Proposal (650) to South American Classification Committee
A. Resurrect Mustelirallus for Porzana albicollis
B. Transfer Neocrex to Mustelirallus
Porzana albicollis (Ash-throated Crake) has remained in the Porzana for most of its history, starting in 1868 with Sclater and Salvin, although Bonaparte in 1858 had described a new monotypic genus, Mustelirallus, for what was then Rallus albicollis. It basically “looks like” a Sora, so no surprise (to me) that its placement has been uncontroversial. The only contrary opinion that I can find is that of Benson and Winterbottom (1968), who proposed that it was the sister to African Crecopsis egregia.
Garcia-R. et al. (2014) produced a phylogeny for the Rallidae based on a fairly large analysis of DNA sequence data, both mitochondrial and nuclear, largely compiled from GenBank etc. Their taxon sampling was fairly good for a family that is cosmopolitan and difficult to collect: 70 species in 22 of 33 extant genera. They found that Porzana albicollis was not particularly closely related to true Porzana (of which our P. carolina is a member; type species = P. porzana) but rather was the sister to Neocrex. A section of their tree is pasted in below – see the original paper for other trees and the full data set; let me know if you need a pdf.
Analysis: The genetic data require removal of albicollis from Porzana. That Porzana albicollis has an antiphonal duet has been reported in the literature, and for is unlike anything I associate with Porzana carolina or can find for P. porzana. Check out this amazing recording by Roger Ahlman on xeno-canto. Also, true Porzana also have greenish or yellowish legs, whereas those of albicollis are a purplish brown. Mustelirallus Bonaparte, 1856, is available; as noted above, Bonaparte described this genus to remove the species from Rallus, which at that time was broadly defined.
However, as pointed out to me by Mark Robbins, a better solution would be to merge Mustelirallus albicollis with Neocrex. From the figure above, as Mark pointed out, based just on branch lengths, this would be consistent with the genetic data. The two species of Neocrex share with albicollis a Neotropical distribution, general similarity in size and bill shape, and a tendency to occur in habitats besides marshes, such as grassy areas. Obviously, this is not evidence for congener status, but the point is that treating albicollis and the two Neocrex as congeners does not violate any important morphological or habitat themes. The two Neocrex have red legs, matching their red bill bases. (Tangentially, red-legged Cyanolimnas of Cuba with its red bill base is almost certainly a Neocrex-derivative, and I see no reason to maintain that monotypic genus … but that’s a NACC issue.)
Mustelirallus Bonaparte1858 has priority over Neocrex Sclater & Salvin 1868. This causes some unfortunate taxonomic instability. Neocrex is feminine whereas I assume Mustelirallus is masculine (because Rallus is). Neocrex colombiana would thus become Mustelirallus colombianus, but erythrops in invariable.
This proposal is separated into two parts:
A. Resurrect Mustelirallus for albicollis.
B. Merge Neocrex into Mustelirallus.
Recommendation: I recommend a YES to both because (1) unless the sample is badly misidentified or the analyses totally botched, albicollis cannot be maintained in Porzana, and (2) treating albicollis and the two Neocrex as congeners is preferable, in my opinion, to resurrecting a monotypic genus. Reasons to vote NO would be (1) to wait for additional corroborating samples or data, or (2) to prefer monotypic Mustelirallus.
BENSON, C. W., AND J. M. WINTERBOTTOM. 1968. The relationship of the Striped Crake Crecopsis egregia (Peters) and the White-throated Crake Porzana albicollis (Vieillot). Ostrich 39: 177-179.
GARCIA-R, J. C., G. C. GIBB, AND S. A. TREWICK. 2014. Deep global evolutionary radiation in birds: diversification and trait evolution in the cosmopolitan bird family Rallidae. Molecular Phylogenetics and Evolution 81: 96–108.
Van Remsen, October 2014
Comments from Stiles: “YES to treating albicollis as congeneric with the two “Neocrex”, and to using Mustirallus as the genus name due to priority.”
Comments from Nores: “YES. The two species are sister taxa and Mustelirallus (Bonaparte 1858) has priority over Neocrex (Sclater & Salvin 1868).”
Comments from Zimmer: “YES. Genetic data look solid, and the antiphonal dueting of albicollis was the first thing that came to my mind when hearing that genetic data places it someplace other than with Porzana. Hearing a recording of that antiphonal duet is one thing – you should see and hear it at close range – very impressive indeed, and not completely unlike anything you would ever hear out of P. carolina.”
Comments from Robbins: “YES, for moving albicollis out of Porzana based on the genetic data. It is unfortunate that an esoteric name like Mustelirallus has to be resurrected to deal to include it with Neocrex. Given that Mustelirallus hasn’t been used for > 150 years, one could argue for using the well-established Neocrex. Regardless of the avenue taken, it is better to unite albicollis and Neocrex into the same genus instead of yet another monotypic genus.”
Comments from Areta: “A. YES to placing albicollis in Mustelirallus, B. NO to merging it with Neocrex. Vocally and in plumage and shape albicollis is very different from Neocrex erythrops, and we know nothing about vocalizations of N. colombiana. I think in this case a conservative change is better than one that goes too far. Crecopsis egregia is now considered Crex egregia, and vocally it doesn't seem very similar to albicollis. Even if egregia is sister to albicollis, Mustelirallus antedates Crecopsis, so the use of Mustelirallus albicollis is a good, solid, taxonomic move.”
Additional comments from Robbins: “Given Nacho’s comments, I vote NO on B.”
Comments from Jaramillo: “YES to both. A – Ash-throated Crake cannot remain in Porzana, due to voice, molecular data, and in my mind morphology. Other than having a short bill, I see very little to unite it with true Porzana.
“B – YES. Although the red on bill and legs is a clear difference between Neocrex and albicollis it is a minor one it seems. Red legs pop in and out in some South American Rails (Red-and-white Crake is one that comes to mind) yet it seems to be closely related to green legged species. At least in coloration and voice. So the coloration does not make me all that concerned. I do think that morphologically Neocrex and Mustelirallus albicollis are not that far off, they share the short bill, the somewhat stout body shape, and they also have curiously barred undertail coverts which are not common in our rails. Agree that the Zapata Rail should be looked at, and it may be in this group. However, few living people have seen it, no vocal data exists, and we only have a handful of specimens so good luck on that one.”
Comments solicited from Dan Lane: “The two parts of Proposal 650 seem to have been passed by SACC a bit too easily without really considering the second part. I agree that Part 1 is a required move given the molecular data, and not a surprising one considering aspects of the life history of M. albicollis. But Part 2 is rather a leap! I don't understand what problem there is with resurrecting a monotypic genus for albicollis (or for any bird that shares no distinct unifying characters with its closest relatives)! I see no SACC proposals to lump Amaurolimnas or Micropygia and abolish those (rightfully maintained, in my opinion) monotypic genera, so what would be so terrible about having yet another?”
“In plumage characters and voice, I would say that there is relatively little to unite Mustelirallus albicollis and Neocrex. I don't have experience with N. colombiana in life, but my experience with Neocrex erythrops is probably about as much as most field ornithologists have had, and I have a bit of experience with Mustelirallus albicollis as well. The two species may share similar habitats (fields and drier marsh, much like most rails), and have short bills (as do all crakes), but the similarities end there, and none seem distinct from other rail lineages to me. On the contrary, there are several aspects that distinguish M. albicollis and N. erythrops. For example, there is mounting evidence that Neocrex erythrops is a long-distance migrant (e.g., the type locality is Lima, Peru, but there seems to be no evidence of resident populations there; I have made several efforts to find them in the only likely habitat I know around the city--coastal wetlands--using playback, and have had no success. To my knowledge, the only modern records from the city are birds found in town parks, city streets, and other unlikely habitats that suggest they are on the move, and there are North American records of the species also supporting long-distance migration). I know of no suggestion of migration with Mustelirallus albicollis.
“As far as I'm aware, the vocalizations of N. erythrops were only discovered with any certainty in the past decade or so, and the repertoire of the species is still being sorted -- factors that belie the inconspicuousness of the sounds; their structure is not like that of Mustelirallus (see http://www.xeno-canto.org/species/Neocrex-erythrops). Mustelirallus albicollis, by comparison, is easily detected by voice, but its primary vocalizations seem to be typically given as a duet (as Kevin mentions above) with the two sexes (presumably) giving rather different sounds in an overlapping way (see http://www.xeno-canto.org/species/Porzana-albicollis). Other (mostly not closely related) Neotropical rails have similar duets using unlike, sex-specific components (e.g., Aramides cajanea, Anurolimnas castaneiceps, Pardirallus nigricans and P. sanguinolentus, and some members of the Rallus crepitans/longirostris group), but this dueting format seems to have no homologous version in Neocrex.
“Given no obvious uniting morphological or vocal characters, I think it is best to take the conservative route suggested by Nacho and maintain Neocrex separate from Mustelirallus until and unless unifying characters are uncovered ... otherwise, what sense does it make to treat them as congeners? What value is a genus if there are no unifying characters among its members?”