepetes, Penelopina, Penelope, Pipile, Aburria and Ortalis), the Curassows (Cracini: Pauxi, Mitu, Nothocrax and Crax) and the Horned Guan (Oreophasini: Oreophasis). Delacour & Amadon (1973) re-lumped Oreophasini into Penelopini. Del Hoyo (1994) and del Hoyo & Motis (2004) ranked Delacour & Amadon (1973)’s tribes as sub-families."

More recently, Fjeldså (2013) recognised chachalacas as a major ancient lineage alongside others in Cracids traditionally recognised as subfamilies or tribes. Recent phylogenetic studies such as Pereira et al. (2002, 2009), Eo et al. (2009), and Frank-Hoelfich et al. (2007) all recognised cracid subfamilies. Following proposal 560, SACC does not now deal with tribes, so the options available here are subfamilies or nothing.

Part A of this proposal would be to recognise the two traditional subfamilies that occur in South America, but (following molecular studies) placing Ortalis tentatively in the Cracinae. Part B would be to treat Ortalidainae separately from the other two groups. Rejection of the proposals would be a vote not to recognise subdivisions at all within this family.

See proposal 285A and the papers cited below for a more detailed discussion of the differences between these groups.

References:

Donegan, T.M. 2012. A new group name for the Chachalacas (Aves: Cracidae: Ortalis). Conservación Colombiana 17: 41–44. http://www.proaves.org/proaves/images/RCC/Con_Col_17_41-44_Ortalidaini.pdf

Fjeldså, J., 2013. Avian classification in flux. pp. 77-146. In: Handbook of the Birds of the World, Special: New species and global index. J. del Hoyo, et al. (Eds.). Lynx Edicions, Barcelona.

Other references cited in Donegan (2012) and the previous proposal.

Thomas Donegan, August 2013

Comments from Stiles: “YES on A and B, given the nearly equivalent ages of the chachalacas vs. the other two cracid lineages according to recent genetic data; subfamily status seems appropriate.”

Comments from Nores: “NO on A and B. Although I do not disagree that subfamilies should be included, I think that it should be considered in a second stage. I do not know when we have to end this stage. The same criteria would apply to include subspecies.”

Comments from Pacheco: “YES on A and B. Considerando que as linhagens são confirmadamente divergentes. Prefiro tratá-las no nível de subfamília que em nível de tribo. Defendo agora que a adoção de subfamílias no SACC, quando bem fundamentadas, pode ser implementada casualmente. “

Comments from Zimmer: “YES on A. These lineages are ancient, and should be formally recognized. YES on B. YES. Given the relatively ancient nature of the lineage relative to other lineages of cracids, I think this is the desired course.”

Comments from Jaramillo: “Yes on both A and B, which unless I am reading it incorrectly would give us three subfamily level groups within the family.”

Comments from Remsen: “YES to both A and B. Even if the dating is only approximately correct, these three lineages are as old as many taxa ranked as families. This is surprising from the phenotype standpoint because in terms of morphology and plumage, Ortalis are superficially very similar to Penelope.

“Note that David (2014) has pointed out in Zootaxa that the formulation of the name itself is incorrect, so if the proposal does pass, we will have to deal with that – if you Google Scholar your way to the article, you can download a pdf or just ask me for a copy. From David (2014): ” Under the circumstances, the misformed tribe name "Ortalidaini" is to be corrected to Ortalisini Donegan, 2012 (Articles 32.5.3.2 and 35.4.1) with authorship, date and place of publication unchanged (Articles 32.2.2 and 33.2.2).” However, Donegan just told me that “Ortalisini” may be pre-occupied in Diptera and that ICZN is now evaluating this. I suggest we defer implementation until this sorts itself out.

Comments from Stotz: “YES. I think given the age of the groups and the fact that most people recognize subfamilies in Cracidae that turn out not to be monophyletic, I think we need to fix that.

Additional comments from Remsen: “After reading Hosner et al. (2016), I switch to a NO on this one. Hosner et al. (2016; Fig. 3) showed that although the Cracidae itself is an ancient group, all modern species likely evolved from a single shared ancestor ca. 12 MYA. Thus the lineage that contains extant taxa is not much older than most genera and in fact younger than many genera (e.g. in Furnariidae). In retrospect, this makes sense because the morphological diversity of extant cracids is remarkably conservative – the chachalacas look basically “identical” to small Penelope guans, etc. Therefore, even though there are deep divergences among the three primary lineages, they are young and morphologically very similar. Although we have no official definition of subfamily, the cracid lineages certainly don’t fit my personal one (which is dominated by lineage age as the only consistently defensible, objective criterion). Also, and this is just a technicality, extralimital Oreophasis should also be ranked as a subfamily if the other three are.”

Additional comments from Stiles: “I agree also, so NO!”

Comments from Cadena: “NO. But not because anything in particular related to clades within the Cracidae. What is a subfamily? I don't think anyone of us knows and because we have not agreed on criteria to recognize clades at this level, I do not think it is worthwhile to introduce this category at all in our classification right now. In addition, is there any particular good reason why we worry about recognizing subfamilies in some families but not in others? I don't think so and doing this in some cases and not in others in the absence of clear guidelines leads to inconsistencies in our classification. The above said, I think it would be extremely good if we could establish some criteria to recognize clades not only at the subfamily level but also at the family and order level (these ranks carry a lot of inertia but they are also defined quite arbitrarily and thus inconsistently). I think Van's ideas about considering clade age are fantastic and I absolutely loved his talk on this topic in the Colombian ornithology congress in 2007 (imagine a classification in which names convey information not only about relationships but also about clade ages!). In addition, I would argue this is a good time to think about this issue because we are now sufficiently close to having good enough time-calibrated trees to establish some sort of highly objective criteria for assessing rank to clades based on their ages. So I say lets talk about this in general, agree on criteria, and then work on particular cases.”