Proposal (143) to South American Classification Committee

Split Lesbia gouldii from Lesbia nuna

This proposal would split into two species a taxon that has been recognized as a single species in virtually all modern literature. The proposal was made by Weller & Schuchmann (2004) in a recent review of the taxonomy of the genus Lesbia, which included a discussion of the status of various forms in the genus, identification of one problematical form as a hybrid, and the description of two new subspecies.

As with several other hummingbird genera, the genus Lesbia has a history of broadening of species limits, with many forms described as species later being reduced to subspecies. In addition, a number of forms described from "Bogotá" skins have later been shown to be hybrids or perhaps artifacts, further complicating the problem. In 1918, Cory recognized six species in the genus: victoriae, including races victoriae, aequatorialis and juliae, eucharis, pallidiventris, berlepschi, nuna (including boliviana) and gouldii (including gracilis and chlorura). Peters (1945) reduced the number of species to two: victoriae (including berlepschi and, with a question, eucharis) and nuna (including the gouldii group of races). Zimmer (1951) essentially followed this arrangement, as have other authors since (Meyer de Schauensee 1966, Hilty & Brown 1986, Sibley & Monroe 1990, Schuchmann 1999, Ridgely & Greenfield 2001). In particular, Schuchmann (1999) described racial variation within nuna as "rather weak". 

In their recent publication, Weller and Schuchmann (2004) resurrected gouldii as a species-level taxon. Their reasons for so doing so are basically differences in the coloration of the tails of the males (more extensive green distally, more bronzy color basally and at the tips of the outer rectrices and more extensive pale outer webs of these feathers) and females (absence of bluish-green central areas of the inner rectrices in females), and smaller size. They state that the southern forms (restricted nuna) "show an intermediate morphological position between the northern forms (= gouldii) and L. victoriae in one or more morphometric traits (hardly unexpected if nuna is the larger form).

However, a detailed analysis of their paper raises some doubts. First, the differences in tail coloration are mostly quantitative in nature, and are equaled or exceeded among the races of victoriae (whose specific allocation is not questioned). More worrisome is the fact that their newly described race huallagae, the southernmost form in their gouldii, is intermediate in size in several measures, notably bill length (the most distinct feature they cite for gouldii); in fact, there is a general increase in size among the forms of gouldii from N to S in most features (although tail length of males appears to vary widely and in a rather mosaic fashion). More worrisome still is the fact that, according to their distribution map (their figure 2), specimens of the next race of gouldii to the north, pallidiventris, are virtually sympatric with those of huallagae … although they do not mention this fact specifically in connection with the description of the latter: one wonders whether, in view of this situation, huallagae should rather go with nuna rather than gouldii if a split should be made (although this would call into question the differences in details of tail color as specific characters). The authors reported statistical differences in several dimensions between huallagae and nuna, but apparently did not compare the former with other forms of gouldii, which appear in some cases to be equally different (in the opposite direction); in any case, the sample of huallagae was small (n = 4). Field study in the putative region of overlap between these two forms seems required. They present a detailed table of color characters of the races of gouldii but do not include those of nuna therein, as might be expected to help to justify a split; the various color characters justifying recognition of huallagae should be evaluated in this larger context. Except for a few outliers, the elevational ranges of all the forms in the nuna-gouldii group appear similar except for aureliae, a new subspecies described N of pallidiventris, which appears centered at slightly lower elevations (their Fig. 3). Taken as a whole in the context of the genus, I do not think that the evidence for splitting gouldii at the species level is consistent or convincing as it stands. Therefore, I recommend a NO vote on this proposal, at least until satisfactory resolution of several doubtful points is forthcoming.

Literature Cited:

Cory 1918.

Peters 1945.

Meyer de Schauensee 1966.

Hilty & Brown 1986.

Sibley & Monroe 1990.

Schuchmann 1999.

Ridgely & Greenfield 2001.

Weller, A. A. & K-L. Schuchmann 2004. Biogeographic and taxonomic revision of the trainbearers Lesbia (Trochilidae) with the description of two new subspecies. Ornithol. Anz. 43:115-136.

ZIMMER, J. 1951. Studies of Peruvian birds, No. 61. The genera Aglaeactis, Lafresnaya, Pterophanes, Boissonneaua, Heliangelus, Eriocnemis, Haplophaedia, Ocreatus, and Lesbia. American Museum Novitates 1540: 1-55.

Gary Stiles, December 2004

_____________________________________________________________________________________

Comments from Remsen: "NO. Evidence for treating gouldii as a species-level taxon is weak, as outlined by Gary."

Comments from Pacheco: "NO. As inconsistências apontadas por Stiles ao trabalho de Weller & Schuchmann (2004), convenceram-me de que tal proposta de desmembramento não é, ainda, aceitável."

Comments from Jaramillo: "NO.  Evidence sounds weak for the split, in fact the evidence as I read from Gary's summary suggests a cline."

Comments from Nores: "NO; como en los casos anteriores las diferencias son a mi modo de ver sólo subespecíficas, lo mismo que en la distribución geográfica."

Comments from Zimmer: "NO. In fact, from the perspective of having to sort these things out in the field, I'd be ecstatic if we could find a reason for lumping victoriae and nuna!"