Proposal (188) to South American Classification Committee

Treat Amazilia rondoniae as a subspecies of Amazilia versicolor

Amazilia rondoniae was described recently by Ruschi (1982) as a new species, based on distinctive plumage characters and sympatry with A. v. nitidifrons. Its type locality is Puerto Velho, on the Río Madeira in Rondonia and its distribution was cited as being limited to the state of Rondonia, in extreme W Brazil, with specimens of A. v. nitidifrons having been taken both NW and E of this area. It was accepted as a good species by Schuchmann (1999) based on Ruschi's description. I have not been able to obtain a copy of this description, but a subsequent publication (Grantsau 1988) includes paintings of two specimens from the type locality in the private collection of Dr. C. A. Campos Seabra in Río de Janeiro, although he also mentions having consulted the Museu "Mello-Leitao" where the two cotypes of rondoniae upon which Ruschi based his description are housed. These specimens were examined by Vielliard (1994) in the course of his preparation of a catalogue of the hummingbird specimens in this collection, in which he also makes numerous corrections of misidentified specimens, including several of related Amazilia species, that bear upon the supposed sympatry of rondoniae and forms of versicolor. 

The diagnostic character of rondoniae is the glittering blue crown, malar area, sides of the neck and throat (cf. also Schuchmann 1999). In milleri the crown, malar area and sides of the neck are glittering green, the center of the throat is white; the neighboring form of versicolor to the E, nitidifrons, is similar in pattern to rondoniae but with the blue replaced by glittering green. The paintings by Grantsau (cited by Vielliard) of two male specimens from the type locality show considerable variation: in one, there is considerable green in the malar area and on the hind crown; in neither is the throat wholly blue, but shows an admixture of white, possibly in the bases of the feathers, and variation also in the amount of dull brownish-green on the sides; in both, the distinctive pattern of the tail is matched by specimens illustrated of several races of versicolor including milleri and nitidifrons. Grantsau (1988) concluded that rondoniae was best treated as a distinctive subspecies of versicolor. Grantsau also illustrated the locations of specimens he cited on maps of Brazil showing state boundaries (no. 12 treats versicolor and rondoniae, as well as fimbriata and lactea; but it places the location of rondoniae in the nearby state of Acre! According to this map, nitidifrons occurs only to the E and S of rondoniae).

Vielliard (1994) reached the same taxonomic conclusion, stating that the overall pattern of rondoniae was "obviously that of the species versicolor and, except for the blue reflections of the head, strongly resembles that of nitidifrons. It represents a clear subspecies, whose recent discovery and apparently restricted distribution are remarkable". He also corrected misidentifications of several specimens that make it clear that milleri occurs only N of the Río Amazonas (Solimões) and nitidifrons to the S and E. A specimen of "nitidifrons" from S Venezuela was re-identified as milleri and another from Benjamin Constant on the N bank of the Amazon, as A. fimbriata laeta. This effectively eliminates the supposed sympatry of rondoniae with forms of versicolor (no form of which has been taken at or near Puerto Velho).

Regarding the blue vs. green of rondoniae, I am rather doubtful that this is a species-level character, given the considerable variation in crown color in a series of males of milleri from E Colombia in the ICN collection here: crowns range from nearly wholly blue to nearly wholly green with only a trace of blue anteriorly; also, the specimens illustrated by Grantsau show variation in this feature. I might mention that a number of Ruschi's taxonomic conclusions have been questioned recently (see LeCroy & Vuilleumier 1992, Vuilleumier et al. 1992, Vielliard 1994); Edwin O. Willis and Yoshika Oniki, who have done much work on Brazilian hummingbirds, stated that "lamentably, without independent confirmation, a considerable portion of Ruschi's work lacks a firm foundation" (Willis and Oniki, in litt.).

In sum, I recommend accepting the conclusion of Vielliard and Grantsau, to consider rondoniae as a subspecies of versicolor rather than a separate species; that is, a YES on this proposal.

REFERENCIAS

Grantsau, R. 1988. Os Beija-Flores do Brasil. Espressao e Cultura, Río de Janeiro.

Ruschi, A. 1982. Bol. Mus. M. Leitao 100:1-2.

LeCroy, M. & F. Vuilleumier 1992. Guidelines for the description of new species in ornithology. Bull. BOC 112A:191-198.

Schuchmann, K.-L. 1999. HBW, vol. 5.

Vielliard, J. 1994. Catálogo dos troquiílideos do Museu de Biologia Mello-Leitao. Ministerio de Cultura, Instituto Brasileiro do Patrimonio Cultural.

Vuilleumier, F., M. LeCroy & E. Mayr. 1992. New species of birds described from 1989 through 1990. Bull. BOC 112A: 267-310.

F. Gary Stiles, November 2005

________________________________________________________________________________________

Comments from Remsen: "YES. Burden-of-proof falls now on treatment of rondoniae as a species-level taxon."

Comments from Zimmer: "YES". I agree with Gary that the blue color to the head of rondoniae is not impressive as a species-level character in this particular group. The versicolor group is in serious need of some real work. In southeastern Brazil alone, there are two distinct types of Amazilia that are passing under the guise of nominate versicolor - white-throated birds and turquoise-throated birds. HBW Vol. 5 treats these as morphs, with white-throated birds being coastal breeders and green-throated birds representing an inland population, with intermediates floating around in between. However, it is common in some areas to see the two forms side-by-side at feeding stations, and off the top of my head, I can't think of any parallels to this situation among other hummingbirds. If these two very different looking birds are just morphs of the same species, then I have no confidence that rondoniae is a valid species-level taxon in the same group."

Comments from Robbins: "YES. As usual, Gary has done a good job of laying things out (as I have not consulted the Grantsau or Vielliard papers), so that the decision is straightforward."

Comments from Silva: "YES, although Amazilia versicolor requires a lot of taxonomic work."

Comments from Pacheco: "NO. Unfortunately, I have more intransigent opinion in view of the several serious inconsistencies (and even fraud evidences) of the works of Augusto Ruschi (e.g. Pacheco 1995, Pacheco & Bauer 2001). I do not recommend consider Amazilia rondoniae (with "two designated holotypes Ruschi 1982) neither as species nor subspecies, but simply as taxon inquirenda."

References:

Pacheco, J. F. (1995) O Brasil perde cinco espécies de aves! Uma análise crítica dos registros de Ruschi para alguns beija-flores das fronteiras setentrionais brasileiras. Atualidades Orn., 66:7.

Pacheco, J. F. e C. Bauer (2001) A lista de aves do Espírito Santo de Augusto Ruschi (1953): uma avaliação crítica. Pp. 261-278. Em: Ornitologia e conservação: da ciência às estratégias (J. L. B. Albuquerque, J. F. Cândido Jr., F. C. Straube e A. L. Roos, eds.) Tubarão: Editora Unisul.

[Fernando can provide pdfs of these papers for those interested.]

Comments from Jaramillo: "YES - The standing of this taxon seems quite weak."

Comments from Nores: "YES. Yo también considero que las diferencias son sólo a nivel de subespecie. Incluso otra subespecie, hollandi tiene para mi más diferencias que rondoniae de versicolor. Yo coincido con Pacheco que el trabajo de Ruschi tiene muchas serias inconsistencias y aún fraudes, pero en este caso los ejemplares existen y consecuentemente el taxón es válido."