Proposal (200) to South American Classification Committee

New genus name for "Cacicus" solitarius

The recent molecular phylogeny has changed our view of the evolutionary tree of the caciques and oropendolas (Price & Lanyon 2002, 2004). Their phylogenetic tree has remained similar with the incorporation of new data, and therefore appears robust. Further sequencing of taxa classified as subspecies, the use of nuclear genes, and additional non-molecular characters, may help to resolve a few problematic nodes in the tree.

The latest phylogenetic tree of the group (Price & Lanyon 2004), as well as Hellmayr (1937), should be consulted to understand this proposal. I'm using in this proposal data from my forthcoming chapter on the Icteridae for the HBW (e.g. data on body mass, size dimorphism, vocalizations, nests, etc.).

1) The true caciques and oropendolas together form a monophyletic clade in the molecular phylogeny. An important shared ancestral character in the group is the weaving of pendant, bag-shaped nests, of various plant or fungal fibers. Some cacique species nest solitarily and are monogamous, but all oropendolas are colonial and polygynous. The simplest nomenclatorial solution would be to use Cacicus Lacépède 1799 for all the true caciques and oropendolas, but this procedure would create an avian genus of Linnean scope and size. Botanists accept such genera, but ornithologists usually do not. Notice that the type species for Cacicus is the Red-rumped Cacique Cacicus haemorrhous, a fact that creates some nomenclatorial problems.

2) The basal or ancestral position in the cacique-oropendola clade, according to the molecular phylogeny, is occupied by the Yellow-winged Cacique, perhaps better known as Mexican Cacique (Cacicus melanicterus). If alternative 1) is discarded, the inclusion of melanicterus in Cacicus would invalidate the monophyly of Cacicus. Although the species is geographically outside the SACC committee, I propose that the genus Cassiculus Swainson 1827 should be resurrected for Cacicus melanicterus.

On purely biogeographic grounds melanicterus is isolated from the remaining caciques by most of Central America, but two oropendolas occur in Mexico.

3) The molecular phylogeny also shows that the genus Cacicus (sensu lato) would not be monophyletic, because another species in particular, the Solitary Cacique (Cacicus solitarius), is closer to the subclade containing all the Psarocolius oropendolas. This result has not changed in successive versions of the phylogeny (Price & Lanyon 2002, 2004). Some additional characters are shared by Cacicus solitarius and the species of Psarocolius: their nest has an elongated, cylindrical shape, and is similarly woven with non-fungal, brownish plant fibers. Along the rivers of the eastern Chaco of Argentina and Paraguay, the nests of Solitary Caciques look more similar in size, shape and texture to nearby nests of Crested Oropendolas (Psarocolius decumanus), than to the blackish fungal nests of the sympatric Golden-winged Caciques (Cacicus chrysopterus) (pers. obs.). See for nesting materials of Southern Cone cacique taxa the paper by Chatellenaz and Ferraro (2000). The simplest solution would be to transfer Cacicus solitarius to the genus Psarocolius. Solitary Caciques have a plain black plumage, entirely lacking the conspicuous yellow tail feathers so typical of Psarocolius. Besides, Solitary Caciques differ in important ways from Psarocolius oropendolas. Solitary Caciques usually forage in dense lower vegetation, whereas oropendolas rather favor the canopy. At difference of oropendolas, Solitary Caciques do not flock, nest solitarily and form monogamous pairs. Their sexual size dimorphism is slight, the ratio male mass/female mass being 1.13. The size dimorphism scores for oropendolas range from 1.64 to 2.12. The vocalizations of Solitary Caciques are shared by both sexes, and include mimicry of several bird species. Songs of oropendola species are reported only (or mostly) for males, and occur during elaborate and rather stereotyped courtship displays (Price & Lanyon 2002). Delimitation of genera is quite subjective, but I believe that the Solitary Cacique is too different from the oropendolas to be included in Psarocolius.

4) If neither Psarocolius nor Cacicus is appropriate for Solitary Caciques, another generic name is needed. In the past Solitary Caciques were included in the genus Archiplanus Cabanis 1851. However, the type species of Archiplanus is Golden-winged Cacique Cassicus albirostris Vieillot 1816, based on Azara's no.59, 'yapú negro y amarillo' (Azara 1802). The current scientific name of Golden-winged Cacique is Cacicus chrysopterus. In the molecular phylogenetic tree this species is located within the main cacique clade, distant from the Psarocolius clade containing the Solitary Cacique (Price & Lanyon 2002, 2004). The generic name Archiplanus could be resurrected for this and other Cacicus species, but not for the Solitary Cacique.

According to Hellmayr (1937) no other generic names are available for Solitary Caciques. Therefore, in my recent paper (Fraga 2005) I proposed the new generic name Procacicus for the Solitary Cacique. I invented a simple name, resembling the old one, to facilitate the transition. Spanish and Portuguese speakers, be ornithologists or birders, commonly use scientific names in everyday conversation.

5) For the time being I will not suggest further nomenclatural changes in the genus Cacicus, although a glimpse at the molecular tree suggests that further changes will be needed. An ongoing extensive project on the molecular phylogeny of the Emberizinae, using mitochondrial DNA and nuclear genes, would probably clarify some of the problems. It would be ideal to use also additional characters, like displays, voices and nest structure. I would expect the nearest relatives of Golden-winged Caciques to weave dark nests of fungal fibers, and that seems to be the case.

References

Chatellenaz, M. L., and L. I. Ferraro. 2000. Materiales vegetales y fúngicos en nidos de aves del Noreste Argentino y Paraguay. Facena 16:103-119 (Corrientes, Argentina).

Hellmayr, C. E. 1937. Catalogue of the birds of the Americas and the adjacent islands. Vol. 12. Field Museum of Natural History, Chicago, USA.

Fraga, R. M. 2005. A new generic name for the Solitary Cacique. Bull. Brit. Orn. Club 125: 286-287.

Price, J. J. and S. M. Lanyon. 2002. Reconstructing the evolution of complex bird song in the oropendolas. Evolution 56: 1514-1529.

Price, J. J., and S. M. Lanyon. 2004. Patterns of song evolution and sexual selection in the oropendolas and caciques. Behavioral Ecology 15:485-497

Rosendo Fraga, January 2006

Comments from Stiles: "YES. Genetic data mandate a new generic name; morphological and behavioral data emphasize the "solitary" position of solitarius. The only acceptable alternative, placing it as congeneric with the oropendolas, is offensive to my conception of what a genus should be as well as my aesthetic sense (whatever that implies)."

Comments from Zimmer: "YES. Including the Solitary Cacique within Psarocolius goes against the grain on way too many counts, and this seems like a logical alternative."

Comments from Robbins: "NO. For now, I vote "no" based on Scott Lanyon's comments below:"

"I do think that the classification of Solitary Cacique needs to be changed, but I would not recommend doing so now. Part of the problem is that our 2004 Behavioral Ecology paper shows that Cacicus melanicterus is even more distantly related than is Cacicus solitarius. If you elevate Cacicus solitarius to its own genus based on this study, then you must also elevate Cacicus melanicterus to its own genus. Based on Figure 2 (see attached jpg file) the simplest classification change would be to:

1) lump all oropendolas (except O. latirostris and P. oseryi) AND Cacicus solitarius into the genus Psarocolius,

2) lump all caciques (except C. solitarius and C. melanicterus) AND O. latirostris and P. oseryi into the genus Cacicus. And 3) place C. melanicterus in its own genus as the sister taxon to Psarocolius and Cacicus.

"That said, I think it is premature to make such a change because the placement of C. solitarius does not have especially strong support and the monophyly of the cacique clade (proposal #2 above) is not consistently supported by this data set. Given this level of uncertainty, we shouldn't be changing nomenclature."

Scott M. Lanyon, May 2006

Comments from Silva: "NO, based on Scott's comments. I think we should wait a more comprehensive phylogeny to make decision."

Comments from Nores: "YES. Los fundamentos dados por Fraga para crear un nuevo género para Cacicus solitarius son muy convincentes, especialmente cuando se ve que la otra posibilidad sería poner a esta especie junto con las oropéndolas en el género Psarocolius."

Comments from Pacheco: "YES. Não apenas a partir das evidências moleculares; mas, sobretudo, considerando a morfologia e o comportamento, a alocação de solitarius, em gênero monotípico parece ser a mais apropriada."

Comments from Jaramillo: "YES - Even if the molecular data sort out slightly differently next time an analysis of the group is done, the available evidence now, and not only from genetic datasets is that solitarius should be given its own genus. If melanicterus needs a different genus, well, that is a separate question and needs a separate proposal. I think we should vote on what we have now, if new evidence arrives 2-3 years down the line that suggests a change in this decision, that is fine, but unless a publication is imminent (and maybe it is?) then I see no reason in delaying the creation of this new genus."

Comments from Schulenberg: "NO. Clearly the nomenclature is a mess, and does not reflect phylogeny. There's an argument to be made for doing what we can when we see the opportunity, but the taxonomy of Cacicus is so far removed from the phylogeny that I would prefer to deal with a series of votes that collectively address all of the issues. Otherwise, down the line we may lump a lot of taxa (including solitarius) back into Cacicus - which I'm not sure I'd recommend, but as Scott noted it's an option - or we'll need several *other* generic names for different subunits of Cacicus. I'm suggesting that which direction to go is something we should confront all at once, not on a piecemeal basis."