Proposal (147) to South American Classification Committee
Recognize Eriocnemis sapphiropygia as a species separate from E. luciani
This proposal would elevate to species status a taxon (E. sapphiropygia) considered to be conspecific with E. luciani in all publications between ca.1945 and 1999, when Schuchmann (1999) separated them.
The forms involved are luciani of extreme S Colombia and Ecuador, catharina of N Peru and sapphiropygia of C Peru to Bolivia, all described in the mid-nineteenth century, and baptistae of S Ecuador, recently described by Schuchmann et al. (2001). The first three were considered separate species by Cory (1918), and if his sequence of species in Eriocnemis is anything to go by, he considered catharina closer to luciani than to sapphiropygia. Chapman (1926) did not discuss species limits beyond noting that a supposed sapphiropygia specimen from Ecuador was really a "young luciani". Peters (1945) lumped catharina and sapphiropygia into luciani without comment, as was his wont, and this arrangement was followed without questioning by Zimmer (1951), Meyer de Schauensee (1966), Fjelds & Krabbe (1990) and Sibley & Monroe (1990). Zimmer also discussed the diagnostic characters and distributions of these three forms while describing a new subspecies, marcapatae, from S Peru (now often not recognized).
The first author to raise sapphiropygia to species rank and include catharina within it was Schuchmann (1999), who stated that "intermediate specimens [with luciani] exist, but morphological differences support treatment as separate species". This split was accepted by Clements & Shany (2001) and Ridgely & Greenfield (2001) without comment.
More details on the split were presented by Schuchmann et al. (2001) in a review of the genus Eriocnemis that included the description of the southernmost population of luciani as a new subspecies, baptistae, based mainly on its more bronzy overall coloration. Unfortunately, their statements and descriptions are not always consistent with those of Zimmer (1951) or Schuchmann (1999), creating some doubts regarding their conclusions. Their arguments may be summarized as follows:
a) in external morphology, all forms of the group are very similar in bill length and length of closed wing; the only clear-cut difference separating the two races of luciani (luciani, baptistae) from those of sapphiropygia (catharina, sapphiropygia) is the length and shape of the tail: tails of the latter are shorter and less forked than those of the former (lengths of the first and fifth rectrices presented). No statistics were employed, but the differences in tail shape are surely significant by inspection (virtually no overlap).
b) the critical forms with regard to plumage differentiation are the adjacent (albeit separated by a ca. 300 km gap) baptistae and catharina. The features of catharina are somewhat equivocal: it shows a tendency toward luciani in its blue-tinged forehead and lack of at least pronounced coppery on the nape (one of the features used by Zimmer to distinguish it from sapphiropygia). On the other hand, the presence of a white midventral "stripe" on the abdomen in females is shared with the latter. It is greener in overall coloration than baptistae and sapphiropygia, and males differ from all other forms in the group in having a strong blue sheen on the abdomen and more bluish (less violet) lower tail-coverts. Schuchmann et al. cited specimens showing intergradation between catharina and sapphiropygia in geographically intermediate localities but provided no detailed descriptions.
c) Schuchmann (1999) mentioned "intermediate specimens" between the species luciani and sapphiropygia (presumably between baptistae and catharina?) but I can find no mention of these in Schuchmann et al. (2001), which is a bit worrisome.
d) the coppery nape is not mentioned as a diagnostic feature of the species sapphiropygia (because it is lacking in catharina??) leading me to wonder about their use of "Coppery-naped Puffleg" as the English name for this taxon.
In sum, the hard evidence for this split boils down to two characters: the difference in tail shape (which is in fact much greater than that found among the races of other polytypic species of Eriocnemis, viz. E. vestitus) and the ventral pattern of the females (also mentioned by Zimmer). The characters of catharina are equivocal in several other respects, although the existence of specimens intermediate between this form and sapphiropygia (albeit not described in detail) would also favor uniting it with the latter. The gap in distribution between baptistae and catharina (if I read Schuchmann et al.'s maps correctly) is large but may reflect lack of collecting rather than lack of birds, and probably should not be used as evidence either way at this point. I feel that the more detailed morphological data of Schuchmann et al. incline the balance in favor of a split, although I would have been more convinced had they paid more attention to some of the details mentioned above. Having no personal knowledge of the forms in question, I am hesitant to make a strong recommendation and await comments from those with more familiarity with them. For now, a tentative YES on this proposal.
Clements & Shany 2001
Fjelds & Krabbe 1990
Meyer de Schauensee 1966
Ridgely & Greenfield 2001
Schuchmann 1999: HBW, vol. 5.
Schuchmann, K.-L., A-A. Weller & I. Heynen. 2001. Systematics and biogeography of the Andean genus Eriocnemis (Aves: Trochilidae). J. Ornithol. 142:433-481.
Sibley & Monroe 1990
ZIMMER, J. 1951b. 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 Robbins: "NO. I certainly don't find anything that has been presented that compels one to support the recognition of Eriocnemis sapphiropygia as a separate species from E. luciani."
Comments from Pacheco: "YES. As informaes disponveis parecem suportar um "tentativo" sim proposta."
Comments from Jaramillo: "YES _ This is difficult, as the data seems lacking but suggestive that the split is good. I don't know these taxa, and admit that I am going along with Gary as he has a well-trained feel for the level of difference shown by species of hummingbirds, and he seems to be (albeit hesitantly) convinced that this split is good. I apologize for my lame justification on this one."
Comments from Nores: "SI. Pienso que la forma y tamao de la cola son caracteres propios de especies y no de subespecies."
Comments from Remsen: "YES, with reluctance. After 2.5 years, I finally gave up on waiting for comments from our two Peru bird book authors. Acknowledging all Gary's doubts on the analysis, forced to vote Y or N, I'll go with Y because (1) Peters and subsequent authors did not provide rationale for the original lump, and (2) the difference in tail shape, as Manuel noted, is typically associated with species rank (e.g., the NACC recognition of multiple species with Chlorostilbon canivetii group). If hummingbird experts Stiles and Schuchmann are comfortable with species rank, then so am I."
Comments from Zimmer: "YES. Evidence isn't overwhelming, but this is another of those unjustified Peters' lumps, and I think the burden of proof is on that camp."