Proposal (142) to South American Classification Committee

Split Heliangelus clarisse from H. amethysticollis, lump H. spencei into H. clarisse

This proposal would essentially overturn the revision of Zimmer (1951), as followed by Meyer de Schauensee (1966) and most subsequent authors, and return to the taxonomy of Peters (1945), as proposed by Schuchmann (1999), followed by Hilty (2002).

A considerable number of forms in this genus were described as separate species prior to the synthesis of Cory (1918), often based upon very minor differences in male plumages; the problem is exacerbated by the fact that the group as a whole is morphologically quite uniform, as well as by the number of apparent hybrids and aberrant forms described from one or two specimens and evident confusion regarding type localities in a few cases. It thus seems best to start with Cory (1918), who listed four species in the group: spencei of Venezuela, Clarisse of NE Colombia, laticlavius of E Ecuador and = amethysticollis of E Peru and E Bolivia (ignoring several forms of dubious affinities or status, and leaving aside strophianus of W Ecuador and extreme SW Colombia, which seems to be clearly distinct; see below). This was essentially followed by Chapman (1926), who did not mention spencei. In 1945, Peters reduced this group to two species, clarisse + spencei and laticlavius + amethysticollis, without comment. 

The first serious revision of the group was that of Zimmer (1951), who described the race decolor from NE Peru, essentially filling in the gap between E Ecuador and SE Peru, comparing it in some detail with laticlavius and amethysticollis. He then stated that "Clarisse belongs in the amethysticollis group" without further comment, and mentioned several characters justifying separating spencei as a separate species including color and "texture" of the male's gorget and throat pattern of females. He also noted numerous characters of both sexes justifying maintaining species status for strophianus. The differences between the three southern races were in details of the color of the males' gorgets, frontlets and pectoral bands and the grayness vs. rufescence, as it were, of their lower underparts; variation in most characters seems rather of a mosaic type rather than showing a clear N-S trend (except possibly pectoral band), which might have influenced him regarding the inclusion of clarisse, which does not appear to differ strikingly in these respects, its chief claim to difference seeming to be a more rosy gorget. This arrangement was followed by Meyer de Schauensee (1966) without comment, as well as by Meyer de Schauensee & Phelps (1969) and Hilty & Brown (1986). In the meantime, Phelps & Phelps (1953) had described violiceps as a race of clarisse, from the Serranía de Perijá on the Colombian-Venezuelan frontier but without mentioning spencei although its gorget is more violaceous than that of clarisse, perhaps approaching the latter (?).

The next useful information was published by Fjeldså & Krabbe (1990), who described the gorget of spencei as "violet, often with a distinct coppery sheen, not strongly glistening" (the "different texture" of Zimmer?) and noted similarities in the throat color of females to that of mavors, males of which have an orange to orange-green gorget. They also noted the "peculiar silvery-green" frontlet of spencei, apparently unlike those of other members of the group including clarisse and violiceps. They called the enlarged amethysticollis of Zimmer a "megaspecies", noting that clarisse had often been split in the past. Finally, Schuchmann (1999) returned to the classification of Peters (1945), adducing as reasons "subtle morphological differences" (unspecified) and "markedly disjunct range" (certainly the case) as "supporting treatment as separate species", noting also that "spencei was sometimes considered a separate species, especially when the races of (clarisse) were lumped into amethysticollis". This treatment was followed by Hilty (2003).

There would seem to be no really clear-cut resolution to this problem, as much depends upon what one takes to be the "status quo" (or starting point). Although it was evident that the group was probably oversplit, the lumping of Peters was done without explicit rationale. Zimmer was explicit on several points (such as excluding spencei) but similarly gave no reasons for considering clarisse a subspecies of amethysticollis. Finally, Schuchmann gave no clear morphological reasons (the differences between all these forms are more or less "subtle") for returning to the arrangement of Peters, the only clear argument being distribution: the obvious division into northern and southern groups, separated by a gap. My reading of Fjeldså & Krabbe leads me to question the lumping of at least spencei with the others since they implicitly raise the possibility that it might be closer to mavors based upon female plumage and the "coppery sheen" of the male's gorget, which also might be the motive for Zimmer's citing of differences in color and texture of its gorget. This is a lot of "might be's", but in the context of the generally vague descriptions of most plumages, this could justify maintaining this form as a species, pending future study. More problematic is what to do with clarisse -- follow Zimmer in lumping it with amethysticollis, or follow Schuchmann (in part) in separating it (based largely on distribution, also a whiter pectoral bar and rosier gorget -- features that I am not wholly convinced are of specific value given their variation among the S races -- and there is N-S trend -- white to buffy -- in the pectoral bar, if not in other characters). Hence, I consider that the weight of evidence, such as it is, favors maintaining the Zimmer-Meyer de Schauensee arrangement for the present and suggest (not very strongly) a NO vote on this proposal, while emphasizing that a careful analysis citing explicitly all characters is sadly lacking and that genetic data also could be helpful (at least regarding a possible relationship of spencei with mavors). I could easily be persuaded to change my mind, particularly regarding clarisse, were a careful study to appear.

Literature Cited:

Cory 1918, Catalogue of Birds of the Americas, vol. II, part 1.

Chapman 1926, Distribution of Bird-Life in Ecuador, Bull. AMNH vol. 55.

Peters 1945, Checklist of Birds of the World, vol. 5.

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.

Phelps & Phelps Jr. 1953, Proc. Biol. Soc. Wash. 66:1-12.

Meyer de Schauensee 1966, Species of birds of South America

Phelps & Meyer de Schauensee 1969, Guide to the birds of Venezuela

Hilty & Brown 1986, Guide to the birds of Colombia

Fjeldså & Krabbe 1990, Birds of the High Andes

Schuchmann 1999, HBW vol. 5.

Hilty 2003, Guide to the birds of Venezuela ("second edition")

Gary Stiles, November 2004

_____________________________________________________________________________________

Comments from Remsen: "NO, but like Gary, with little conviction. Neither classification provides convincing rationale, so I'll go with status quo until that latter is provided."

Comments from Robbins: "NO. Based on what is now available any decision would be arbitrary, thus I support the 'status quo'."

Comments from Pacheco: "NO.  Diante da complexidade do presente caso, é preferível continuar com o tratamento proposto por Zimmer (1951); porquanto, este ainda seja a única revisão do grupo (parafraseando Stiles) seriamente apresentada."

Comments from Jaramillo: "NO.  More work needed on this front. My guess is that more than one species is involved, but the division would seem to be arbitrary at the moment."

Comments from Nores: "NO; pienso que las diferencias de plumajes son sólo subespecíficas y que el argumento indicado por Schuchmann en varias oportunidades 'markedly disjunct range o disjunct range' para separar especies es más un fundamento para considerarlas subespecies que especies."

Comments from Zimmer: "NO. This seems like a case where more data is needed. Any decision on where to split at this point seems arbitrary."