Proposal (#318) to South American Classification Committee

Transfer Piranga, Habia, and Chlorothraupis to Cardinalidae


Effect on SACC: This would transfer three genera that we have already excised from Thraupidae to the Cardinalidae.

Background: SACC classification has already removed these three genera from their traditional home in the Thraupidae to Incertae Sedis, with the following footnotes:

2. [Piranga] There is strong generic evidence that the genus Piranga belongs in the Cardinalidae (Burns 1997, Klicka et al. 2000, Yuri and Mindell 2002, Burns et al. 2003, Klicka et al. 2007). Proposal badly needed.

5. [Habia] Genetic data (Burns 1997, Burns et al. 2002, 2003, Klicka et al. 2000, 2007) indicate that the genus Habia does not belong in the Thraupidae, but in the Cardinalidae. Proposal badly needed.

6. [Chlorothraupis] Genetic data (Burns 1997, Burns et al. 2002, 2003, Klicka et al. 2000, 2007) indicate that the genus Chlorothraupis does not belong in the Thraupidae, but in the Cardinalidae. Similarity in behavior to Habia had been noted previously by Willis (1966). Proposal badly needed. Klicka et al. (2007) found that Habia is paraphyletic with respect to Chlorothraupis, with H. rubica closer to Chlorothraupis than to H. fuscicauda+ H. gutturalis.

New information: Klicka et al. (2007) confirmed these findings with broader taxon-sampling. Their combined analysis included 102 genera of tanagers, emberizines, and cardinalines. Because the three genera above have been sampled in most of the previous genetic work, with similar results, they are treated here as a package. [Klicka et al. (2007) also found that Granatellus is embedded in Cardinalidae but Saltator is not cardinaline, and made other taxonomic recommendations, such as a merger of Habia and Chlorothraupis, but these are best dealt with in separate proposals.] The genetic sampling consisted of 2281 bp of two mitochondrial genes, ND2 and cyt-b ... a nice sample.

The critical node (#1 in their Fig. 1) that places these three genera within a group that also consists of Cardinalis, Caryothraustes, Periporphyrus, Rhodothraupis, Pheucticus, Granatellus, Cyanocompsa, Amaurospiza, Cyanoloxia, Passerina, and Spiza has strong support (100% Bayesian, 78% MP bootstrap, 92% ML bootstrap); see the MS for additional details.

Analysis and Recommendation: mtDNA is widely considered a reliable predictor of phylogeny at these levels of taxonomy, and certainly these data sets represent the first truly scientific estimates of the phylogeny and classification of this group. As hard as it will be for some people, North Americans at least, to accept that what they think of as "tanagers" are not true tanagers, the genetic data leave no option to but to transfer them to Cardinalidae. Although Klicka et al. (2007) consistently found support for a Thraupidae-Cardinalidae sister relationship, that node does not have strong support. Therefore, merging Cardinalidae into Thraupidae cannot be just justified. Klicka et al. (2007) treated these families as well as parulids, icterids, and emberizids as tribes of Sibley & Ahlquist's massive Emberizidae, and once we get more confidence in estimating divergence times and hopefully tying family rank to absolute age of the lineage, then in fact they may all end up in one family. But that is not relevant to group monophyly and our current classification, which ranks them as families.

To comfort those who might be disturbed by such a radical change, note that Piranga shares with Pheucticus and some Passerina complex age and seasonal plumage changes and complex songs that are rare or unknown in core Thraupidae. The resemblance in plumage between Cardinalis and Habia cristata is not the coincidence that we once thought. Having placed all the cardinalines sensu Klicka et al. in a separate synoptic series section in our collection, I am impressed by the overall phenotypic similarities in plumage patterns, subtle shades, and texture; these sort of fuzzy things don't count for anything, of course, but I speculate that had earlier ornithologists not been so mesmerized by bill shape differences in their classifications and focused more on patterns and textures that the "new" Cardinalidae wouldn't be so radical.

I recommend a YES vote on this one -- we've had these genera dangling in Incertae Sedis limbo waiting for just one more data set, and those data have arrived.


KLICKA, J., K. BURNS, AND G. M. SPELLMAN. 2007. Defining a monophyletic Cardinalini: A molecular perspective. Molecular Phylogenetics and Evolution 45: 1014­1032.

(see SACC Biblio for rest).

Van Remsen (in consultation with Kevin Burns and John Klicka), November 2007


Note: This move will undoubtedly cause some major anxiety among those concerned with making English names "perfect." Rather than consider changing the names to something besides "Tanager", I recommend considering the name "tanager" to refer to an ecomorph (intermediate bill shape between warbler and finch), just like "sparrow," "grosbeak," "finch," "warbler," "chat," "flycatcher," and so, rather to refer to a taxonomic group, with "tanager" the name used for those with beak morphology intermediate between "finch" and "warbler."


Comments from Robbins: "YES, for transferring these three genera from Incertae Sedis status to the currently recognized family Cardinalidae. As an aside, I fully support changing the English names of these taxa to reflect their true relationships. Lets call all the Piranga "tanagers", Scarlet Piranga, Hepatic Piranga, etc. In my opinion, there is no difference in using that as an English name as there is in using Euphonia or a host of others. Now is the time to correct these misnomers."

Comments from Stiles: "YES. This transfer is clearly mandated by the data. I really don't have a strong opinion on the English names. I see no problem with continuing to call them tanagers in the generic sense; although Mark's idea of calling them "Pirangas" is not bad, I suspect that the rather enormous inertia of 150+ years of "tanagers" is enough to tip the balance towards conservatism in this case."

Comments from Stotz: "YES.  This has been in the works for a while, and this dataset seems to make it clear that these genera all belong in Cardinalinae.  In terms of the English names, I am in favor of leaving them be, at least for the time being.  Once everything is moved around the Thraupidae is going to be full of things not called tanagers and there will be tanagers scattered through various families.  We have grosbeaks, buntings and finches already in multiple families, and seem to have survived.  I think we can survive the name tanagers not providing much phylogenetic information."

Comments from Zimmer: "YES. The genetic data seem clear. With respect to English names, I like Mark's suggestion because it cleans up a messy situation, but I would have to agree with Doug that we are probably better off waiting on any changes given that there is bound to be more upheaval."

Comments from Schulenberg: "YES. I don't have any problems with leaving all with "tanager" as part of the English. As Van says (for once I agree with Van on English names - take note!), we long ago accepted that many other group names have no phylogenetic meaning. I'd like to see if this same approach can be accepted for ex-thraupid tanagers."

Comments from Nores: "YES. El análisis genético muestra bien esto, pero yo sería de la idea que en caso como estos en que existe un análisis confiable, como parece ser el de Klicka et al., y que todas las propuestas están basadas en ese análisis, se hiciera una sola propuesta con todos los cambios. Por ejemplo .... 319, 320, 321."