Coeligena torquata was described by Biossoneau (1840) from Bogotá (it does not occur in Bogotá per se, but is common just over the W edge of the altiplano to the W). Various subspecies have been described since, the most relevant to this question being insectivora Tschudi 1844, margaretae Zimmer 1848 and eisenmanni Weske 1985.

Coeligena inca was described by Gould in 1852, and the subspecies omissa was described by Zimmer in 1948.

The two groups differ strikingly in the color of the "collar", this being white in torquata and rufous in inca. They were considered separate species by Cory (1918) and Peters (1945), although Peters indicated doubts by referring to the latter as "Coeligena (torquata?) inca", however, he did not specify his reasons for so doing.

The first to specifically suggest that they were conspecific was Zimmer (1948), in the course of describing margaretae and omissa. He noted various characters in which these two were intermediate between nominate torquata and inca (among others, that this collar was paler rufous in omissa), and noted that in the rufous collars of inca and omissa, the feathers were rufous only at their tips, and white basally. When he wrote, there was a gap of over 300 km between the nearest localities of the southernmost known specimen of torquata (race insectivore) and inca (race omissa). In 1985 Weske described the final race in this complex, eisenmanni, and also reported specimens of insectivora from further S, such that the distributional gap had largely been closed. He noted that eisenmanni and insectivora were separated by only 20 km on either side of the deep valley of the Río Apurímac and that both eisenmanni and omissa occurred on the Cordillera de Vilcabamba at localities separated by ca. 130 km, with no specimens from the intervening area (in which available habitat appeared to be fairly continuous). He listed a further series of features in which eisenmanni was intermediate between insectivora and omissa as well as features unique to eisenmanni, and concluded that Zimmer had been correct in lumping torquata and inca. This conclusion had been accepted by Meyer de Schauensee (1966), and was followed by Sibley & Monroe (1990).

However, Schuchmann (1999) resplit torquata and inca based on "plumage differences and disjunct distributions". He did not elaborate on these distinctions, nor did he specifically address the numerous points of intermediacy of races margaretae, insectivora, eisenmanni, and omissa noted by Zimmer and Weske. These involved various parts of the plumages including the pattern of the frontlet and crowns, gorget, general colors of upper- and underparts, tail patterns, width of the "collar" etc. It is noteworthy that he also did not describe in detail or illustrate the races insectivora or omissa. Although I do not make such a proposal, I suspect that purely on the basis of plumage one could make a better case for splitting the Venezuelan conradi from adjacent nominate torquata ­ they are probably the most different-looking forms in the entire complex.

In this case, after reading the accounts in HBW and looking at the illustrations I was inclined to accept the split of inca from torquata, but upon reading the much more detailed analyses of Zimmer and Weske I was led to conclude that Schuchmann's treatment was much more superficial ­ and unconvincing. I therefore feel that Coeligena inca should continue to be considered conspecific with C. torquata, and recommend a NO vote on this proposal.

Literature Cited:

Boissoneau 1840, Revue Zool. 1840:6
Gould 1852, Contr. Orn :136
Cory 1918, Catalogue of Birds of the Americas, vpl. 2 pt. 1.
Peters 1945, Checklist of Birds of the World, vol. 5
Zimmer 1948, Auk 65:410
Weske 1985, AOU Monogr. 36:35
Sibley & Monroe 1990, Distribution and Taxonomy of Birds of the World
Schuchmann 1999, Handbook of Birds of the World, vol. 5

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

Comments from Remsen: "NO. The seems to be a classic example of a split based on an incomplete appreciation of the geographic variation involved."

Comments from Stotz: "NO. These may very well be separate species. In fact, perhaps they probably are. The two basic types are both known from the Urubamba valley in southern Peru. But while this is suggestive, there aren't sufficient samples from the region to say anything about whether interbreeding might be occurring, or if the taxa are actually in contact. The subject needs more work."

Comments from Pacheco: "NO. A proposição de Schuchmann não foi devidamente imunizada para ser aceita sem contestação. Voto pela manutenção da subordinação de inca, até que mais informação esteja disponível."

Comments from Schulenberg: "YES. Recently Dan Lane (in press) observed a white-breasted bird in the middle Urubamba valley; dark breasted birds (omissa) have been collected in the upper Urubamba. So, this is only a single observation, and of course sight records are no way to assess potential introgression. Still, this record fit a pattern of observations by Lane of birds otherwise known south only to the Apurimac valley (in other words, not reported from the relatively well known upper Urubamba), so something interesting seems to be going on down there.

"The evidence for a split admittedly is pretty weak, as it stands now, but I'm predicting (based on Lane's record) that white- and rufous-breasted birds "must" meet in the Urubamba valley. So I'll stir the pot and vote yes."

Comments from Jaramillo: "YES ­ Reading Weske was important in my decision. I find it odd that right at the area where these two groups meet there is a population that is somewhat intermediate, but also shows several features not found in either (longer winged, less sexual size dimorphism, bronze central rectrices, more white in tail, bronze uppertail coverts). The pattern I would expect is that this intermediate population should be pretty clearly intermediate in its features, and also that variability of the characters should be high. Weske's paper does not note that eisenmanni is particularly variable, which would be the expectation if it is an intermediate population. Based on the measurements and plumage features, eisenmanni does not fit clearly as an intermediate population to me, it just seems to be something different from the other two, maybe with hybrid origins but not a classic intermediate population.
The fact that omissa and insectivora come close in range and don't show a clear cline suggests to me that two species are involved. The interpretation of what eisenmanni is, well I think that is an open question. An intermediate population, even in a steep step cline should not be different and distinct from the two end points. Perhaps the dynamics of how these two end points are intermingling in the area of contact are complex, and as such I would rather seek that more work be done in this region of potential intergradation and keep the two as separate species than to lump them. I am more impressed by the fact that there is no clear cline here, more than by the presence of this odd somewhat intermediate population called eisenmanni."

Additional comments from Stiles: "Based on published, info I'll maintain my
recommendation to not split C. inca, at least until Dan Lane or somebody
collects a decent series from the middle Urubamba."

Comments from Nores: "NO, aunque en este caso parecería más justificado que en los anteriores. De todos modos el análisis de Zimmer parece concluyente, especialmente que las razas inca y omissa tienen las plumas del collar sólo la punta rufa y el resto blanco."

Comments from Zimmer: "YES. Gary's points are well taken, and Schuchmann's analysis clearly doesn't take into account the extent of geographic variation. But as both Tom and Alvaro point out, something funny is going on in the contact zone. While the morphological differences between these two forms may not be that far apart, they remain substantial compared to interspecific differences in several other hummingbird genera. So, until we have a clearer picture of what is happening in the contact zone, I'm for treating them as separate."