Proposal (345) to South American Classification Committee

Treat Phoenicopterus roseus as separate species from P. ruber

Effect on SACC: None, other than (1) change in English name, and (2) future distribution statements.

Background: The fifth edition (1957) AOU Check-list treated the American Flamingo, Phoenicopterus ruber, as a distinct species. In the next two editions, 1983 and 1998, the European P. roseus was merged with it. A Feb. 1978 memo in my files from Les Short provides the reason. He had checked with Phil Kahl, as apparently the committee had asked him to do. Kahl strongly supported the merger of those two forms (but not chilensis, as some have done). Kahl said "at the 1976 "Flamingo Group" [of ICBP?} meeting it was consensus that roseus is conspecific with ruber." Also the "new" Birds of the Western Palearctic merged them, as did the revised volume 1 of Peters (for which Kahl wrote that part). No biological evidence was cited.

New information: The BOU (Knox et al. 2002) and other European groups have more recently recognized Greater roseus and Caribbean ruber as distinct again. This is based on several papers summarized by Sangster (1997), as follows:

1. Plumage color is distinct; roseus is predominately white, often with a pale pinkish hue on neck; ruber is mainly pinkish orange, including head, neck, and underparts. Intensity of color varies with diet and season.

2. In roseus, the basal part of the bill, as the bare skin bordering it, is relatively dark pink; the black on the bill is less extensive-on the upper mandible black does not extend beyond the ventral curvature. The bill of ruber has a pale orange-red base and an extensive black tip, which does not extend below the nostril.

3. Studies of group displays in captivity by Studer-Thiersch indicate that at several stages of the display the taxa attain different postures of head, neck, body, and wings. Calls of roseus are short and bi-syllabic; in ruber they consist of 3 syllables and are drawn out. However, Studer-Thiersch did not separate these forms on the basis of this behavior, considering them closer to one another than to chilensis (not discussed here).

Whether the color and behavioral differences act as reproductive barriers is not really known, because of the wide geographic separation of the taxa. In a colony of roseus in France, mixed pairs of roseus and both (escaped) ruber and chilensis have been observed. Sangster mentioned several observations of hybrids of roseus and chilensis, but there is no further discussion of hybrids with ruber. At any rate, hybridization or lack thereof in such a situation would not necessarily be meaningful.

Perhaps most important here is the fact that the forms were merged with no analysis or reasons given. We often reverse such arbitrary decisions. In this case, we might catch up with the rest of the world by recognizing two species, and I so recommend.

Recommendation: I recommend returning to the English name American Flamingo, following AOU (1957) and most earlier references, as well as Gill and Wright (2006), rather than using Caribbean, as done by Sangster (1997) and Knox et al. (2002). There is a population in the Galapagos, which is American but not Caribbean.

Literature Cited

Gill and Wright. 2006. English names.

Knox, A. G., et al. 2002. Taxonomic recommendations for British birds. Ibis 144: 707-710.

Sangster, G. 1997. Species limits in flamingos, with comments on lack of consensus in taxonomy. Dutch Birding 19: 193-198.

Richard C. Banks, May 2008

Note from Remsen: This proposal recently passed NACC, and with Dick's permission, I have forwarded the same proposal to SACC. Because NACC went with American Flamingo, I see no reason not to follow if this is the predominant historical name.

These are the comments from NACC members, available from NACC website:

YES on separating the two species, NO on the common name. It seems it was an arbitrary decision to merge these taxa to begin with. The data for species status isn't great, either, but reversing the original arbitrary decision and bringing us in line with the BOU and other European groups seems to outweigh any hesitation. The proposed English name American Flamingo seems inappropriate. There are other flamingos in the Americas, so why should this one be designated the American Flamingo? I vote to keep Caribbean Flamingo.

YES. As the proposal notes, the merger appears to have been arbitrary and in this case I think it makes sense to follow the Europeans. I like American Flamingo, although Van's comments about the Galapagos population are worth considering. I guess that I would follow the conservative route and keep "Caribbean Flamingo" for now.

YES. I favor the English name of American Flamingo.

YES. It would seem that plumage color and bill color are important in reproductive isolation in flamingos, given the sympatry of the three species in the Andes and their differences in these characters. Behavioral (mating) displays are probably also important. These differences, coupled with initially having very poor reason to merge them, provide enough support to treat them as two species. I prefer American Flamingo over Caribbean, though there are other "American" flamingoes.

YES. Also Yes on the change to 'American' Flamingo, given the new species' distribution in the Galapagos (a tangent -- it is interesting that finches, mockingbirds, and flamingos all show a Caribbean -- Galapagos connection).

A strong YES. In my book I went with specific status, partly because they are vocally distinctly different. And NO on American Flamingo because of the existence of three other American flamingo species, the fact that they are primarily Caribbean, and that my impression is that Caribbean is the more entrenched name.

YES. Barely. Relevant data for split are weak, but as Dick noted, evidence for the lump is at least as weak. However, I think an English name proposal should be separate. I would vote NO on "American" because there is no reason to change the historical name for something slightly "better." Or is it better? The vast majority of the range is indeed Caribbean, the Galapagos themselves have some surprising Caribbean connections, and some have proposed that the Galapagos population of flamingos warrants species rank (which would then cause a flip-flop back to Caribbean).

YES. Go with "American" Flamingo, although I don't like it. There are at least 2 other "American" flamingos, if I recall correctly. I generally dislike adjectives like "American," or "Common", or "Northern" for names, but -- again -- go with what has been generally used, and Caribbean doesn't work well.

YES. I prefer the English name of American Flamingo.

YES. I agree on arbitrariness historically and sufficient published evidence to consider original treatment as two species just as likely. I abstain on English name.

______________________________________________________________________________________________________

Comments from Stiles: "YES, for the reasons given by Dick - it is effectively the lesser of two evils. as for the English name, I have no strong feelings - "American" would seem to have the stronger track record in the past in spite of not being the only "American" flamingo, so my lukewarm preference is for it - although I rather like "Caribbean" as emphasizing the main distribution of the species and the avoidance of having to change it if the Galapagos form were to be split - incidentally, just how serious is the idea of splitting them? Is it based upon anything more than distribution?"

Comments from Pacheco: "YES. Entendo que há boas razões -- mais que simplesmente arbitrárias -- para decidir pela separação."

Comments from Stotz: "YES. The lumping of ruber with roseus always seemed a little forced to me anyway. I prefer to use American Flamingo as the long-established name for this species. It also means that if somehow the Galapagos population were ever split Caribbean would be available for the Caribbean species as a distinctive "new" name."

Comments from Jaramillo: "YES - I was thinking of writing this proposal myself, and suggest these two belonged as separate species. They differ as much from each other as either does from Chilean Flamingo, so it was not logical to have the two New World taxa of this group separate and yet the Old World taxon be lumped with one of the NW taxa. The three seem like good species, although they do form a well defined superspecies in my opinion. I prefer the name American Flamingo as opposed to Caribbean Flamingo."