Proposal (231) to South American Classification Committee
Split Acestrura from Chaetocercus
The two genera in question had been recognized as separate through most of the twentieth century (e. g., Simon 1921, Peters 1945, Meyer de Schauensee 1966, Sibley & Monroe 1990). However, Schuchmann (1999) lumped them with the bald statement: "..no differences in external morphology justify their separation" and this was followed (equally without explicit justification, except for similarities in the figures of the birds in the plates of HBW) in the SACC list. Before Schuchmann, the last authority to have combined these genera (under the older name Chaetocercus) was probably Cory (1918), equally without comment. The re-splitting after Cory was due to Simon (1921), the recognized expert on hummingbirds of his day but also a notorious splitter at the generic (and specific) levels. In sum, the situation is sufficiently confusing that a review of the characters of these genera seems desirable to resolve this question.
Unfortunately, I have been unable to consult the original descriptions of Chaetocercus Gray 1855, Acestrura Gould 1861, Myrmia Mulsant 1875 or Microstilbon Todd 1913. However, for the first three the key presented by Salvin (1892) probably follows the original descriptions closely, as do keys in the other volumes of the "Birds of the British Museum" series edited by Sharpe. My own experience with other genera described by Gould and Mulsant et al. is in accord: these authors often presented their generic diagnoses in the form of keys, with the key characters permitting allocation of specimens in their respective genera. Salvin's key treats exclusively adult males and it seems likely that the characters of these formed the principal (if not the only) basis for generic distinctions. This is also indicated by the generic names themselves: Chaetocercus means "tooth-tailed" whereas Acestrura means "needle-tailed", both descriptive of the tails of the respective adult males.
In Salvin's key, the small hummingbirds of the "bee" group are characterized by having the tail "very varied, not regularly cuneate or spatuled; throat usually glittering red, sometimes violet or pale green, with a violet border, sometimes plain.." The first group to key out has the tail "tail deeply and evenly forked with the outermost primary [apparently an error for "rectrix"] not shorter than any of the others, the median much the shortest; none of the rectrices filiform" and includes, among others, Calliphlox. The second major group of genera has tails "more even, more or less irregularly formed, the median rectrices, however, not abnormally short". This group includes the North American genera and Acestrura, which is separated from the others by its "filiform" outer rectrices (4 and 5), which are indeed "needle-like" (being little more than the rachises) in mulsanti, the type species.
A third group with the "tail nearly normal (rounded or forked)" includes Stellula, Mellisuga and others not relevant here. The fourth and final group, including Chaetocercus, (all four key out at the same level, strictly dichotomous keys not being universally used at that time) has the "tail abnormal" [i.e., neither rounded nor forked], the median rectrices very short, the lateral ones shorter than the submedian ones". This effectively says that rectrix 1 is very short, 5 is shorter than 2. Examination of male specimens of Chaetocercus jourdanii shows that rectrices 1 and 5 are indeed very short, the latter sharp-pointed and spinelike; 2 is longer and blunt, whereas 3 and 4 are much longer, narrow but blunt-tipped.
However, after this things get messier. In Salvin's key one of the species later assigned to Acestrura (bombus) is placed in Chaetocercus as is burmeisteri, later separated in Microstilbon.
Besides mulsanti, Acestrura includes heliodor, micrura (later separated in Myrmia due to its extremely short tail) and "decorata", now considered a hybrid. Upon examining specimens, I find that one can arrange the species in a gradient of tail shapes from mulsanti with filiform, rather long outer rectrices, through heliodor in which the outer rectrices are much shorter and, while slender and very acute, are not really filiform and in which the central rectrices are nearly as short as in Chaetocercus - in fact, the tail of this species is almost exactly intermediate between those of the type species of Acestrura and Chaetocercus. The tails of the females of all these species are double-rounded with rufous tips and bases and blackish subterminal areas on the lateral rectrices, differing in whether the central rectrices are more rufous or green. My conclusion from all this is that there is indeed little justification for not subsuming Acestrura into Chaetocercus, as did Schuchmann, and I recommend a NO vote on this proposal. A further recommendation is that the generic status of Microstilbon and Myrmia should also be reexamined critically, but I leave that to someone with access to specimens and the original descriptions.
Meyer de Schauensee 1966
Salvin, O. 1892. Trochili. In Catalogue of Birds in the British Museum, vol. 16.
Schuchmann 1999 (HBW vol. 5)
Sibley & Monroe 1990
Simon, E. 1921. Histoire naturelle des Trochilidae (Synopsis et Catalogue). E. Mulo, Paris.
Gary Stiles, August 2006
Comments from Cadena: "NO. It would be interesting to see whether the two "genera" come out as distinct clades in the working phylogeny of McGuire et al., but since the status quo is to merge them in a single genus and morphology provides no evidence to split them (according to Gary's analysis), this should probably not be changed unless the two do not turn out to be sister to each other, which seems quite unlikely anyway."
Comments from Robbins: "NO. Gary provides good rationale for not recognizing Acestrura from Chaetocercus."
Comments from Stotz: "NO. I suspect that current arrangements of woodstars are not to be taken seriously. I rather prefer Acestrura to Chaetocercus as a name, but not much you can do about that. I suspect that some of the other small hummingbird genera like Atthis and Mellisuga will end up playing a role eventually. There doesn't seem to be a strong argument currently for any treatment, so I think we should leave things be."
Comments from Jaramillo: "NO - I liked the reasoning behind Gary's analysis."
Comments from Pacheco: "NO. Concordo com as raz›es apresentadas e consequente recomendaŤ‹o de Stiles."