Background: Threnetes leucurus (Pale-tailed Barbthroat) with a range through most of Amazonia, has been long considered a separate species from Threnetes niger (Sooty Barbthroat), found in French Guiana and adjacent Brazil (Amapa). Grantsau (1969) described a new species in this complex, Threnetes loehkeni from Amapá. Hinkelmann (1988) suggests that Threnetes loehkeni is best treated as a subspecies of leucurus, but at that time does not include Threnetes niger in the mix. In this paper, he also nixes a couple of Ruschi taxa (Threnetes cristinae and T. niger freiri described from specimens from Amapá). In HBW, Hinkelmann (1999) lumps all of the Amazonian Threnetes into one species under the oldest name niger. He states "niger represents a melanistic form of geographically limited distribution, based on apomorphic characters within the range of variation found in leucurus; otherwise the two are apparently inseparable." Because of Hinkelmann's uncertainty about Threnetes niger and the fact that Hinkelmann and Schuchmann (1997) did not correctly apply the names in this case, the taxonomic issues did not really become generally appreciated until the publication of HBW, Vol. 5.

Analysis:
The taxon niger is as Hinkelmann states a melanistic bird. I have seen two specimens, one at the Field Museum, and one in Sao Paulo. On these specimens, the basic Threnetes plumage pattern is visible underlying the dark coloration of the underparts. The basic plumage pattern is that of loehkeni. This led Parker et al (1996) to suggest that loehkeni was probably a pale morph of niger. The logic of Hinkelmann's treatment of loehkeni and leucurus as conspecific with niger, is not made explicit, but seems to be based on loehkeni as a subspecies of leucurus and niger being a melanistic population allied to one or the other of these (see Hinkelmann and Schuchmann 1997 for the most explicit discussion, although they mess up the name there).

It looks to me like Hinkelmann is correct that niger and loehkeni are basically melanistic and normal morphs of the same thing. It isn't completely clear that whether there is a geographical element to this, so that they should be considered two subspecies, or whether they are morphs within the same subspecies. In HBW, Hinkelmann takes the former view. In Hinkelmann and Schuchmann (1997) they suggest that the latter view is more likely. The question is whether the treatment of niger plus loehkeni as conspecific with the widespread leucurus is appropriate. Hinkelmann (1988) suggests that the plumage characters that Grantsau (1969) mentions for loehkeni are invalid, except for the tail pattern. In loehkeni (and in niger), the tail is dark green, with very narrow pale tips to the outer four rectrices on each side. In leucurus, the outer four tail features are mostly pale (ranging from white to ochraceous depending on the subspecies), with a black subterminal band followed by pale tips. The pale on the tail is broadest on the innermost of the four outer feathers, so nearly completely pale, and gets progressively more black as you move outward, so that the outermost feather is about 50% black. The result is a very different and distinctive tail pattern from that of niger and loehkeni (the plate in HBW does a good job on these birds).

Hinkelmann (1988) treats loehkeni as a "distinctive subspecies" of leucurus, stating "the colouration of loehkeni does not go beyond the variation found in all subspecies of T. leucurus, variation which is probably due to age and sex." I would agree with Hinkelmann on this, except for the tail pattern. Although females and young birds of leucurus have less white in their tails than do adult males, in the 100 +/- specimens of leucurus at the Field Museum, none approach the condition of loehkeni.
Whether to treat niger as a distinct species from leucurus or as conspecific is difficult, however, I think that recognizing two species is the best course. The two taxa approach one another moderately closely, and may occur together. Typical leucurus occurs in Surinam and southern Amapá, at least (Hinkelmann 1988). Tostain et al. (1992) list both species from French Guiana, and gives both from two localities, St. Georges and Saul. However, since they do not give subspecies, I don't know whether the leucurus may refer to loehkeni rater than nominate. As an aside I would note that the range of niger within French Guiana includes most of the border with Surinam, so it should occur in Surinam, but I a unaware of any records.

Given the close approach of niger and leucurus and the possibility of sympatry without any indication of intermediate tail-patterned birds being found, and the lack of any reports of melanistic birds in leucurus from near the range of niger, I would argue that there is no evidence of intergradation occurring between niger and leucurus. Further, the other species in the genus, Threnetes ruckeri, allopatric to these birds in Trans-Andean forests, is very similar to leucurus in plumage, except for its distinctive tail pattern. The tail pattern of niger is even more distinctive than that of ruckeri in my view.

Proposed change: Recognize Threnetes leucurus as a distinct species from Threnetes niger, treating loehkeni as a subspecies of niger. The English name Pale-tailed Barbthroat would apply to leucurus. Threnetes niger has been called Sooty Barbthroat, except when treated as conspecific with leucurus. This name is not particularly appropriate when loehkeni is lumped into it. Loehken's Barbthroat seems inappropriate as well. Maybe Dark-tailed Barbthroat?

Recommendation:
I recommend treating Threnetes niger as a distinct species from Threnetes leucurus, treating loehkeni as a subspecies of niger.

References:
Grantsau, R. 1969. Uma nova espécie de Threnetes (Aves, Trochilidae). Pap. Avuls. Zool. 22:245-247.

Hinkelmann, C. 1988. Comments on recently described new species of hermit hummingbirds. BBOC 108:159-169.

Hinkelmann, C. 1999. Pale-tailed Barbthroat (Threnetes niger) [species account], in del Hoyo, J., Elliot, A. & J. Sargatel, eds. Handbook of the Birds of the World. Vol. 5. Page 539.
[How should you cite species accounts in this?]

Hinkelmann, C. and K. Schuchmann. 1997. Phylogeny of the hermit hummingbirds (Trochilidae: Phaethornithinae). Studies on Neotropical Fauna and Environments 32:142-163.

Parker, T. A., III, D.F. Stotz, and J. W. Fitzpatrick. 1996. Ecological and Distributional Databases, in Stotz, D. F., J. W. Fitzpatrick, T. A. Parker III, and D. K. Moskovits, Neotropical Birds: Ecology and Conservation. University of Chicago, Chicago.

Tostain, O., J.-L. Dujardin, Ch. Érard, and J.-M Thiollay. Oiseaux de Guyane. Société d'Etudes Ornithologiques, Brunoy, France.

=====================

Comments from Robbins: "[YES.] Doug has done a good job of providing details to the convoluted treatment of the T. niger complex."

Comments from Stiles: "NO. Sounds reasonable but lacks solid published evidence - if Doug were to publish something on this, I would probably go for it."

Comments from Zimmer: "A reluctant "NO". I think Doug's arguments are persuasive, but once again, I think there needs to be some published analysis, particularly since the historical treatments have been so muddled."

Comments from Jaramillo: "YES ­ although reluctantly. A quick note stating what has been stated here with illustrations of tail pattern and a map would be highly encouraged."

Comments from Nores: "YES. Acepto reconocer a Threnetes leucurus como una especie distinta de Threnetes niger y tratar a loehkeni como una subspecie de niger. Pienso que el patrón de la cola de leucurus es concluyente."

Comments from Remsen: "YES. In retrospect, and after reading the information assembled by Doug, it seems premature in my opinion to consider leucurus and niger conspecific -- the burden-of-proof in this case falls on the treatment that would consider them conspecific."